Showing 1–50 of 867 results for author: Tan, Z

RecPFN: Prior-Fitted Networks for In-Context-Based Recommendations

Authors: En Zhi Tan, Jia Xiang Lim, Bryan Lijie Chew, Tze Minh Ng, Benjamin Yan Han Yap

Abstract: We introduce RecPFN, a prior-fitted network that brings in-context learning to sequential recommendation. RecPFN is pretrained entirely on synthetic clickstream environments sampled from a broad structural causal prior, enabling it to amortize Bayesian-style inference from a small support set. At inference, a lightweight decoder-only transformer conditions on a handful of domain sequences and prod… ▽ More We introduce RecPFN, a prior-fitted network that brings in-context learning to sequential recommendation. RecPFN is pretrained entirely on synthetic clickstream environments sampled from a broad structural causal prior, enabling it to amortize Bayesian-style inference from a small support set. At inference, a lightweight decoder-only transformer conditions on a handful of domain sequences and produces next-item predictions for queries in a single forward pass, without any weight updates. Across eight public benchmarks, RecPFN achives state-of-the-art zero-shot performance while remaining strongly competitive with supervised methods in low-compute and low-data regimes. It is deployment-efficient and robust to domain shift, outperforming strong zero-shot baselines that rely on large real-interaction corpora. RecPFN provides a practical path toward generalizable, data-efficient recommenders and opens avenues for richer priors, longer-context ICL, and multimodal extensions. Code for training and evaluation is publicly available at https://github.com/SAP-samples/tabular-ai-recpfn/. △ Less

Submitted 20 August, 2026; originally announced August 2026.

Comments: 12 pages, 4 figures, 8 tables

Journal ref: In Proceedings of the 49th International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 1731-1742. 2026

Unregularized Convergence of Single-Loop, Entropy-Regularized Natural Actor-Critic

Authors: Zhiqiang Tan

Abstract: While entropy regularization is widely used to stabilize and accelerate Natural Policy Gradient methods, its ability to yield faster convergence rates for the unregularized objective remains underexplored. Existing analyses often rely on double-loop architectures and invoke a linear entropy penalty. To bridge the gap between theory and practice, we analyze a single-loop, entropy-regularized Natura… ▽ More While entropy regularization is widely used to stabilize and accelerate Natural Policy Gradient methods, its ability to yield faster convergence rates for the unregularized objective remains underexplored. Existing analyses often rely on double-loop architectures and invoke a linear entropy penalty. To bridge the gap between theory and practice, we analyze a single-loop, entropy-regularized Natural Actor-Critic algorithm under compatible linear function approximation. By training an uncentered critic, our critic tracking can remain stable even as the training policy approaches determinism and the Fisher information matrix degenerates. We focus on two primary regimes for the optimization landscape: a Stochastic Regime, where we fuse coupled actor-critic updates into a joint Lyapunov recurrence, and a Deterministic Regime, where we pivot to a Policy Mirror Descent framework to circumvent the collapse of Euclidean geometry. By exploiting a positive Minimal Action Gap in the unregularized Markov decision process, we introduce an Exponential Translation mechanism that maps the regularized gap to the unregularized one up to an exponentially decaying tail. By tuning the fixed temperature, our algorithm achieves accelerated unregularized convergence rates, up to approximation-error terms: $\tilde{\mathcal{O}}(T_{total}^{-1})$ in the Stochastic Regime, and $\tilde{\mathcal{O}}(T_{total}^{-2/3})$ for the average iterate alongside $\tilde{\mathcal{O}}(T_{total}^{-1/3})$ for the last iterate in the Deterministic Regime. Here, $T_{total}$ denotes the total number of stochastic critic updates (or Monte Carlo rollouts). Furthermore, in the tabular setting, our positive-action-gap analysis yields a $\tilde{\mathcal{O}}(T_{total}^{-2/3})$ average-iterate rate, surpassing the $\mathcal{O}(T_{total}^{-1/2})$ worst-case statistical barrier that applies without a positive action margin. △ Less

Submitted 19 August, 2026; originally announced August 2026.

EDITBRIDGE: Towards Faithful and Efficient Ultra-High-Resolution Image Editing

Authors: Jiayi Song, Shijie Huang, Fangtai Wu, Yubo Huang, Zhenxiong Tan, Songhua Liu, Jiaming Liu, Ruihua Huang

Abstract: High-resolution image editing is increasingly demanded in professional workflows, yet existing diffusion-based models remain constrained to resolutions below 1K due to quadratic attention complexity and prohibitive memory requirements. A prevalent workaround employs a two-stage pipeline: editing at low resolution followed by independent super-resolution. However, this approach suffers from two cri… ▽ More High-resolution image editing is increasingly demanded in professional workflows, yet existing diffusion-based models remain constrained to resolutions below 1K due to quadratic attention complexity and prohibitive memory requirements. A prevalent workaround employs a two-stage pipeline: editing at low resolution followed by independent super-resolution. However, this approach suffers from two critical issues: information divergence, where hallucinated details contradict the original high-resolution (HR) source, and texture degradation, manifesting as over-smoothed or over-sharpened artifacts. We propose EditBridge, a diffusion bridge framework for efficient ultra high-resolution editing. Unlike conventional diffusion that regenerates from noise, we formulate refinement as structured data-to-data translation from the low-resolution (LR) edited result to its HR counterpart, explicitly conditioned on the original HR source to preserve authentic details. To efficiently incorporate HR source guidance, we introduce a prior-guided block-wise sparse attention mechanism that exploits semantic correspondence from first-stage editing to constrain cross-image interactions to spatially aligned regions, significantly reducing computational overhead. Extensive experiments demonstrate that EditBridge achieves high-fidelity editing with superior perceptual quality at resolutions up to 4K, delivering 3.6--8.4$\times$ speedup at 2K and enabling practical 4K editing in 61 seconds. △ Less

Submitted 18 August, 2026; originally announced August 2026.

CABLE: Extending the Reach of Memory Retrieval via Complementary Antecedent-Based Linking and Expansion

Authors: Zheling Tan, Jin Gao, Dequan Wang

Abstract: As LLM agents operate across structured workflows and sessions, preserving long-term history does not ensure that later contexts can recover relevant evidence through a bounded memory interface. We study this evidence-reachability problem in long-term conversational memory, where retrieval still relies heavily on semantic similarity. This works well for topical recall, but it often misses earlier… ▽ More As LLM agents operate across structured workflows and sessions, preserving long-term history does not ensure that later contexts can recover relevant evidence through a bounded memory interface. We study this evidence-reachability problem in long-term conversational memory, where retrieval still relies heavily on semantic similarity. This works well for topical recall, but it often misses earlier experiences, plans, or motivations that are semantically distant from the later events they help explain. Existing memory graphs provide cross-memory structure, yet links driven mainly by semantic overlap can duplicate what the host retriever already recovers. We argue that link construction should instead prioritize a sparse set of retriever-complementary associations. We present CABLE (Complementary Antecedent-Based Linking and Expansion), a plug-in augmentation that constructs links designed to extend the host retriever's direct semantic reach. For each new memory, CABLE generates antecedent-oriented queries, retrieves prior memories, subtracts candidates in the direct semantic neighborhood, and verifies the remainder before adding the accepted complementary associations into a sparse directed graph. At retrieval time, CABLE expands the host system's retrieved seeds along these links to surface implicit supporting evidence. We evaluate CABLE with A-MEM on LoCoMo and MA-LongMemEval, and further integrate it into SimpleMem and Mem0g on LoCoMo, using Qwen3.5-27B, DeepSeek-chat, and GPT-4o-mini. CABLE yields higher mean LLM-judge scores in every evaluated system-level setting, with the largest gains in categories where useful evidence is distributed across memories or sessions, including open-domain, multi-session, and preference-oriented questions. These results support prioritizing sparse, reasoning-relevant associations that complement rather than duplicate the host retriever. △ Less

Submitted 18 August, 2026; originally announced August 2026.

Comments: Accepted by COLM 2026

I-SDPO: Instance-Level Adaptive Self-Distillation Policy Optimization

Authors: Yubo Zhang, Xinhong Ma, Zezhong Tan, Ziqiang Dong

Abstract: Group Relative Policy Optimization (GRPO) learns from reward differences within a rollout group, but receives no useful relative signal when every sampled response is incorrect. Privileged self-distillation can fill this gap with dense token supervision, yet applying it throughout training creates a different failure mode: the teacher is a biased, low-variance surrogate for the reward objective, s… ▽ More Group Relative Policy Optimization (GRPO) learns from reward differences within a rollout group, but receives no useful relative signal when every sampled response is incorrect. Privileged self-distillation can fill this gap with dense token supervision, yet applying it throughout training creates a different failure mode: the teacher is a biased, low-variance surrogate for the reward objective, so persistent imitation can oppose reward-improving updates after the policy becomes capable of producing successful trajectories. We introduce I-SDPO (Instance-Level Adaptive Self-Distillation Policy Optimization), which treats teacher reliance as capability-dependent. I-SDPO makes one routing decision per input instance and shares it across that instance's rollout group: all-incorrect groups use a privileged self-distillation objective, whereas any-success groups remain intact for GRPO. This design uses imitation only where group-relative rewards are uninformative. A local analysis characterizes when teacher and reward directions align and shows that a non-vanishing biased distillation weight induces an optimization bias floor. The routing rule automatically reduces the expected distillation rate as success probability rises, withdrawing teacher influence without a hand-designed schedule. On SciKnowEval, I-SDPO obtains the best result in all four scientific domains and improves average mean@16 accuracy from 56.67% with GRPO to 70.31%, with a maximum domain gain of 18.24 points. △ Less

Submitted 13 August, 2026; originally announced August 2026.

Comments: 14 pages, 3 figures

PatchGen: Learning Soft Intra-Image Predictive Subsets for Visual Generalization

Authors: Zhaorui Tan, Weimiao Yu, Xi Yang

Abstract: Visual classifiers are expected to generalize under data shifts, target shifts, and their combinations, yet most existing methods focus on domain invariance while failing to address intra-image predictive sufficiency. We investigate the structural hypothesis that each image contains a sample-adaptive oracle intra-image predictive subset sufficient for label prediction, while the remaining patches… ▽ More Visual classifiers are expected to generalize under data shifts, target shifts, and their combinations, yet most existing methods focus on domain invariance while failing to address intra-image predictive sufficiency. We investigate the structural hypothesis that each image contains a sample-adaptive oracle intra-image predictive subset sufficient for label prediction, while the remaining patches form non-essential complementary context that may correlate with the label. The theoretical analysis shows that restricting prediction to this oracle subset preserves the Bayes risk achievable by the full-patch representation while admitting a complexity bound that tightens with the oracle-subset size. Based on this view, we propose PatchGen, a text-free module that learns a sample-dependent soft predictive-subset mask as a task-driven proxy for the unobserved oracle subset mask. Specifically, histopathology visualizations suggest that PatchGen assigns higher scores to tumor-consistent regions than to some frequently co-occurring inflammatory context. Extensive experiments on natural and histopathological image benchmarks spanning all three shift settings show that PatchGen improves average performance over matched-backbone baselines in most evaluated configurations, enhances generalization to unknown classes, and remains competitive with vision-language methods without text supervision. △ Less

Submitted 12 August, 2026; originally announced August 2026.

From Points to Edges: Edge-Conditioned Spectral Operators for Physics-Sensitive PDE Learning

Authors: Zhentao Tan, Ruijie Quan, Yi Yang

Abstract: Neural operators have become a central tool for solving partial differential equations (PDEs), with spectral operators offering efficient global mixing across spatial locations. However, many PDEs contain physics-sensitive local structures that are critical to the underlying physical behavior. For example, in Darcy flow, local material interfaces are often reflected by sharp changes in the permeab… ▽ More Neural operators have become a central tool for solving partial differential equations (PDEs), with spectral operators offering efficient global mixing across spatial locations. However, many PDEs contain physics-sensitive local structures that are critical to the underlying physical behavior. For example, in Darcy flow, local material interfaces are often reflected by sharp changes in the permeability field and can strongly influence the solution. Existing spectral operators primarily adapt modal mixing based on center-point representations, making them insufficiently responsive to such localized structural variations. We propose the Edge-Conditioned Spectral Operator (ESO), a novel spectral operator framework that modulates global spectral mixing using local edge-wise variations. By incorporating the Pairwise-Variation Modal Mixer (PVMM) to inject local edge information into spectral mode selection, ESO preserves the global approximation capability of spectral neural operators while enabling the learned kernel to adapt to physics-sensitive local structures. Furthermore, we introduce a task-adaptive Physics-Aware Reweighting (PAR) that emphasizes physically important regions, identified by taskspecific physical quantities. Across nine PDE benchmarks, ESO consistently achieves state-of-the-art performance. Visual and region-wise analyses further demonstrate that ESO reduces solution errors near coefficient jumps, high-gradient flow structures, and other physically sensitive regions. The code is available at https://github.com/Tanpig-X/ESO. △ Less

Submitted 7 August, 2026; originally announced August 2026.

Unleashing the Potential of Vision-Language Models for Generalizable AI-Generated Image Detection

Authors: Weihan Cai, Hao Tan, Zichang Tan, Jun Wan, Xinping Gao

Abstract: Recent work has shown that a simple linear probe on frozen representations from modern vision foundation models (VFMs) can achieve state-of-the-art AIGI detection performance, substantially outperforming specialized detectors in challenging in-the-wild scenarios. This finding has established DINOv3 as the dominant foundation-model baseline for subsequent improvements. However, we find that the vis… ▽ More Recent work has shown that a simple linear probe on frozen representations from modern vision foundation models (VFMs) can achieve state-of-the-art AIGI detection performance, substantially outperforming specialized detectors in challenging in-the-wild scenarios. This finding has established DINOv3 as the dominant foundation-model baseline for subsequent improvements. However, we find that the vision-language model Perception Encoder (PE) holds greater potential for AIGI detection, because its language-aligned representation preserves high-level provenance semantics. Specifically, PE exhibits stronger local provenance organization than DINOv3 in its frozen feature space. However, semantic-agnostic linear probing fails to exploit this structure, as PE-Linear still underperforms DINOv3-Linear by 4.1% on In-the-Wild. Based on this observation, we propose Semantic Prototype Calibration (SPC), which constructs category prototypes from forensic semantic information and calibrates them with supervised data. We apply SPC to PE and refer to the resulting detector as PE-SPC. Our analysis shows that this simple design achieves stronger generalization. Across cross-generator, post-processing, and in-the-wild benchmarks, PE-SPC surpasses the previous DINOv3 baseline and achieves new state-of-the-art results. △ Less

Submitted 7 August, 2026; v1 submitted 5 August, 2026; originally announced August 2026.

HoRFFI: High-Openness RF Fingerprint Identification with a Similarity-Enhanced Variational Information Bottleneck

Authors: Shuiguang Zeng, Yuxiang Shen, Yuanyu Zhang, Yulong Shen, Zhiyuan Tan, Houbing Herbert Song

Abstract: Radio frequency fingerprint identification (RFFI) is a promising technique for wireless device authentication. However, practical RFFI systems must enroll newly authorized devices while rejecting previously unseen ones, even when the feature extractor is trained on only a few labeled base-device classes, giving rise to a high-openness RFFI problem. Existing open-set recognition methods typically r… ▽ More Radio frequency fingerprint identification (RFFI) is a promising technique for wireless device authentication. However, practical RFFI systems must enroll newly authorized devices while rejecting previously unseen ones, even when the feature extractor is trained on only a few labeled base-device classes, giving rise to a high-openness RFFI problem. Existing open-set recognition methods typically rely on feature spaces learned from a large and diverse set of known-device classes, limiting their applicability in practical scenarios. To address this challenge, we propose HoRFFI, a high-openness RFFI framework that supports scalable device identification and unknown-device rejection using only a small number of labeled training devices. HoRFFI employs a similarity-enhanced variational information bottleneck (SVIB)-based supervision mechanism, which reduces the encoder's dependence on training-class diversity and learns a more transferable embedding space. This supervision mechanism uses feature-space augmentation and clustering to derive inter-sample similarity information, which provides supplementary supervision for regularizing the embedding space. Experiments on public LoRa and Wi-Fi datasets show that HoRFFI achieves absolute improvements of \(0.112\) and \(0.288\) in novel-class accuracy, respectively, and corresponding absolute AUC improvements of \(0.029\) and \(0.060\) over the best-performing baselines. △ Less

Submitted 5 August, 2026; originally announced August 2026.

Comments: Submitted to IEEE INFOCOM 2026

AgentPanel: Toward a New Paradigm for Human--AI Collaboration in Exploring Scientific Questions

Authors: Zhiyao Cui, Qianyi Wang, Haoyang Yan, Yiqun Zhang, Siyue Ren, Hangfan Zhang, Zelin Tan, Hao Li, Chunjiang Mu, Dexian Cai, Shao Zhang, Chen Zhang, Meng Li, Jianan Chai, Yuting Fan, Zichao Ye, Xiaolei Yang, Xinyao Lu, Yuyang Yu, Wenjie Lou, Xiaosong Wang, Fenghua Ling, Shiyang Feng, Mao Su, Qiaosheng Zhang , et al. (4 additional authors not shown)

Abstract: Identifying promising scientific ideas remains an important challenge in research practice. Researchers commonly rely on small-group discussions or one-to-one interactions with a single large language model, yet these approaches often expose them to only a limited range of perspectives and directions. We present AgentPanel, a multi-agent forum for human--AI collaboration in scientific exploration.… ▽ More Identifying promising scientific ideas remains an important challenge in research practice. Researchers commonly rely on small-group discussions or one-to-one interactions with a single large language model, yet these approaches often expose them to only a limited range of perspectives and directions. We present AgentPanel, a multi-agent forum for human--AI collaboration in scientific exploration. Heterogeneous agents asynchronously discuss scientific questions in a forum-style environment, while researchers can submit questions, browse and organize candidate ideas, engage agents in follow-up interactions, and optionally generate post-hoc summary reports. We evaluate AgentPanel in terms of idea quality, exploration breadth, interaction effectiveness, candidate-selection efficiency, and practical utility. Offline experiments show that AgentPanel outperforms a centralized multi-agent debate baseline. A human study with 20 participants further shows that users value AgentPanel for perspective diversity and exploration support. In experience-based comparisons with commonly used LLM tools, 65\% of participants favored AgentPanel for both breadth of research directions and overall suitability for early-stage exploration. The platform is publicly available at https://agentpanel.cc/. △ Less

Submitted 4 August, 2026; originally announced August 2026.

DiverseDiT++: Quantifying, Analyzing, and Promoting Representation Diversity in Diffusion Transformers

Authors: Binglei Li, Mengping Yang, Zhiyu Tan, Xiaomeng Yang, Zhizhong Huang, Junping Zhang, Hao Li

Abstract: Recent advances in Diffusion Transformers (DiTs) have enabled remarkable progress in visual synthesis, benefiting from their superior scalability. To facilitate DiTs' capability of capturing meaningful internal representations, recent works such as REPA incorporate external pretrained encoders for representation alignment. However, the underlying mechanisms governing representation learning within… ▽ More Recent advances in Diffusion Transformers (DiTs) have enabled remarkable progress in visual synthesis, benefiting from their superior scalability. To facilitate DiTs' capability of capturing meaningful internal representations, recent works such as REPA incorporate external pretrained encoders for representation alignment. However, the underlying mechanisms governing representation learning within DiTs remain poorly understood in the community. To this end, this paper first presents a systematic analysis of the representation dynamics of DiTs via quantifying the diversity of block-wise representations. Specifically, we introduce a novel metric, termed the Weighted Diversity Score (WDS), to measure the representational discrepancies across different blocks. Through extensive investigations on the evolution and influence of internal representations under various settings, we reveal that representation diversity across blocks is a critical factor for effective representation learning in DiTs. More importantly, WDS exhibits a strong correlation with synthesis quality across diverse settings, model scales, and training stages (Pearson's $r=-0.869$ with $\log(\text{FID})$), suggesting its potential as an indicator to reflect model performance and a principled guide for model optimization. Based on this key finding, we propose DiverseDiT++, a novel framework that explicitly promotes diverse representation learning. Concretely, our method incorporates long residual connections to diversify input representations across blocks and a representation diversity loss to encourage blocks to learn distinct features. Extensive experiments on ImageNet $256\times256$ and $512\times512$ demonstrate that our DiverseDiT++ yields consistent performance gains and convergence acceleration when applied to different backbones with various sizes,... △ Less

Submitted 3 August, 2026; originally announced August 2026.

Comments: 35 pages, 32 figures

SKT: Skill-Use Training at Scale via Verified Synthetic Data Generation

Authors: Zelin Tan, Yiqun Zhang, Hao Li, Zhiyao Cui, Hejia Geng, Shao Zhang, Hangfan Zhang, Yang Chen, Xiaosong Wang, Lilong Wang, Zhenfei Yin, Shuyue Hu, Chen Zhang, Lei Bai

Abstract: Agent skills have become an important mechanism for equipping language-model agents with reusable procedural knowledge. However, providing skills alone does not guarantee that current models can effectively identify, apply, and coordinate them. To improve skill-use capabilities, we introduce SKT, a verified data synthesis pipeline that constructs skill-grounded tasks and executable trajectories fr… ▽ More Agent skills have become an important mechanism for equipping language-model agents with reusable procedural knowledge. However, providing skills alone does not guarantee that current models can effectively identify, apply, and coordinate them. To improve skill-use capabilities, we introduce SKT, a verified data synthesis pipeline that constructs skill-grounded tasks and executable trajectories from large collections of agent skills. SKT selects suitable single-skill and multi-skill configurations, synthesizes tasks through rule-based and agent-based verification with feedback-guided repair, and retains only successful trajectories that substantially use every required skill. Using 2,000 public skills, SKT produces 4,000 task packages and 27,164 verified trajectories. Based on the same pipeline and a disjoint test pool, we further construct SkillEval, a held-out executable benchmark for evaluating skill use. Experiments across diverse models, benchmarks, and agent harnesses show that supervised fine-tuning on SKT-generated trajectories consistently improves skill-use performance. Verification ablations, cross-harness evaluation, and scaling experiments further demonstrate that these gains depend on high-quality supervision, extend beyond a single agent interface, and increase with broader skill coverage. Together, these results establish verified data synthesis as an effective and scalable approach for skill-use training. △ Less

Submitted 3 August, 2026; originally announced August 2026.

Comments: 24 pages,8 figures, Version 1

DoubleHelix: Structured Cross-Modal Fusion for Audio-Visual Speech Recognition with LLMs

Authors: Ziwei Cheng, Zhenhua Tan, Zhuomin Zhu

Abstract: Audio-visual speech recognition (AVSR) relies on effective fusion of audio and visual modalities, yet existing approaches treat cross-modal interaction as a single-step operation without structured iterative refinement. We present DoubleHelix, a multimodal fusion framework that reformulates fusion as an iterative cross-modal interaction process with adaptive degradation-aware enhancement. The fram… ▽ More Audio-visual speech recognition (AVSR) relies on effective fusion of audio and visual modalities, yet existing approaches treat cross-modal interaction as a single-step operation without structured iterative refinement. We present DoubleHelix, a multimodal fusion framework that reformulates fusion as an iterative cross-modal interaction process with adaptive degradation-aware enhancement. The framework comprises three components including ReverseParallelHelix for multi-turn structured interaction with learned alignment constraints, QualitySensor for learning degradation-aware gating signals, and HelixReplication for consistency-guided conditional feature enhancement. Experiments on LRS3 demonstrate that DoubleHelix achieves 0.68% WER on clean audio, outperforming previous best results by 5.6% relative improvement under matched backbone settings. Comprehensive ablation studies validate each component contribution, including targeted analysis of design choices such as asymmetric pathway weighting. The framework shows improved robustness under evaluated babble-noise conditions, achieving 11.6% WER at SNR -5dB. △ Less

Submitted 31 July, 2026; originally announced July 2026.

Comments: ACM MM2026 ACCEPTED

MSC Class: 68T10 ACM Class: I.2; I.4

TextCloak: Thwarting Unauthorized LLM Exploitation via RL-Driven Unlearnable Text

Authors: Chengshuai Zhao, Pingchuan Ma, Dawei Li, Bohan Jiang, Zhiyuan Yu, Zhen Tan, Huan Liu

Abstract: The rapid development of Large Language Models (LLMs) has led to significant advances across a wide range of language tasks, while simultaneously raising growing concerns about unauthorized data exploitation and privacy leakage. Unlearnable examples (UEs) offer a promising defense by introducing carefully designed perturbations into data such that models trained on them exhibit degraded utility. H… ▽ More The rapid development of Large Language Models (LLMs) has led to significant advances across a wide range of language tasks, while simultaneously raising growing concerns about unauthorized data exploitation and privacy leakage. Unlearnable examples (UEs) offer a promising defense by introducing carefully designed perturbations into data such that models trained on them exhibit degraded utility. However, existing methods for text protection are primarily designed for classification tasks (e.g., sentiment analysis) in discriminative language models and often rely on injecting class-specific linguistic cues, which limits their effectiveness in the open-ended generation settings of LLMs. In this work, we propose TextCloak, an RL-driven framework for protecting textual data against unauthorized LLM exploitation. TextCloak employs a generative policy that transforms batches of clean text into unlearnable examples while preserving semantic fidelity and linguistic naturalness. To optimize the policy, we introduce GRPO-UE, which rewards generated unlearnable text based on the downstream degradation they induce in fine-tuned surrogate LLMs and updates the generator parameters via group-relative policy optimization. This bi-level optimization enables the generator to discover generalizable protective patterns beyond class-specific cues. Comprehensive experiments on six publicly available datasets and nine state-of-the-art LLMs demonstrate that TextCloak consistently impairs unauthorized fine-tuning while maintaining text utility for legitimate use. Further analyses establish its transferability and robustness across model architectures, training configurations, and adaptive attacks, highlighting its broad applicability as a practical defense against unauthorized LLM exploitation. △ Less

Submitted 30 July, 2026; originally announced July 2026.

Veritas++: Value-aware On-Policy Distillation for Perception-Enhanced AIGI Detection

Authors: Hao Tan, Jun Lan, Zichang Tan, Ajian Liu, Zijian Yu, Chuanbiao Song, Huijia Zhu, Weiqiang Wang, Jun Wan, Zhen Lei

Abstract: The growing capability of image generation models has made synthetic images a routine presence in open media, making robust and generalizable AI-Generated Image (AIGI) detection increasingly essential. While multi-modal large language models (MLLMs) offer a transparent alternative to black-box binary scoring, we observe that current MLLM-based detectors still exhibit notable perception bottlenecks… ▽ More The growing capability of image generation models has made synthetic images a routine presence in open media, making robust and generalizable AI-Generated Image (AIGI) detection increasingly essential. While multi-modal large language models (MLLMs) offer a transparent alternative to black-box binary scoring, we observe that current MLLM-based detectors still exhibit notable perception bottlenecks in capturing fine-grained anomalies. They primarily focus on how visual evidence is organized and synthesized, leaving the intrinsic perception less optimized. To mitigate this gap, we present Veritas++, a perception-enhanced reasoning framework that establishes reliable perception as the foundation of authenticity reasoning. Rather than directly optimizing the model's explanatory ability, we ground AIGI detection on three basic perception abilities, i.e., capturing fine-grained visual details, semantic anomalies and pixel-level differences. Building on this insight, we introduce Perception-oriented Learning (PoRL), which replaces open-ended description supervision with verifiable rewards to explicitly strengthen these capacities. To further integrate enhanced perception with reasoning, we introduce Value-aware On-Policy Distillation (VaOPD), an adaptive distillation mechanism that prioritizes high-value distillation signals over uniform supervision, internalizing perception-aware reasoning through a privileged self-teacher. Extensive experiments across standard, in-the-wild and emerging benchmarks demonstrate that Veritas++ achieves promising generalization. The perception learning effectively bridges the perception gap and yields seamless gains on detection, while VaOPD further enables efficient capability evolvement without sacrificing existing performance. Code and checkpoints are available at https://github.com/EricTan7/VeritasPP. △ Less

Submitted 29 July, 2026; originally announced July 2026.

Monkey King Bang: A Unified Scientific Multimodal Foundation Model

Authors: Hesen Chen, Xinyu Su, Xiaomeng Yang, Yuetan Lin, Zixiong Yang, Junyi An, Fenglei Cao, Yifeng Jiao, Yunqi Zhang, Yuan Cheng, Zhiyu Tan, Hao Li, Libo Wu, Yuan Qi

Abstract: Scientific discovery is increasingly shifting from isolated disciplines to multi-domain reasoning, and AI for science faces a similar transition. Existing systems are either specialised for individual domains or unify scientific data mainly through text tokenisation and prompt-based interfaces, limiting their ability to handle diverse scientific inputs, produce modality-native outputs, and support… ▽ More Scientific discovery is increasingly shifting from isolated disciplines to multi-domain reasoning, and AI for science faces a similar transition. Existing systems are either specialised for individual domains or unify scientific data mainly through text tokenisation and prompt-based interfaces, limiting their ability to handle diverse scientific inputs, produce modality-native outputs, and support joint understanding, reasoning, and generation across scientific domains. We introduce MKB, a unified scientific multimodal model for both understanding and generation, built around a shared Transformer backbone and modality-tailored encoders, adapters, and decoders. MKB covers six scientific branches, including DNA, RNA, proteins, small molecules, earth science, and medical images, and supports native outputs such as biological sequences, molecular strings, meteorological fields, and segmentation masks. Training follows a two-stage modality-then-language curriculum: Stage 1 aligns modality-specific components with the frozen backbone, and Stage 2 consolidates them with the language backbone using mixed scientific and general corpora. Experiments show that MKB achieves competitive scientific understanding across biological and molecular benchmarks, produces high-fidelity native outputs for weather forecasting, biological generation, and medical-image segmentation, and largely retains the general capabilities of its Qwen3-VL backbone. These results demonstrate the feasibility of the proposed paradigm, suggesting that shared-backbone models with modality-tailored components can provide a promising foundation for future cross-domain scientific multimodal exploration. The model and code are publicly available at https://github.com/Shanghai-Academy-of-AI-For-Science/MKB and https://huggingface.co/sais-org/MKB. △ Less

Submitted 17 July, 2026; originally announced July 2026.

Translating AI into scientific impact: Field context, career position, and institutional capability in AI-enabled research

Authors: Zhiyong Tan, Hongkan Chen, Yi Bu

Abstract: Artificial intelligence (AI) is increasingly embedded in scientific research, but its scientific value is unlikely to be distributed evenly. This study examines how AI knowledge integration is associated with scientific impact and asks who benefits from AI-related knowledge in science. Using large-scale bibliographic data, we measure AI integration through references to papers in the OpenAlex Arti… ▽ More Artificial intelligence (AI) is increasingly embedded in scientific research, but its scientific value is unlikely to be distributed evenly. This study examines how AI knowledge integration is associated with scientific impact and asks who benefits from AI-related knowledge in science. Using large-scale bibliographic data, we measure AI integration through references to papers in the OpenAlex Artificial intelligence subfield and link it to five-year citation impact. The results show that AI references are generally associated with higher citation impact, but the returns vary substantially across scientific fields. Career stage also matters: senior scholars benefit more from the extensive margin of AI referencing, whereas junior scholars benefit more from intensive AI referencing and tend to cite newer and higher-impact AI papers. At the institutional level, returns are non-monotonic: institutions with intermediate AI capability achieve the largest proportional gains, while leading AI institutions are more deeply embedded in AI-centered knowledge spaces, attract more AI-related audiences, and more often become substitute citation gateways to cited AI sources. These findings suggest that the value of AI knowledge depends not only on technical capability, but also on translational capacity: the ability to make AI knowledge meaningful, legitimate, and useful across scientific communities. △ Less

Submitted 18 July, 2026; originally announced July 2026.

Bidirectional Resource Scheduling for Disaggregated and Asynchronous RL Post-Training

Authors: Zhiqiang Tan, Maoxin Wang, Sijie Wang, Yiming Yin, Qiang Wang, Xiaowen Chu, Shaohuai Shi

Abstract: It is well established that the reasoning capabilities of large language models (LLMs) can be improved by applying reinforcement learning (RL) in a post-training stage. In a standard RL iteration, the current model (the policy) generates experience through rollouts, and the resulting data is then used to update the policy during training. High-performance RL frameworks such as StreamRL and AReaL e… ▽ More It is well established that the reasoning capabilities of large language models (LLMs) can be improved by applying reinforcement learning (RL) in a post-training stage. In a standard RL iteration, the current model (the policy) generates experience through rollouts, and the resulting data is then used to update the policy during training. High-performance RL frameworks such as StreamRL and AReaL employ a disaggregated architecture and asynchronous rollouts to better exploit both rollout and training resources, thereby increasing overall system throughput. Nonetheless, across varying RL setups (e.g., hardware configurations, model scales, staleness levels, and hyperparameters) and under changing workloads, it remains common for both rollout and training resources to experience idle periods. In this paper, we present BiDiRL, a hybrid time-space multiplexing architecture for asynchronous, disaggregated RL designed to reduce resource idleness. First, we develop a hot-switch runtime that enables rapid switching between rollout and training resources with negligible overhead. Second, we propose a static, scheduling-aware planner based on time-performance modeling that chooses a hot-switch-friendly resource partition, so that rollout and training durations are roughly balanced at a coarse level. Third, at execution time, we introduce a bidirectional scheduler that further exploits runtime bubbles through fine-grained resource switching, allowing the bottleneck stage to temporarily borrow idle resources from the other pool. Across a wide range of workloads, datasets, and models on two 32-GPU testbeds, BiDiRL increases RL training throughput by up to 1.94x compared with RL systems including veRL, AReaL, and ROLL, without affecting convergence behavior. △ Less

Submitted 5 August, 2026; v1 submitted 10 July, 2026; originally announced July 2026.

Comments: 16 pages

PhyMRI-SR: Toward Physics-Aware MRI Image Super-Resolution

Authors: Lihua Wei, Huatong Gao, Jia Gong, Zhiyu Tan, Hao Li, Jun Liu, Zhihua Ren

Abstract: Magnetic resonance imaging (MRI) super-resolution is vital for improving diagnostic accessibility, yet most methods treat it as a deterministic mapping from a fixed low-resolution input to a high-resolution target. This overlooks a key property of MRI acquisition physics: spatial resolution and signal-to-noise ratio (SNR) are inherently coupled, making any given low-resolution scan merely one of m… ▽ More Magnetic resonance imaging (MRI) super-resolution is vital for improving diagnostic accessibility, yet most methods treat it as a deterministic mapping from a fixed low-resolution input to a high-resolution target. This overlooks a key property of MRI acquisition physics: spatial resolution and signal-to-noise ratio (SNR) are inherently coupled, making any given low-resolution scan merely one of many possible realizations under varying acquisition trade-offs. We rethink MRI super-resolution as a physics-aware reconstruction problem, in which the goal is to identify the optimal resolution-SNR configuration and then super-resolve it to obtain high-quality MRI results. A key implication of this formulation is that MRI resolution becomes dynamic rather than fixed. To handle such resolution-heterogeneous inputs, we adapt 2D Gaussian Splatting (2D GS) to MRI by formulating reconstruction as a coordinate-based, resolution-agnostic rendering problem. To further enhance fidelity, we introduce three innovations: (1) a prior-aware Gaussian representation that combines an Anatomical Structure Prior for tissue-specific kernel initialization with an Imaging System Prior that captures hardware characteristics via a covariance dictionary; (2) a physics-constrained signal modeling scheme that predicts intrinsic tissue parameters (proton density rho and effective relaxation rate R2) and synthesizes intensities through governing physical equations, ensuring biophysically plausible contrast; and (3) a meta-learning framework that alleviates paired-data scarcity by pretraining on simulated data and adapting to real-world conditions. Extensive experiments on dynamic-resolution datasets and standard benchmarks demonstrate that our method achieves state-of-the-art performance, highlighting its strong potential for clinical deployment. △ Less

Submitted 7 July, 2026; originally announced July 2026.

Comments: Project Page: https://bio-med-i2-lab.github.io/projects/PhyMRI-SR

InvWeaver: Deductive Feedback for Invariant Synthesis in Interacting-Loop Programs

Authors: Guangyuan Wu, Weining Cao, Zehui Tan, Yuan Yao, Hengfeng Wei, Taolue Chen, Xiaoxing Ma

Abstract: Loop invariant inference is a fundamental yet challenging problem in program verification. Recent LLM-aided guess-and-check techniques have shown strong performance on single-loop programs, but they often struggle with programs containing multiple interacting loops. This paper presents InvWeaver, a neuro-symbolic framework for synthesizing invariants for such programs. The key idea is to expose in… ▽ More Loop invariant inference is a fundamental yet challenging problem in program verification. Recent LLM-aided guess-and-check techniques have shown strong performance on single-loop programs, but they often struggle with programs containing multiple interacting loops. This paper presents InvWeaver, a neuro-symbolic framework for synthesizing invariants for such programs. The key idea is to expose inter-loop dependencies and propagate proof obligations through a combination of loop-level abstraction, obligation-guided inference, and weakest-precondition-based refinement. We evaluate InvWeaver on a comprehensive benchmark suite, including a newly curated dataset derived from classic algorithms. Experimental results show that InvWeaver substantially outperforms existing invariant inference methods, solving 72 out of 82 multi-loop benchmark problems and maintaining strong performance on single-loop tasks. △ Less

Submitted 6 July, 2026; originally announced July 2026.

Ranking the Impact of Contextual Specialization in Neural Speech Enhancement

Authors: Peter Leer, Svend Feldt, Zheng-Hua Tan, Jan Østergaard, Jesper Jensen

Abstract: We systematically investigate neural speech enhancement systems, ranging from very small ($\sim$10\,k parameters) to medium-large ($\sim$2-5\,M parameters), which specialize to acoustic conditions using contextual information such as speaker identity, noise type, speaker gender, spoken language, and SNR. By fine-tuning generalist models on specific data subsets, we find that specializing to a spea… ▽ More We systematically investigate neural speech enhancement systems, ranging from very small ($\sim$10\,k parameters) to medium-large ($\sim$2-5\,M parameters), which specialize to acoustic conditions using contextual information such as speaker identity, noise type, speaker gender, spoken language, and SNR. By fine-tuning generalist models on specific data subsets, we find that specializing to a speaker's identity consistently yields the largest gains in estimated speech intelligibility and quality. In contrast, specializing to SNR, noise type, or gender offers only marginal benefits. Crucially, we show that a small model specialized to both a specific speaker and a specific noise type can match or exceed the performance of a generalist model ten times its size. Further, cross-lingual tests reveal that models specialized to a target language outperform multilingual generalists, suggesting that language is a salient feature for specialization. These findings highlight the potential of small, adaptive models for resource-constrained applications like hearing aids, which specialize on-the-fly to contextual information. △ Less

Submitted 6 July, 2026; originally announced July 2026.

Comments: Accepted to ICASSP 2026

CoGen3D: An Agentic Human-AI Co-Design Pipeline for 3D Asset Generation for Virtual Reality

Authors: Weiwei Jiang, Wanyu He, Zheyu Tan, Zheyuan Kuang, Difeng Yu, Shinobu Hasegawa, Sven Mayer, Zhanna Sarsenbayeva

Abstract: Creating 3D assets for virtual reality requires modeling expertise, which restricts the authorship of immersive experiences. Existing generative AI tools rely on unconstrained, command-driven prompting, lacking the conversational scaffolding needed for users to articulate their intent and validate designs prior to rendering. To address this, we introduce CoGen3D, an agentic human-AI co-design pipe… ▽ More Creating 3D assets for virtual reality requires modeling expertise, which restricts the authorship of immersive experiences. Existing generative AI tools rely on unconstrained, command-driven prompting, lacking the conversational scaffolding needed for users to articulate their intent and validate designs prior to rendering. To address this, we introduce CoGen3D, an agentic human-AI co-design pipeline that proactively guides users through conversational intent elicitation, a concept image confirmation, and image-to-3D generation that directly deploys to immersive scenes. We evaluated this system through a user study (N=120) across six affectively diverse immersive scenes, observing 60 Design group participants who co-created 3D assets for the scenes, and 60 Validation group participants who experienced the scenes with generated assets. Our findings show that co-designed assets are associated with higher scene engagement and shifted affective responses, while participants generally preferred concept images over the final 3D assets, with no increased leniency toward degradation in their own creations. Analysis of the human-AI conversations further shows that target environments shape users' conversational patterns. Our results suggest that our staged, intent-based co-design can democratize virtual reality authoring and shift immersive content creation from technical execution toward collaborative spatial design. △ Less

Submitted 4 July, 2026; originally announced July 2026.

ACM Class: H.5.2; H.5.3; I.3.7; I.4.8

Learning Taxonomic Trees with Hierarchical Representation Regularization for Large Multimodal Models

Authors: Hulingxiao He, Zhi Tan, Yuxin Peng

Abstract: Taxonomies provide key information about the semantic relationships between concepts and the inherent organization of vision and language. Despite their impressive capabilities, large multimodal models (LMMs) often lack taxonomic knowledge, leading to low hierarchical visual recognition (HVR) consistency. These models typically only rely on language modeling objectives during fine-tuning and lack… ▽ More Taxonomies provide key information about the semantic relationships between concepts and the inherent organization of vision and language. Despite their impressive capabilities, large multimodal models (LMMs) often lack taxonomic knowledge, leading to low hierarchical visual recognition (HVR) consistency. These models typically only rely on language modeling objectives during fine-tuning and lack explicit taxonomy-aware regularization. To address this, we propose Hierarchical Representation Regularization ($HiR^2$), a simple plug-and-play regularizer that improves hierarchical consistency in LMMs. Specifically, we introduce a semantic-aware visual tree construction framework that extracts coarse-to-fine visual features from intermediate LLM layers guided by textual cues. The regularizer combines two complementary objectives: a taxonomic entailment loss that enforces hierarchy via hyperbolic entailment cones in the Lorentz model, and a discriminative dispersive loss that promotes angular separation of semantically similar embeddings on the unit sphere without disturbing the radial hierarchical structure. Extensive experiments demonstrate that $HiR^2$ effectively captures taxonomic structures across diverse LMMs and fine-tuning methods. Code is available at https://github.com/PKU-ICST-MIPL/HiR2_ICML2026. △ Less

Submitted 2 July, 2026; originally announced July 2026.

Comments: Published as a conference paper at ICML 2026

Paths and Intersections: Minimum Realization of Okamura-Seymour Instances

Authors: Yu Chen, Pavlo Pylyavskyy, Zihan Tan

Abstract: We study the inverse problem for shortest-path metrics of Okamura-Seymour (OS) instances. Given an OS metric $D$ on a cyclically ordered terminal set $T$, the goal is to find minimum realizations of $D$, where minimum means having the fewest edges among all disk-embedded realizations with the prescribed terminal order. We show that $D$ determines a canonical medial graph template and every minimum… ▽ More We study the inverse problem for shortest-path metrics of Okamura-Seymour (OS) instances. Given an OS metric $D$ on a cyclically ordered terminal set $T$, the goal is to find minimum realizations of $D$, where minimum means having the fewest edges among all disk-embedded realizations with the prescribed terminal order. We show that $D$ determines a canonical medial graph template and every minimum realization is the primal graph of an arrangement of this template. Consequently, the underlying embedded graphs of minimum realizations of $D$ can be recovered, and for each such graph one can efficiently compute edge lengths realizing $D$. Our algorithm follows a recent approach of analyzing graph structures, by viewing graphs as paths and their intersections, which we believe is of independent interest. △ Less

Submitted 2 July, 2026; originally announced July 2026.

Teaching Vision-Language-Action Models What to See and Where to Look

Authors: Yuguang Yang, Canyu Chen, Zhewen Tan, Yizhi Wang, Zichao Feng, Chunyang Liu, Kehua Sheng, Juan Zhang, Linlin Yang, Baochang Zhang, Yan Wang, Bo Zhang, Xianbin Cao

Abstract: Vision-Language-Action (VLA) models have emerged as a promising paradigm for end-to-end autonomous driving. However, existing VLAs' training relies heavily on text-centric visual question answering and chain-of-thought reasoning data, which emphasizes linguistic reasoning rather than action-grounded planning. As a result, the learned representations capture semantic knowledge but lack spatial depe… ▽ More Vision-Language-Action (VLA) models have emerged as a promising paradigm for end-to-end autonomous driving. However, existing VLAs' training relies heavily on text-centric visual question answering and chain-of-thought reasoning data, which emphasizes linguistic reasoning rather than action-grounded planning. As a result, the learned representations capture semantic knowledge but lack spatial dependencies crucial for reliable trajectory prediction. We propose DriveTeach-VLA, a framework that explicitly teaches VLAs what to see and where to look. Driving-aware Vision Distillation (DVD) injects driving-specific perceptual priors into the vision encoder, while 2D Trajectory-Guided Prompts (2D-TGP) provide spatial conditioning aligned with feasible driving trajectories. Together, they form a vision-guided learning pipeline: what to see (DVD pretraining) - where to look (TGP-guided SFT) - how to act (TGP-guided GRPO). DriveTeach-VLA achieves the state-of-the-art performance on NAVSIM and nuScenes. Our code is available at: https://github.com/ShivaTeam/DriveTeach-VLA. △ Less

Submitted 1 July, 2026; originally announced July 2026.

Comments: The paper has been accepted by ECCV 2026

Generative Skill Composition for LLM Agents

Authors: Xinyu Zhao, Zhen Tan, Vaishnav Tadiparthi, Nakul Agarwal, Kwonjoon Lee, Ehsan Moradi Pari, Hossein Nourkhiz Mahjoub, Tianlong Chen

Abstract: Recent LLM agents benefit from skills for solving complex tasks. Skills encapsulate modular packages of procedural knowledge and instructions for performing specialized tasks, such as setting up a sandboxed environment, running a test suite, or refactoring a function across multiple files. As skill libraries grow and become reusable across tasks and domains, selecting an appropriate skill composit… ▽ More Recent LLM agents benefit from skills for solving complex tasks. Skills encapsulate modular packages of procedural knowledge and instructions for performing specialized tasks, such as setting up a sandboxed environment, running a test suite, or refactoring a function across multiple files. As skill libraries grow and become reusable across tasks and domains, selecting an appropriate skill composition has emerged as a central bottleneck. Existing approaches fall into two categories. One exposes the agent's reasoning to the entire skill collection; the other performs skill retrieval via embeddings or LLM-based rerankers. Both provide useful insights; however, they miss the structural nature of skill composition, which is a joint decision over which skills, how many, and in what order -- three dimensions that cannot be decoupled. We formalize this as structured skill composition: given a task and a skill library, predict an executable skill plan that jointly specifies the activated subset, count, and execution order. We propose SkillComposer, which instantiates structured skill composition as task-conditioned skill sequence prediction. SkillComposer uses a constrained autoregressive decoder over skill identifiers, so subset, count, and order emerge jointly from a single decoding pass, and dependencies between successive skills are captured naturally. We build a training set of task-composition pairs from a real, human-curated skill library. We then evaluate SkillComposer along two axes: composition quality on a held-out test set, and downstream task success on SkillsBench across two production-grade coding agents. On GPT-5.2-Codex, Gemini-3-Pro-Preview, SkillComposer raises the pass rate by +23.1, +18.2pp over the no-skill baseline, surpassing top-3 retrieval and matching the gold-skill retrieval upper bound at lower prompt-token cost. △ Less

Submitted 30 June, 2026; originally announced June 2026.

Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent

Authors: Lei Bai, Zongsheng Cao, Yang Chen, Zhiyao Cui, Shangheng Du, Yue Fan, Shiyang Feng, Zijie Guo, Haonan He, Liang He, Xiaohan He, Shuyue Hu, Yusong Hu, Songtao Huang, Yichen Jiang, Hao Li, Xin Li, Dahua Lin, Weihao Lin, Fenghua Ling, Dongrui Liu, Zhuo Liu, Wenjie Lou, Runmin Ma, Chunjiang Mu , et al. (28 additional authors not shown)

Abstract: We introduce Agents-A1, a 35B Mixture-of-Experts Agentic Model that reaches trillion-parameter-level performance by scaling the agent horizon. We investigate agent-horizon scaling from two perspectives: scaling long-horizon trajectories and scaling heterogeneous agent abilities. To support this goal, we build a long-horizon knowledge-action infrastructure that connects external knowledge, actions,… ▽ More We introduce Agents-A1, a 35B Mixture-of-Experts Agentic Model that reaches trillion-parameter-level performance by scaling the agent horizon. We investigate agent-horizon scaling from two perspectives: scaling long-horizon trajectories and scaling heterogeneous agent abilities. To support this goal, we build a long-horizon knowledge-action infrastructure that connects external knowledge, actions, observations, and verifier outcomes, producing agentic trajectories with an average length of 45K tokens. Based on this, we train Agents-A1 with a three-stage recipe. First, we perform full-domain supervised fine-tuning to align the base model with broad agentic behaviors. Second, we train domain-level teacher models to capture specialized expertise in each domain. Third, we propose a multi-teacher domain-routed on-policy distillation with salient vocabulary alignment to improve knowledge transfer efficiency across different domains, unifying six heterogeneous domains into one deployable student model. Agents-A1 achieves strong and broad performance for long-horizon agent benchmarks. Compared with 1T-parameter model such as Kimi-K2.6 and DeepSeek-V4-pro, Agents-A1 achieves leading results on SEAL-0 (56.4), IFBench (80.6), HiPhO (46.4), FrontierScience-Olympiad (79.0), and MolBench-Bind (56.8), and remains highly competitive on SciCode (44.3), HLE (47.6) and BrowseComp (75.5). We hope this work provides the community with a practical path for scaling the horizon using a 35B agent that can reach or match the performance of 1T models on long-horizon tasks. △ Less

Submitted 13 July, 2026; v1 submitted 29 June, 2026; originally announced June 2026.

Comments: The model checkpoints and evaluation codebase are available at https://huggingface.co/collections/InternScience/agents-a1 and https://github.com/InternScience/Agents-A1

Parametric Skills

Authors: Xuan Zhao, Haonan He, Qingyu Yang, Minglei Li, Jingqi Ye, Zelin Tan, Bo Wan, Peng Ye

Abstract: Since intelligence fundamentally relies on efficient skill acquisition (Chollet, 2019), the ability to leverage skills is critical. For LLMs, skills, manually authored or extracted from task trajectories, are textual recipes encoding mature problem-solving experience and are critical to agentic capabilities. Despite widespread deployment, their utility is limited by the model's ability to comprehe… ▽ More Since intelligence fundamentally relies on efficient skill acquisition (Chollet, 2019), the ability to leverage skills is critical. For LLMs, skills, manually authored or extracted from task trajectories, are textual recipes encoding mature problem-solving experience and are critical to agentic capabilities. Despite widespread deployment, their utility is limited by the model's ability to comprehend and follow skill instructions, especially under complex and long-context scenarios, where key instructions are difficult to locate and adhere to. To address this limitation, we propose ParametricSkills, a framework that can convert free-form textual skills into parameters at test time, enabling context-free skill exploitation. Specifically, we first construct a large-scale, high-quality skill library, and synthesize single-turn and multi-turn skill exploitation trajectories built around these skills with OpenCode. Using these data, we then train a hypernetwork that parameterizes both the skill content and the test-time exploitation methodology by receiving textual skills and converting them into LoRA adapters. Experimental results on six complex software engineering (SWE) subtasks demonstrate that, the proposed ParametricSkills averagely outperforms in-context learning by 6.44 points as judged by DeepSeek-V4-Flash, while also achieving significantly higher BERT Score and F1 score, confirming its effectiveness. Beyond performance, we further find that parametric skills, being inherently accumulative, offer a preliminary yet promising avenue toward test-time continual learning. △ Less

Submitted 29 June, 2026; originally announced June 2026.

Comments: Preprint, Under Review

Demystifying the Design Space and Best Practices for Heterogeneous LLM Inference and Serving

Authors: Zhixin Wang, Zhengbo Wang, Fangcheng Fu, Yinhui Lu, Jinlong Hou, Yijie Chen, Xiaowei Shen, He Liu, Xiangbin Li, Jun Chen, Ruya Gu, Dian Wang, Zhou Tan, Yuan Cheng, Hongzhou Zhang, Xiangjun Huang, Ping Zhang, Xiaohe Hu

Abstract: Heterogeneous prefill-decode (PD) inference is now in production: prefill on cost-efficient or supply-available accelerators, decode on bandwidth-strong ones, and KV state crossing mixed interconnects in mixed numerical formats. Each deployment makes these decisions on its own. What is missing is the picture across configurations-which decisions must be made jointly at the PD boundary, and which c… ▽ More Heterogeneous prefill-decode (PD) inference is now in production: prefill on cost-efficient or supply-available accelerators, decode on bandwidth-strong ones, and KV state crossing mixed interconnects in mixed numerical formats. Each deployment makes these decisions on its own. What is missing is the picture across configurations-which decisions must be made jointly at the PD boundary, and which can be made independently. We propose a design space organized along four design axes-accelerator, precision, interconnect, and KV residency and the workload regime (stage pressure) they respond to. We show that only a subset of interactions among these factors become binding constraints once PD inference becomes heterogeneous. These interactions surface through three recurring boundary decisions: compute placement, KV representation, and KV ownership. The resulting analysis yields concrete guidance. Precision policy belongs to runtime roles rather than to a single system-wide setting, because the same low-bit format relieves different bottlenecks on each side of the boundary. KV transfer engines move bytes rather than tensor semantics, making representation compatibility an explicit boundary concern whenever producer and consumer differ. The KV handoff also carries a lifecycle-reservation, release, and failure recovery-that spans prefill and decode and requires explicit ownership. Two further interactions remain open. Cross-vendor and interconnect-related claims are stated as design guidance grounded in industrial deployment observations and source-code inspection of the runtimes involved. △ Less

Submitted 29 June, 2026; v1 submitted 28 June, 2026; originally announced June 2026.

Model Merging to Evolution: Parameter Space Exploration for Expert Models

Authors: Chao Wang, Yuchen Guo, Zheng Tan, Guanchun Wang, Yanbiao Ma, Qiqi Duan, Peng Wu

Abstract: Model merging integrates the capabilities of multiple expert models to create strong models for multiple tasks without additional training, thereby reducing computational resource requirements. However, existing methods operate within the convex combination space of expert models, failing to explore high-performance regions outside this space. This paper proposes the MERGEvolve framework, which un… ▽ More Model merging integrates the capabilities of multiple expert models to create strong models for multiple tasks without additional training, thereby reducing computational resource requirements. However, existing methods operate within the convex combination space of expert models, failing to explore high-performance regions outside this space. This paper proposes the MERGEvolve framework, which unifies model merging and evolution within an evolution strategy by treating the merged model as the initialization for evolutionary exploration of the parameter space. During the merging phase, expert models act as deterministic sources to build a strong initial point. The evolution phase then explores the parameter space using random noise. Theoretical analysis shows that MERGEvolve explores regions outside the convex combination space. Extensive experiments on single-task and multi-task benchmarks demonstrate that MERGEvolve consistently achieves performance competitive with advanced model merging baselines. Ablation studies confirm that a high-quality initial point is critical for efficient exploration of the parameter space. △ Less

Submitted 17 June, 2026; originally announced June 2026.

Comments: Chao Wang and Yuchen Guo contributed equally

Paths and Intersections: Recognizing Outerplanar Metrics

Authors: Yu Chen, Zihan Tan

Abstract: We study the following distance realization problem: given a metric $D$ on a set $T$ of terminals, does there exist an (edge-weighted) outerplanar graph $G$, such that $T\subseteq V(G)$, and for every pair $t,t'\in T$, $\textsf{dist}_G(t,t')=D(t,t')$? We first prove that there is no ``local characterization'', forming a contrast with trees and Okamura-Seymour instances. Our main result is an effic… ▽ More We study the following distance realization problem: given a metric $D$ on a set $T$ of terminals, does there exist an (edge-weighted) outerplanar graph $G$, such that $T\subseteq V(G)$, and for every pair $t,t'\in T$, $\textsf{dist}_G(t,t')=D(t,t')$? We first prove that there is no ``local characterization'', forming a contrast with trees and Okamura-Seymour instances. Our main result is an efficient algorithm for this problem whose running time is polynomial in $|T|$. Both our proof and our algorithm utilize a recent new approach of analyzing graph structures, by viewing graphs as paths and their intersections, which we believe is of independent interest. △ Less

Submitted 4 July, 2026; v1 submitted 24 June, 2026; originally announced June 2026.

Accelerating Disaggregated RL for Visual Generative LLMs with Diffusion-Based Parallelism and Trainer-Assisted Generation

Authors: Sijie Wang, Zhengyu Qing, Zhiqiang Tan, Yiming Yin, Yeqing Zhang, Yaoyuan Wang, Qiang Wang, Xiaowen Chu, Shaohuai Shi

Abstract: Reinforcement learning (RL) has become a dominant post-training paradigm, driving the emergence of high-performance RL systems such as veRL for autoregressive large language models (LLMs). In parallel, diffusion-oriented RL algorithms, e.g., DanceGRPO and FlowGRPO, have rapidly expanded the scope of RL from language reasoning to diffusion-based visual and flow-based generation. However, efficient… ▽ More Reinforcement learning (RL) has become a dominant post-training paradigm, driving the emergence of high-performance RL systems such as veRL for autoregressive large language models (LLMs). In parallel, diffusion-oriented RL algorithms, e.g., DanceGRPO and FlowGRPO, have rapidly expanded the scope of RL from language reasoning to diffusion-based visual and flow-based generation. However, efficient RL systems for diffusion generative LLMs remain underexplored. Existing implementations, e.g., veRL-Omni, still rely on colocated execution, which simplifies synchronization but couples rollout and training resources, limits heterogeneous deployment, and constrains independent scaling. To this end, we introduce DigenRL, a disaggregated RL framework for diffusion-based generative LLMs that supports flexible resource allocation, accommodates heterogeneous GPUs, and facilitates efficient task scheduling. To maximally reduce the execution bubbles in the disaggregated architecture, we propose: 1) a generation-axis pipeline (GAP) and time-step parallelism (TSP) in the diffusion architecture to enable finer-grained pipelining between rollout and training; 2) an elastic trainer-assisted generation (TAG) approach to enable the trainer GPU resources to dynamically assist in executing rollout generations; and 3) a tightly one-step constrained asynchronous strategy to further utilize the tail bubble in the pipeline. Extensive experiments are conducted on three hardware testbeds with 16-32 GPUs using HunyuanVideo-13B, Wan2.1-14B, FLUX.1-12B, and QwenImage-20B generative models. Experimental results show that DigenRL achieves 1.56-2.10x throughput improvements over state-of-the-art diffusion RL systems, veRL-Omni and GenRL. △ Less

Submitted 24 June, 2026; v1 submitted 23 June, 2026; originally announced June 2026.

Comments: Withdrawn by the authors pending resolution of intellectual property and institutional disclosure requirements

World Action Models: A Survey

Authors: Qiuhong Shen, Shihua Zhang, Yue Liao, Qi Li, Zhenxiong Tan, Shizun Wang, Shuicheng Yan, Xinchao Wang

Abstract: World Action Models (WAMs) are embodied predictive-action models that make a forecast of the future available to action. Recent WAMs repurpose large video generation models, and a parallel line relies on language or vision-language backbones without a video-generation core. This rapid expansion has blurred the boundary among broad world models, video generation models, action-grounded video world… ▽ More World Action Models (WAMs) are embodied predictive-action models that make a forecast of the future available to action. Recent WAMs repurpose large video generation models, and a parallel line relies on language or vision-language backbones without a video-generation core. This rapid expansion has blurred the boundary among broad world models, video generation models, action-grounded video world models, Vision-Language-Action policies, and WAMs. This survey gives the field a common account. It first clarifies these boundaries, then organizes existing works through two complementary views. The first view asks what each method is required to generate, spanning rendered futures, latent futures, and video-generation-free action reasoning. The second view decomposes each method by predictive substrate, backbone, action coupling, and deployment regime. This anatomy supports a unified discussion of interactability, causality, persistence, physical plausibility, and generalization, followed by data, evaluation, and open challenges. Across these axes, a consistent design pattern emerges: WAMs are not simply video generators with action heads, but predictive-action methods whose design choices trade representational richness against compute, memory, latency, and action-label cost. The field is moving toward methods that generate less of the future while preserving what control requires. The survey homepage is available at https://world-action-models.github.io/. △ Less

Submitted 18 June, 2026; originally announced June 2026.

Comments: 57 pages, 6 figures

HydraHead: From Head-Level Functional Heterogeneity to Specialized Attention Hybridization

Authors: Zhentao Tan, Wei Chen, Jingyi Shen, Yao Liu, Xu Shen, Yue Wu, Jieping Ye

Abstract: The quadratic complexity of attention poses a critical bottleneck for long-context processing, spurring interest in hybrid attention designs. Most open-source hybrid models adopt a layer-wise strategy. Yet, prior work has noted the inherent difficulty of integrating Linear Attention (LA) with Full Attention (FA), suggesting that the design space of attention hybridization remains underexplored. To… ▽ More The quadratic complexity of attention poses a critical bottleneck for long-context processing, spurring interest in hybrid attention designs. Most open-source hybrid models adopt a layer-wise strategy. Yet, prior work has noted the inherent difficulty of integrating Linear Attention (LA) with Full Attention (FA), suggesting that the design space of attention hybridization remains underexplored. To probe this space, we conduct interpretability analysis and observe that layers exhibit block-wise functional similarity, while individual heads within the same layer display distinct functional specialization despite sharing input features. This head-level heterogeneity suggests that the head dimension provides a natural and principled granularity for fusing heterogeneous attention signals. Building on this insight, we introduce HydraHead, a novel architecture that hybridizes FA and LA along the head axis. HydraHead features two key innovations: (1) an interpretability-driven selection strategy that identifies retrieval-critical heads and preserves FA only for them, and (2) a scale-normalized fusion module that reconciles the distributional gap between FA and LA head outputs. By leveraging a three-stage transfer pipeline with parameter reuse and distillation, we achieve high-performance hybrid models with minimal training overhead. Under a unified training setup, HydraHead outperforms other hybrid designs in long-context tasks while maintaining strong general reasoning. With interpretability-driven head selection, it matches a 3:1 layer-wise hybrid's long-context performance at a 7:1 LA-to-FA ratio. Crucially, trained on only 15B tokens, HydraHead achieves over 69% improvement over the baseline at 512K context length, approaching Qwen3.5, a leading model of comparable size with a native context length of 256K. This highlights the significant scaling potential of head-level hybridization. △ Less

Submitted 18 June, 2026; originally announced June 2026.

FuseChain: Runtime Evidence Reconstruction for Software Supply-Chain Attacks

Authors: Zhuoran Tan, Yutian Tang, Jeremy Singer, Christos Anagnostopoulos, Ke Xiao

Abstract: Software supply-chain (SSC) attacks are increasingly multi-stage, cross-source, and temporally distributed. A single attack campaign may leave weak and fragmented traces across multi-source telemetry that captures different granularities and perspectives of runtime behavior. Existing runtime detection systems often analyze these sources independently, making it difficult to identify low-frequency… ▽ More Software supply-chain (SSC) attacks are increasingly multi-stage, cross-source, and temporally distributed. A single attack campaign may leave weak and fragmented traces across multi-source telemetry that captures different granularities and perspectives of runtime behavior. Existing runtime detection systems often analyze these sources independently, making it difficult to identify low-frequency attack evidence or reconstruct the temporal context in which it appears. We present FUSECHAIN, a runtime detection framework that represents multi-source software supply-chain telemetry as a temporal heterogeneous provenance graph over a unified event-time axis. By aligning package/runtime traces, process events, network telemetry, DNS/HTTP metadata, and security alerts on a unified temporal graph, FuseChain captures cross-source dependencies and sparse attack evidence that may be ambiguous within any individual source. It learns anomaly-centric temporal representations from benign-prefix telemetry and performs deployable attack-stage reconstruction through a lightweight decoder on top of a frozen anomaly backbone. Our experiments show that jointly optimizing anomaly detection and stage prediction is ineffective under sparse and imbalanced runtime supply-chain telemetry. Across seven SSC attack scenarios, FuseChain improves deployable stage reconstruction from 0.369 to 0.881 Stage Recall@500 with a frozen-backbone decoder, while adaptive retrieval further increases observable-stage recall from 0.524 to 0.655 without modifying the detector. These results highlight the deployable value of decoupling runtime SSC anomaly detection from downstream attack-stage interpretation. △ Less

Submitted 14 June, 2026; originally announced June 2026.

Competitive Analysis for Online Fair Division under Multiple Fairness Notions

Authors: Tianqi Chen, Zhiyi Tan

Abstract: We study the online fair division of indivisible items with additive utilities, where items arrive sequentially and must be irrevocably allocated upon arrival. Considering various fairness notions, we focus on designing online algorithms that produce fair or approximately fair allocations for any instance. We measure algorithm performance using the competitive ratio, defined as the worst-case rati… ▽ More We study the online fair division of indivisible items with additive utilities, where items arrive sequentially and must be irrevocably allocated upon arrival. Considering various fairness notions, we focus on designing online algorithms that produce fair or approximately fair allocations for any instance. We measure algorithm performance using the competitive ratio, defined as the worst-case ratio between the fairness guarantee achieved by the online algorithm and that of an optimal offline allocation with full knowledge of future arrivals. We examine a broad spectrum of models, including the allocation of goods or chores, normalized versus non-normalized utilities, and identical versus general utility functions. We address the majority of the unresolved cases by providing online algorithms or proving the limits of the competitive ratio achievable by online algorithms. In most cases, the algorithms are shown to be optimal. △ Less

Submitted 13 June, 2026; originally announced June 2026.

FairGen: Preference-Aligned Diffusion for Demographically Equitable Medical Image Synthesis

Authors: Zhimin Li, Ruichen Zhang, Zhen Tan, Howard J Aizenstein, Jingtong Hu, Tianlong Chen

Abstract: Medical imaging is central to modern diagnostics, and artificial intelligence (AI) systems are increasingly used to support image-based analysis by improving efficiency, accuracy, and access to care. However, inequities in healthcare access and differential disease prevalence create severe demographic imbalances in clinical image data. Such imbalances are compounded by the fact that diseases can m… ▽ More Medical imaging is central to modern diagnostics, and artificial intelligence (AI) systems are increasingly used to support image-based analysis by improving efficiency, accuracy, and access to care. However, inequities in healthcare access and differential disease prevalence create severe demographic imbalances in clinical image data. Such imbalances are compounded by the fact that diseases can manifest with distinct features across demographic groups, rendering certain phenotypic presentations naturally rare. AI models trained on such imbalanced data risk perpetuating diagnostic bias and widening healthcare disparities. Here we introduce FairGen, a fairness-aware diffusion framework that synthesizes demographically balanced medical images while preserving pathology-relevant visual features. By embedding physician-aligned preferences into the generation process, FairGen improves subgroup coverage during synthesis and downstream classification. Applied to dermatology, radiology, and neuroimaging benchmark tasks, FairGen achieves fairness improvements of 95.9% for skin images, 80.0% for chest radiography, and 35.2% for brain MRI, while maintaining competitive diagnostic accuracy relative to models trained on original clinical data. Clinician-facing expert review and external validation on independent cohorts further support that these gains extend beyond standard fidelity metrics and are not confined to the original in-distribution datasets. △ Less

Submitted 1 June, 2026; originally announced June 2026.

Comments: Accepted for publication in npj Digital Medicine. 20 pages, 6 figures

Does AI Reviewer See the Full Picture? Attacking and Defending Multimodal Peer Review

Authors: Xinyu Zhao, Rana Muhammad Shahroz Khan, Zhen Xu, Zhen Tan, Tianlong Chen

Abstract: The integration of Large Language Models (LLMs) and Multimodal LLMs (MLLMs) into scientific peer-review workflows introduces novel and significant risks for adversarial manipulation, especially given the multimodal nature of scientific papers where figures, not just text, convey core evidence. This creates a significant gap: current robustness studies on AI peer-review are overwhelmingly text-only… ▽ More The integration of Large Language Models (LLMs) and Multimodal LLMs (MLLMs) into scientific peer-review workflows introduces novel and significant risks for adversarial manipulation, especially given the multimodal nature of scientific papers where figures, not just text, convey core evidence. This creates a significant gap: current robustness studies on AI peer-review are overwhelmingly text-only. Moreover, the problem is distinct from standard jailbreaking, as a peer-review attack seeks to induce a domain-specific, targeted failure (e.g., "inflate this score") rather than a general safety policy violation, for which no practical defenses exist. To address this, we introduce PaperGuard, the first comprehensive benchmark designed to systematically evaluate and defend AI-generated peer-review against these domain-specific, cross-modal attacks. Our framework is built on three pillars: (1) a new multimodal peer-review dataset spanning multiple scientific domains; (2) a unified suite of attacks, including black-box prompt injections and white-box perturbations, specifically designed to target both text (GCG) and figures (PGD); and (3) a practical defense, motivated by the long-context challenge of academic papers, that uses chunk-based embedding search to efficiently localize and mitigate harmful instructions. Our extensive experiments, conducted across state-of-the-art models, confirm that AI reviewers are pervasively vulnerable. PaperGuard establishes the foundational benchmark, protocols, and actionable defense necessary to pioneer trustworthy, attack-resilient AI-assisted scholarly reviewing. △ Less

Submitted 24 June, 2026; v1 submitted 10 June, 2026; originally announced June 2026.

Comments: Accepted to ICML 2026, Project Page: https://paper-guard.github.io/

JOIN: Anchor-Grasp-Conditioned Joining via Opposition, Inference, and Navigation for Bimanual Assistive Manipulation

Authors: Drake Moore, Matt Cheng, Xiang Zhi Tan, Taşkın Padır

Abstract: Assistive mobility and manipulation platforms have received increasing attention as a means of restoring independence to individuals with disabilities. While effective for many basic activities of daily living (ADLs), a significant percentage of everyday tasks such as opening a jar, pouring a liquid, lifting a tray, or basic meal preparation, is fundamentally bimanual and remains out of reach for… ▽ More Assistive mobility and manipulation platforms have received increasing attention as a means of restoring independence to individuals with disabilities. While effective for many basic activities of daily living (ADLs), a significant percentage of everyday tasks such as opening a jar, pouring a liquid, lifting a tray, or basic meal preparation, is fundamentally bimanual and remains out of reach for any single-arm system. Adding a second arm to a wheelchair is impractical, due to the additional power draw, cost, and the loss of space required for transfers and mobility. We instead propose a heterogeneous, on-demand bimanual system, in which a wheelchair-mounted anchor arm is joined when needed by a summoned mobile manipulator that serves as a complement arm. The central technical problem, which we call bimanual joining, is conditional: the anchor has already committed to a grasp, and the complement arm must choose where to stand and what to grasp to complete the task. We formulate bimanual joining as a three-phase decomposition (plan, drive, grasp) and show that a vision-language model (VLM), coupled with standard geometric tools, provides task-level knowledge sufficient to solve a representative class of bimanual ADLs. Our system JOIN, contributes (i) a wheelchair-referenced opposition score, and (ii) task-conditioned directional manipulability. We evaluate JOIN on a Kinova Gen3 anchor and a Hello Robot Stretch~3 complement on representative same-object and different-object tasks. JOIN accomplished more attempts (19/20) than state-of-the-art methods (14/20) and required markedly less correction by the operator. △ Less

Submitted 9 June, 2026; originally announced June 2026.

Comments: Xiang Zhi Tan and Taşkın Padır share equal advising

ConMem: Structured Memory-Guided Adaptation in Training-Free Multi-Agent Systems

Authors: Zhixun Tan, Qiang Chen, Tairan Huang, Xiu Su, Yi Chen

Abstract: Recent advances have improved the adaptive capabilities of LLM-based multi-agent systems (MAS) through memory-, skill-, and learning-based approaches, yet these approaches remain challenged by noisy trajectories, insufficient modeling of memory-skill relations, and reliance on additional training or high-quality supervision. To address these limitations, we propose ConMem, a relation-aware and tra… ▽ More Recent advances have improved the adaptive capabilities of LLM-based multi-agent systems (MAS) through memory-, skill-, and learning-based approaches, yet these approaches remain challenged by noisy trajectories, insufficient modeling of memory-skill relations, and reliance on additional training or high-quality supervision. To address these limitations, we propose ConMem, a relation-aware and training-free framework that enables efficient multi-agent adaptation through cross-experience coordination. Specifically, ConMem distills historical interaction trajectories into structured memory cards to capture reusable strategies and cues, organizing them into a relation-aware memory graph. At runtime, ConMem retrieves cards according to task needs and coordinates them through the card graph to resolve strategy conflicts and recover their dependencies. Combined, these modules yield structured and relation-aware guidance, enabling robust, lightweight adaptation in multi-agent systems without additional training. Extensive experiments across multiple benchmarks and mainstream MAS architectures show consistent gains over existing memory architectures, with improved inference-time efficiency through pruning more than 50% of expanded candidates and reducing planning overhead by over 80%. Our codes are available at https://anonymous.4open.science/r/ConMemCode △ Less

Submitted 7 June, 2026; originally announced June 2026.

IRAF: Interference-Resilient Adaptive Fusion for Noise-Robust End-to-End Full-Duplex Spoken Dialogue Systems

Authors: Tao Zhong, Jiajun Deng, Nikita Kuzmin, Yinke Zhu, Tianxiang Cao, Tristan Tsoi, Zhili Tan, Simon Lui, Xunying Liu

Abstract: Full-duplex spoken dialogue models allow voice agents to listen and speak concurrently, enabling natural interaction with real-time overlap. However, end-to-end dual-channel models that jointly encode user and agent streams may degrade in realistic acoustic environments: interfering speakers leaking into the user microphone can be encoded as part of the user query, corrupting the LLM's conditionin… ▽ More Full-duplex spoken dialogue models allow voice agents to listen and speak concurrently, enabling natural interaction with real-time overlap. However, end-to-end dual-channel models that jointly encode user and agent streams may degrade in realistic acoustic environments: interfering speakers leaking into the user microphone can be encoded as part of the user query, corrupting the LLM's conditioning and causing unstable turn-taking and reduced response quality. We propose Interference-Resilient Adaptive Fusion (IRAF), a lightweight, streaming-compatible module that modulates the contribution of user audio to the LLM frame by frame. IRAF predicts a scalar reliability gate from target-speaker and user audio embeddings and rescales user representations before fusion with agent embeddings. Experiments on MS-MARCO and InstructS2S-200K show consistent gains in response quality and full-duplex interaction under interfering-speaker conditions. △ Less

Submitted 4 June, 2026; originally announced June 2026.

Towards Worst-case Hardness for Low-Noise LPN

Authors: Divesh Aggarwal, Rishav Gupta, Hai Hoang Nguyen, Kel Zin Tan, Prashant Nalini Vasudevan

Abstract: The hardness of the Learning Parity with Noise (LPN) problem is a foundational assumption in cryptography, forming the basis of constructions ranging from symmetric-key primitives to public-key encryption and beyond. A central open question is whether the average-case hardness of LPN can be based on worst-case complexity assumptions, as has been achieved for the analogous Learning With Errors (LWE… ▽ More The hardness of the Learning Parity with Noise (LPN) problem is a foundational assumption in cryptography, forming the basis of constructions ranging from symmetric-key primitives to public-key encryption and beyond. A central open question is whether the average-case hardness of LPN can be based on worst-case complexity assumptions, as has been achieved for the analogous Learning With Errors (LWE) problem. Existing worst-case-to-average-case reductions for LPN [BLVW19, YZ21] rely on statistical smoothing of linear codes, which inherently limits the resulting average-case hardness to noise rates as large as $1/2 - 1/\mathrm{poly}(n)$, which is insufficient for public-key applications. We explore a new approach towards obtaining such reductions: rather than requiring that random sparse combinations of the rows of the generator matrix of a code be statistically close to uniform, we only require that they be computationally indistinguishable from uniform. This leads to a clean win-win structure: we show that any efficient LPN solver can be transformed into a pair of efficient algorithms $(S, D)$ such that for every matrix $A$ of appropriate dimensions over $\mathbb{F}_2$, either $S$ decodes the code generated by $A$ from random noise, or $D$ distinguishes random noisy codewords of the dual of this code from uniform. By instantiating this reduction with appropriate parameters, we obtain the average-case hardness of LPN with inverse-polynomial noise rate $n^{-α}$ for any constant $α< 1$, assuming the worst-case simultaneous hardness of decoding a code from random noise and distinguishing random noisy codewords of its dual from uniform. In particular, setting $α= 1/2$, our reduction yields LPN hardness in the parameter regime required for Alekhnovich's construction of public-key encryption [Ale03], a regime that was previously inaccessible via worst-case reductions. △ Less

Submitted 4 June, 2026; originally announced June 2026.

RealClawBench: Live OpenClaw Benchmarks from Real Developer-Agent Sessions

Authors: Zongwei Lv, Zhewen Tan, Yaoming Li, Yilun Yao, Yuxuan Tian, Lin Sun, Xiangzheng Zhang, Weihong Lin, Tong Yang, Guangxiang Zhao

Abstract: Agent benchmarks should reflect what users actually ask deployed agents to do, yet existing benchmarks often miss key realism properties of real developer-agent sessions. We introduce RealClawBench, a live benchmark framework built from real OpenClaw sessions to capture the distribution, diversity, and real-world difficulty of deployed agent use. Real user requests are challenging to benchmark bec… ▽ More Agent benchmarks should reflect what users actually ask deployed agents to do, yet existing benchmarks often miss key realism properties of real developer-agent sessions. We introduce RealClawBench, a live benchmark framework built from real OpenClaw sessions to capture the distribution, diversity, and real-world difficulty of deployed agent use. Real user requests are challenging to benchmark because they often depend on local execution environments, involve implicit or underspecified intent, and require nontrivial verification. RealClawBench addresses these challenges with two core mechanisms: reconstructed execution environments and deterministic verifiable scorers, which together convert real sessions into reproducible, automatically scored tasks. The resulting release contains 281 executable tasks sampled from a much larger real-session pool while preserving the source distribution, with maximum final-vs-source Jensen-Shannon divergence of 0.0448. Evaluating 14 contemporary models shows that the best system solves only 65.8% of tasks, revealing substantial headroom on realistic developer-agent workloads. By turning real deployed sessions into controlled evaluation instances, RealClawBench provides a practical path toward benchmarks that better measure agent capability in actual use. Code is available at:https://anonymous.4open.science/r/real-claw-bench-582B. △ Less

Submitted 5 June, 2026; v1 submitted 2 June, 2026; originally announced June 2026.

Comments: 19 pages, 5 figures, 8 tables

From 'What' to 'How' and 'Why': Sharing LLM-Generated Retrospective Summaries of Older Adults' Passive Tracking Data with Remote Family Members

Authors: Jiachen Li, Reina Szeyi Chan, Akshat Choube, Xiang Zhi Tan, Elizabeth Mynatt, Varun Mishra

Abstract: With the growing prevalence of modern ubiquitous computing technologies, multi-modal tracking systems hold promise for providing timely awareness and reassurance to stakeholders such as remote family members (RFMs) of older adults, who play a central role in care coordination. However, combining heterogeneous data streams into high-level, meaningful content - such as retrospective summaries - rema… ▽ More With the growing prevalence of modern ubiquitous computing technologies, multi-modal tracking systems hold promise for providing timely awareness and reassurance to stakeholders such as remote family members (RFMs) of older adults, who play a central role in care coordination. However, combining heterogeneous data streams into high-level, meaningful content - such as retrospective summaries - remains challenging. While recent work has demonstrated the promise of large language models (LLMs) for interpreting multi-modal tracking data, less attention has been given to generating narrative accounts for stakeholders like RFMs, who possess rich personal knowledge of older adults and strong emotional responsibility, yet have limited visibility into their daily lives and limited capacity for caregiving. In this work, we explore how LLMs can be used to generate retrospective summaries from multi-modal tracking data for RFMs of older adults. We leveraged and customized an existing system, Vital Insight, to generate initial summaries on different dates and data availability scenarios as technology probes, and conducted interviews with 11 RFMs to gather feedback. Based on these insights, we redesigned the system into a multi-layer, multi-agent, insight-driven summary approach that builds from objective statistics and descriptions to enriched, context-aware narratives. We then compared the redesigned summaries with the initial versions through a survey with the same 11 RFMs and found significant improvements in satisfaction, perceived helpfulness, trust, and willingness to receive the summaries. We conclude by presenting design implications for AI-generated summaries for RFMs and broader contexts, emphasizing the need to support RFMs' sensemaking shift from simply presenting ''What'' data were collected, to explaining ''How'' is my loved one doing and ''Why''. △ Less

Submitted 2 June, 2026; originally announced June 2026.

AnchorMoE: Interpretable Time Series Classification via Anchor-Routed MoE

Authors: Tao Xie, Zexi Tan, Haoyi Xiao, Mengke Li, Yiqun Zhang, Yang Lu, Cuie Yang, Yiu-ming Cheung

Abstract: Multivariate time series classification (MTSC) is pivotal in high-stakes domains, such as clinical diagnosis and industrial fault detection, where safe deployment necessitates transparent decision-making. However, isolating the temporal segments that drive model predictions is challenging because discriminative signals in real-world time series are typically sparse, heterogeneous, and heavily obsc… ▽ More Multivariate time series classification (MTSC) is pivotal in high-stakes domains, such as clinical diagnosis and industrial fault detection, where safe deployment necessitates transparent decision-making. However, isolating the temporal segments that drive model predictions is challenging because discriminative signals in real-world time series are typically sparse, heterogeneous, and heavily obscured by background noise. This paper, therefore, proposes AnchorMoE, an interpretable-by-construction classification framework. Built upon a Mixture-of-Experts (MoE) architecture, AnchorMoE encodes multi-view representations of local patches and routes them to specialized experts, ensuring that the final prediction is formulated as an exact additive decomposition over the input segments, facilitating ante-hoc transparency rather than relying on post-hoc estimations. To maintain the reliability of this decomposition under sparse signal distributions, we introduce a geometric orthogonality constraint that penalizes representational redundancy, compelling distinct experts to specialize in heterogeneous predictive patterns. Furthermore, an uncertainty-aware reliability gate is designed to dynamically calibrate the contribution of each segment, effectively suppressing residual background noise. Extensive experiments on real-world and synthetic benchmarks demonstrate that AnchorMoE achieves highly competitive classification performance while faithfully grounding its decisions in the raw time series. △ Less

Submitted 10 July, 2026; v1 submitted 2 June, 2026; originally announced June 2026.

Comments: Accepted by KDD 2026

FinCom: A Financial Multi-Agent Demo with Disagree-or-Commit Deliberation

Authors: Chao Peter Yang, Zixiao Tan, Kaisen Yao, Ziyu Zhou, Eleanor Jiang, Michael Wu

Abstract: Multi-agent systems powered by large language models (LLMs) are increasingly used for financial analysis and decision support. However, existing coordination schemes, especially those emphasizing consensus or debate, are vulnerable to sycophancy: agents conform to peer reasoning instead of evidence, leading to premature agreement and degraded outcomes. We introduce FinCom (Financial Committee), a… ▽ More Multi-agent systems powered by large language models (LLMs) are increasingly used for financial analysis and decision support. However, existing coordination schemes, especially those emphasizing consensus or debate, are vulnerable to sycophancy: agents conform to peer reasoning instead of evidence, leading to premature agreement and degraded outcomes. We introduce FinCom (Financial Committee), a governed multi-agent framework and interactive system that operationalizes the Disagree-or-Commit (DoC) protocol to embed structured dissent into financial AI committees. A central Supervisor orchestrates three ReAct-enabled specialist agents: Research, Quantitative, and Risk. Each agent is equipped with role-specific tools for retrieval, computation, and stress testing. During deliberation, agents must either explicitly critique or commit to their peers' reasoning before converging on a unified recommendation. This demonstration showcases how FinCom supports committee-style financial analysis through coordinated multi-agent interaction, including structured report generation and interactive decision support. Evaluated across the most recent financial agent benchmark, in addition to 90 internal handcrafted financial tasks using an LLM-as-a-Judge protocol, DoC improves reasoning accuracy and risk awareness significantly over a consensus-seeking baseline on both an in-house and external evaluation set. By reframing disagreement as a governance primitive rather than noise, FinCom offers a lightweight, prompt-only recipe for improving accountability, transparency, and epistemic robustness in agentic financial systems. △ Less

Submitted 30 May, 2026; originally announced June 2026.

ESPO: Early-Stopping Proximal Policy Optimization

Authors: Zihang Li, Rui Zhou, Yingcheng Shi, Wenhan Yu, Zhewen Tan, Zixiang Liu, Zeming Li, Binhua Li, Yongbin Li, Tong Yang, Jieping Ye

Abstract: When a large language model under reinforcement learning commits a wrong reasoning step early in a trajectory, standard algorithms force it to keep generating until the maximum horizon, spending compute on tokens that never receive positive reward and polluting advantage estimates with post-failure noise. We propose ESPO (Early-Stopping Proximal Policy Optimization), which detects trajectory failu… ▽ More When a large language model under reinforcement learning commits a wrong reasoning step early in a trajectory, standard algorithms force it to keep generating until the maximum horizon, spending compute on tokens that never receive positive reward and polluting advantage estimates with post-failure noise. We propose ESPO (Early-Stopping Proximal Policy Optimization), which detects trajectory failure on-the-fly and terminates rollouts early. At each generation step, ESPO computes a surrogate regret using only the logits already computed during sampling, and terminates when the smoothed cumulative regret significantly exceeds its estimated values. Truncated trajectories are treated as absorbing failure states with a terminal reward, concentrating negative temporal-difference (TD) errors near the detected failure step without any additional reward model or human annotation. On DeepSeek-R1-Distill-Qwen-7B trained for mathematical reasoning, ESPO surpasses PPO on AIME~2024 (46.28% vs. 45.25%), AMC~2023 (85.83% vs. 82.94%), and MATH-500 (87.42% vs. 85.43%), while saving more than 20% rollout tokens cumulatively. △ Less

Submitted 28 May, 2026; originally announced May 2026.

Geometry-Guided Modeling of Foundation Features Enables Generalizable Object Shape Deformation Learning

Authors: Yiyao Ma, Kai Chen, Zhongxiang Zhou, Zhuheng Song, Dongsheng Xie, Zelong Tan, Rong Xiong, Qi Dou

Abstract: Monocular 3D shape recovery is fundamental to geometric understanding, yet achieving robust generalization across arbitrary viewpoints and unseen object categories remains a significant challenge. In this paper, we present a generalizable deformation learning framework that reconstructs 3D objects by explicitly deforming a category-level shape template to match the target observation. To address c… ▽ More Monocular 3D shape recovery is fundamental to geometric understanding, yet achieving robust generalization across arbitrary viewpoints and unseen object categories remains a significant challenge. In this paper, we present a generalizable deformation learning framework that reconstructs 3D objects by explicitly deforming a category-level shape template to match the target observation. To address complex shape variations between the template and the target, we introduce a geometry-guided feature modeling mechanism. This process first enriches foundation features with template topology to yield a geometry-aware representation, which is then explicitly correlated with the target observation to guide precise deformation. Furthermore, to bridge the disparity between the fixed template and arbitrary target views, we propose a view-adaptive feature aggregation module. This module leverages multi-view template features and their corresponding camera poses to enrich the canonical template representation, ensuring robust feature alignment regardless of the target's perspective. Extensive experiments demonstrate that our approach significantly outperforms state-of-the-art methods in handling large shape variations and diverse viewpoints, exhibiting strong generalization to novel categories and effectively supporting downstream real-world dexterous robotic manipulation tasks. Project homepage: https://GODeform.github.io/ △ Less

Submitted 1 June, 2026; v1 submitted 28 May, 2026; originally announced May 2026.

Comments: 20 pages, 12 figures, accepted by ICML 2026

Measuring Real-World Prompt Injection Attacks in LLM-based Resume Screening

Authors: Mohan Zhang, Yuqi Jia, Zhen Tan, Steven Jiang, Neil Zhenqiang Gong, Tianlong Chen, Dawn Song

Abstract: LLMs are vulnerable to prompt injection attacks. However, this vulnerability has been primarily demonstrated conceptually in academic studies or through a few anecdotal case studies. Its prevalence and impact in real-world LLM-based applications are largely unexplored. In this work, we present the first systematic study of prompt-injection attacks in a widely used application: LLM-based resume scr… ▽ More LLMs are vulnerable to prompt injection attacks. However, this vulnerability has been primarily demonstrated conceptually in academic studies or through a few anecdotal case studies. Its prevalence and impact in real-world LLM-based applications are largely unexplored. In this work, we present the first systematic study of prompt-injection attacks in a widely used application: LLM-based resume screening. Our analysis is based on approximately 200K real-world resumes collected over multiple years by hireEZ. We first design tailored methods to detect prompt injection in resumes. Manual validation on a small-scale dataset demonstrates that our detectors achieve high precision and outperform state-of-the-art general-purpose detectors. We then apply our detector to the full resume dataset and conduct a comprehensive measurement study of real-world prompt injection attacks. Our analysis reveals several intriguing findings: approximately 1% of resumes contain hidden prompt injections; the prevalence of such injected resumes has increased noticeably over the past one to two years; and more than 90% of injected prompts do not use explicit instructions. These results provide the first evidence of large-scale prompt injection in real-world LLM-based applications and lay the groundwork for future studies to understand and mitigate such attacks. △ Less

Submitted 27 May, 2026; originally announced May 2026.

Comments: Published in USENIX Security Symposium 2026; Code and artifacts are available at https://github.com/UNITES-Lab/resume-injection-measurement

A Digital Twin Framework for Virtual Visuo-Haptic Teleoperation of Complex-Shaped Optical Microrobots

Authors: Zongcai Tan, Lan Wei, Dandan Zhang

Abstract: Optical tweezers (OT) provide piconewton-scale manipulation for delicate biomedical tasks, where visuo-haptic feedback can improve operator awareness by conveying interaction-force cues and trap-stability information. However, visuo-haptic teleoperation frameworks for complex-shaped optical microrobots remain underdeveloped, particularly in multi-trap manipulation scenarios. This paper presents a… ▽ More Optical tweezers (OT) provide piconewton-scale manipulation for delicate biomedical tasks, where visuo-haptic feedback can improve operator awareness by conveying interaction-force cues and trap-stability information. However, visuo-haptic teleoperation frameworks for complex-shaped optical microrobots remain underdeveloped, particularly in multi-trap manipulation scenarios. This paper presents a digital twin framework for virtual visuo-haptic teleoperation of complex-shaped OT-driven microrobots. The framework integrates a digital twin environment, image-based pose and depth estimation, microrobot motion simulation, and model-based haptic rendering within a Robot Operating System (ROS)-connected bimanual teleoperation system. For force modeling, we combine a Multi-Sphere Distributed Manipulation (MSDM) model with optical-force estimation from the Optical Tweezers Toolbox, enabling simulator-driven visuo-haptic feedback. The framework reproduces representative microrobot motion trends and provides haptic force rendering that is numerically consistent with the fitted optical-force model. In simulated cell-delivery tasks, haptic feedback reduced the standard deviations of the contact-force metric and the microrobot-to-trap-center distance metric by 53.2% and 55.2%, respectively, and improved task success from 30% to 80%. These results demonstrate the framework's effectiveness for evaluating visuo-haptic teleoperation strategies for complex-shaped optical microrobots. △ Less

Submitted 27 May, 2026; originally announced May 2026.

Comments: Accepted by 2026 MARSS