Showing 1–50 of 481 results for author: Hao, Y

Programmable Packet Scheduling with Dynamic Reordering at Line Rate

Authors: Zekun Wang, Binghao Yue, Yichen Deng, Weitao Pan, Jiangyi Shi, Yue Hao

Abstract: High-speed switch packet scheduling demands both line-rate performance and programmability. Existing programmable hardware scheduling models, such as PIFO and PIEO, can express a broad range of scheduling algorithms; however, their semantics are restricted to packet-level ordering and cannot dynamically reorder buffered packets, which limits the support for dynamic-ordering algorithms such as pFab… ▽ More High-speed switch packet scheduling demands both line-rate performance and programmability. Existing programmable hardware scheduling models, such as PIFO and PIEO, can express a broad range of scheduling algorithms; however, their semantics are restricted to packet-level ordering and cannot dynamically reorder buffered packets, which limits the support for dynamic-ordering algorithms such as pFabric. To overcome this limitation, we propose UIFO (Update-In-First-Out), a new programmable scheduling model that introduces a two-level abstraction over classes and packets. UIFO enables dynamic updates to the scheduling order at the class level while preserving in-order packet scheduling within each class, thereby supporting dynamic reordering of already-buffered packets. Furthermore, UIFO remains fully compatible with and generalizes existing PIFO and PIEO models. We implement a hardware prototype of UIFO based on priority-queue designs and evaluate it on an FPGA platform and in a 28 nm ASIC process. Overall, UIFO significantly enhances scheduling expressiveness and maintains favorable scalability while sustaining 100 Gbps line-rate throughput. △ Less

Submitted 13 April, 2026; originally announced April 2026.

Comments: 14 pages,12 body

ClawBench: Can AI Agents Complete Everyday Online Tasks?

Authors: Yuxuan Zhang, Yubo Wang, Yipeng Zhu, Penghui Du, Junwen Miao, Xuan Lu, Wendong Xu, Yunzhuo Hao, Songcheng Cai, Xiaochen Wang, Huaisong Zhang, Xian Wu, Yi Lu, Minyi Lei, Kai Zou, Huifeng Yin, Ping Nie, Liang Chen, Dongfu Jiang, Wenhu Chen, Kelsey R. Allen

Abstract: AI agents may be able to automate your inbox, but can they automate other routine aspects of your life? Everyday online tasks offer a realistic yet unsolved testbed for evaluating the next generation of AI agents. To this end, we introduce ClawBench, an evaluation framework of 153 simple tasks that people need to accomplish regularly in their lives and work, spanning 144 live platforms across 15 c… ▽ More AI agents may be able to automate your inbox, but can they automate other routine aspects of your life? Everyday online tasks offer a realistic yet unsolved testbed for evaluating the next generation of AI agents. To this end, we introduce ClawBench, an evaluation framework of 153 simple tasks that people need to accomplish regularly in their lives and work, spanning 144 live platforms across 15 categories, from completing purchases and booking appointments to submitting job applications. These tasks require demanding capabilities beyond existing benchmarks, such as obtaining relevant information from user-provided documents, navigating multi-step workflows across diverse platforms, and write-heavy operations like filling in many detailed forms correctly. Unlike existing benchmarks that evaluate agents in offline sandboxes with static pages, ClawBench operates on production websites, preserving the full complexity, dynamic nature, and challenges of real-world web interaction. A lightweight interception layer captures and blocks only the final submission request, ensuring safe evaluation without real-world side effects. Our evaluations of 7 frontier models show that both proprietary and open-source models can complete only a small portion of these tasks. For example, Claude Sonnet 4.6 achieves only 33.3%. Progress on ClawBench brings us closer to AI agents that can function as reliable general-purpose assistants. △ Less

Submitted 9 April, 2026; originally announced April 2026.

Comments: Project page: https://claw-bench.com

PRISM-MCTS: Learning from Reasoning Trajectories with Metacognitive Reflection

Authors: Siyuan Cheng, Bozhong Tian, YanChao Hao, Zheng Wei

Abstract: PRISM-MCTS: Learning from Reasoning Trajectories with Metacognitive Reflection Siyuan Cheng, Bozhong Tian, Yanchao Hao, Zheng Wei Published: 06 Apr 2026, Last Modified: 06 Apr 2026 ACL 2026 Findings Conference, Area Chairs, Reviewers, Publication Chairs, Authors Revisions BibTeX CC BY 4.0 Keywords: Efficient/Low-Resource Methods for NLP, Generation, Question Answering Abstract: The emergence of re… ▽ More PRISM-MCTS: Learning from Reasoning Trajectories with Metacognitive Reflection Siyuan Cheng, Bozhong Tian, Yanchao Hao, Zheng Wei Published: 06 Apr 2026, Last Modified: 06 Apr 2026 ACL 2026 Findings Conference, Area Chairs, Reviewers, Publication Chairs, Authors Revisions BibTeX CC BY 4.0 Keywords: Efficient/Low-Resource Methods for NLP, Generation, Question Answering Abstract: The emergence of reasoning models, exemplified by OpenAI o1, signifies a transition from intuitive to deliberative cognition, effectively reorienting the scaling laws from pre-training paradigms toward test-time computation. While Monte Carlo Tree Search (MCTS) has shown promise in this domain, existing approaches typically treat each rollout as an isolated trajectory. This lack of information sharing leads to severe inefficiency and substantial computational redundancy, as the search process fails to leverage insights from prior explorations. To address these limitations, we propose PRISM-MCTS, a novel reasoning framework that draws inspiration from human parallel thinking and reflective processes. PRISM-MCTS integrates a Process Reward Model (PRM) with a dynamic shared memory, capturing both "Heuristics" and "Fallacies". By reinforcing successful strategies and pruning error-prone branches, PRISM-MCTS effectively achieves refinement. Furthermore, we develop a data-efficient training strategy for the PRM, achieving high-fidelity evaluation under a few-shot regime. Empirical evaluations across diverse reasoning benchmarks substantiate the efficacy of PRISM-MCTS. Notably, it halves the trajectory requirements on GPQA while surpassing MCTS-RAG and Search-o1, demonstrating that it scales inference by reasoning judiciously rather than exhaustively. △ Less

Submitted 7 April, 2026; originally announced April 2026.

Cactus: Accelerating Auto-Regressive Decoding with Constrained Acceptance Speculative Sampling

Authors: Yongchang Hao, Lili Mou

Abstract: Speculative sampling (SpS) has been successful in accelerating the decoding throughput of auto-regressive large language models by leveraging smaller draft models. SpS strictly enforces the generated distribution to match that of the verifier LLM. This is unnecessarily restrictive as slight variations of the verifier's distribution, such as sampling with top-$k$ or temperature, would also be accep… ▽ More Speculative sampling (SpS) has been successful in accelerating the decoding throughput of auto-regressive large language models by leveraging smaller draft models. SpS strictly enforces the generated distribution to match that of the verifier LLM. This is unnecessarily restrictive as slight variations of the verifier's distribution, such as sampling with top-$k$ or temperature, would also be acceptable. Typical acceptance sampling (TAS) alleviates this issue by accepting more tokens using entropy-based heuristics. However, this approach distorts the verifier distribution, potentially degrading output quality when the verifier encodes critical information. In this work, we formalize the speculative sampling algorithm through the lens of constrained optimization. Based on this formulation, we propose Cactus (constrained acceptance speculative sampling), a method that guarantees controlled divergence from the verifier distribution and increasing acceptance rates. Empirical results across a wide range of benchmarks confirm the effectiveness of our approach. △ Less

Submitted 4 April, 2026; originally announced April 2026.

Comments: Camera-ready version. Accepted at ICLR 2026

Omni123: Exploring 3D Native Foundation Models with Limited 3D Data by Unifying Text to 2D and 3D Generation

Authors: Chongjie Ye, Cheng Cao, Chuanyu Pan, Yiming Hao, Yihao Zhi, Yuanming Hu, Xiaoguang Han

Abstract: Recent multimodal large language models have achieved strong performance in unified text and image understanding and generation, yet extending such native capability to 3D remains challenging due to limited data. Compared to abundant 2D imagery, high-quality 3D assets are scarce, making 3D synthesis under-constrained. Existing methods often rely on indirect pipelines that edit in 2D and lift resul… ▽ More Recent multimodal large language models have achieved strong performance in unified text and image understanding and generation, yet extending such native capability to 3D remains challenging due to limited data. Compared to abundant 2D imagery, high-quality 3D assets are scarce, making 3D synthesis under-constrained. Existing methods often rely on indirect pipelines that edit in 2D and lift results into 3D via optimization, sacrificing geometric consistency. We present Omni123, a 3D-native foundation model that unifies text-to-2D and text-to-3D generation within a single autoregressive framework. Our key insight is that cross-modal consistency between images and 3D can serve as an implicit structural constraint. By representing text, images, and 3D as discrete tokens in a shared sequence space, the model leverages abundant 2D data as a geometric prior to improve 3D representations. We introduce an interleaved X-to-X training paradigm that coordinates diverse cross-modal tasks over heterogeneous paired datasets without requiring fully aligned text-image-3D triplets. By traversing semantic-visual-geometric cycles (e.g., text to image to 3D to image) within autoregressive sequences, the model jointly enforces semantic alignment, appearance fidelity, and multi-view geometric consistency. Experiments show that Omni123 significantly improves text-guided 3D generation and editing, demonstrating a scalable path toward multimodal 3D world models. △ Less

Submitted 2 April, 2026; originally announced April 2026.

Yet Even Less Is Even Better For Agentic, Reasoning, and Coding LLMs

Authors: CodeArts Model Team, Yang Ye, Jingyuan Tan, Tianyue Jiang, Ruizhe Ye, Qiankun He, Jiarui Yang, Jian Dong, Sicong Liang, Chongjian Yue, Peibai Xu, Lufan Lu, Shiguan Pang, Taotao Qian, Junbao Hu, Yuechan Hao, Ensheng Shi, Qi Zhang, Yi Hao, Na Fan, Xin Tan, Shuai Yao, Zhiwei Shen, Zongchen Li, Yanlin Wang , et al. (2 additional authors not shown)

Abstract: Training effective software engineering agents requires large volumes of task-specific trajectories, incurring substantial data construction costs. Inspired by the "Less-Is-More" hypothesis in mathematical reasoning, we investigate its extension to agentic scenarios and propose an end-to-end training framework that achieves superior agentic capabilities with fewer but higher-quality training traje… ▽ More Training effective software engineering agents requires large volumes of task-specific trajectories, incurring substantial data construction costs. Inspired by the "Less-Is-More" hypothesis in mathematical reasoning, we investigate its extension to agentic scenarios and propose an end-to-end training framework that achieves superior agentic capabilities with fewer but higher-quality training trajectories. This is achieved via STITCH (Sliding-memory Trajectory Inference and Task Chunking Heuristic), a coarse-to-fine mechanism that filters low-value noise and retains decision-critical tokens to maximize training signal quality. We conduct experiments across multiple agent frameworks (e.g., mini-SWE-agent, MSWE-agent), model scales (30B to 355B), and multilingual settings (Python, Java, and ArkTS). On SWE-bench Verified, models trained with STITCH achieve up to 63.16% relative improvement over base models. On Multi-SWE-bench (Java), MiniMax-M2.5-STITCH achieves 43.75% with our CodeArts Agent scaffold (+16.67%). On HarmonyOS (ArkTS), GLM-4.7-STITCH improves the compilation pass rate to 61.31% (+43.34%) with less than 1K training trajectories. Our results confirm that the "Less-Is-More" paradigm generalizes effectively to complex agentic tasks across diverse languages and model scales. △ Less

Submitted 6 April, 2026; v1 submitted 1 April, 2026; originally announced April 2026.

Comments: 17 pages, 5 figures

Disentangled Graph Prompting for Out-Of-Distribution Detection

Authors: Cheng Yang, Yu Hao, Qi Zhang, Chuan Shi

Abstract: When testing data and training data come from different distributions, deep neural networks (DNNs) will face significant safety risks in practical applications. Therefore, out-of-distribution (OOD) detection techniques, which can identify OOD samples at test time and alert the system, are urgently needed. Existing graph OOD detection methods usually characterize fine-grained in-distribution (ID) p… ▽ More When testing data and training data come from different distributions, deep neural networks (DNNs) will face significant safety risks in practical applications. Therefore, out-of-distribution (OOD) detection techniques, which can identify OOD samples at test time and alert the system, are urgently needed. Existing graph OOD detection methods usually characterize fine-grained in-distribution (ID) patterns from multiple perspectives, and train end-to-end graph neural networks (GNNs) for prediction. However, due to the unavailability of OOD data during training, the absence of explicit supervision signals could lead to sub-optimal performance of end-to-end encoders. To address this issue, we follow the pre-training+prompting paradigm to utilize pre-trained GNN encoders, and propose Disentangled Graph Prompting (DGP), to capture fine-grained ID patterns with the help of ID graph labels. Specifically, we design two prompt generators that respectively generate class-specific and class-agnostic prompt graphs by modifying the edge weights of an input graph. We also design several effective losses to train the prompt generators and prevent trivial solutions. We conduct extensive experiments on ten datasets to demonstrate the superiority of our proposed DGP, which achieves a relative AUC improvement of 3.63% over the best graph OOD detection baseline. Ablation studies and hyper-parameter experiments further show the effectiveness of DGP. Code is available at https://github.com/BUPT-GAMMA/DGP. △ Less

Submitted 31 March, 2026; originally announced March 2026.

Comments: Accepted for publication in IEEE Transactions on Knowledge and Data Engineering (TKDE)

CReF: Cross-modal and Recurrent Fusion for Depth-conditioned Humanoid Locomotion

Authors: Yuan Hao, Ruiqi Yu, Shixin Luo, Guoteng Zhang, Jun Wu, Qiuguo Zhu

Abstract: Stable traversal over geometrically complex terrain increasingly requires exteroceptive perception, yet prior perceptive humanoid locomotion methods often remain tied to explicit geometric abstractions, either by mediating control through robot-centric 2.5D terrain representations or by shaping depth learning with auxiliary geometry-related targets. Such designs inherit the representational bias o… ▽ More Stable traversal over geometrically complex terrain increasingly requires exteroceptive perception, yet prior perceptive humanoid locomotion methods often remain tied to explicit geometric abstractions, either by mediating control through robot-centric 2.5D terrain representations or by shaping depth learning with auxiliary geometry-related targets. Such designs inherit the representational bias of the intermediate or supervisory target and can be restrictive for vertical structures, perforated obstacles, and complex real-world clutter. We propose CReF (Cross-modal and Recurrent Fusion), a single-stage depth-conditioned humanoid locomotion framework that learns locomotion-relevant features directly from raw forward-facing depth without explicit geometric intermediates. CReF couples proprioception and depth tokens through proprioception-queried cross-modal attention, fuses the resulting representation with a gated residual fusion block, and performs temporal integration with a Gated Recurrent Unit (GRU) regulated by a highway-style output gate for state-dependent blending of recurrent and feedforward features. To further improve terrain interaction, we introduce a terrain-aware foothold placement reward that extracts supportable foothold candidates from foot-end point-cloud samples and rewards touchdown locations that lie close to the nearest supportable candidate. Experiments in simulation and on a physical humanoid demonstrate robust traversal over diverse terrains and effective zero-shot transfer to real-world scenes containing handrails, hollow pallet assemblies, severe reflective interference, and visually cluttered outdoor surroundings. △ Less

Submitted 31 March, 2026; v1 submitted 31 March, 2026; originally announced March 2026.

ResAdapt: Adaptive Resolution for Efficient Multimodal Reasoning

Authors: Huanxuan Liao, Zhongtao Jiang, Yupu Hao, Yuqiao Tan, Shizhu He, Ben Wang, Jun Zhao, Kun Xu, Kang Liu

Abstract: Multimodal Large Language Models (MLLMs) achieve stronger visual understanding by scaling input fidelity, yet the resulting visual token growth makes jointly sustaining high spatial resolution and long temporal context prohibitive. We argue that the bottleneck lies not in how post-encoding representations are compressed but in the volume of pixels the encoder receives, and address it with ResAdapt… ▽ More Multimodal Large Language Models (MLLMs) achieve stronger visual understanding by scaling input fidelity, yet the resulting visual token growth makes jointly sustaining high spatial resolution and long temporal context prohibitive. We argue that the bottleneck lies not in how post-encoding representations are compressed but in the volume of pixels the encoder receives, and address it with ResAdapt, an Input-side adaptation framework that learns how much visual budget each frame should receive before encoding. ResAdapt couples a lightweight Allocator with an unchanged MLLM backbone, so the backbone retains its native visual-token interface while receiving an operator-transformed input. We formulate allocation as a contextual bandit and train the Allocator with Cost-Aware Policy Optimization (CAPO), which converts sparse rollout feedback into a stable accuracy-cost learning signal. Across budget-controlled video QA, temporal grounding, and image reasoning tasks, ResAdapt improves low-budget operating points and often lies on or near the efficiency-accuracy frontier, with the clearest gains on reasoning-intensive benchmarks under aggressive compression. Notably, ResAdapt supports up to 16x more frames at the same visual budget while delivering over 15% performance gain. Code is available at https://github.com/Xnhyacinth/ResAdapt. △ Less

Submitted 31 March, 2026; v1 submitted 30 March, 2026; originally announced March 2026.

Comments: work in progress

The First Issue Matters: Linking Task-Level Characteristics to Long-Term Newcomer Retention in OSS

Authors: Yichen Hao, Weiwei Xu, Kai Gao, Xiaofang Zhang

Abstract: Sustaining newcomer participation is critical for the long-term health of open-source communities. Although prior research has explored various task recommendation approaches to help newcomers resolve their first-issue, these methods overlook how characteristics of first-issues may influence newcomers' long-term retention, limiting our understanding of whether initial success leads to sustained pa… ▽ More Sustaining newcomer participation is critical for the long-term health of open-source communities. Although prior research has explored various task recommendation approaches to help newcomers resolve their first-issue, these methods overlook how characteristics of first-issues may influence newcomers' long-term retention, limiting our understanding of whether initial success leads to sustained participation and hindering effective onboarding design. In this paper, we conduct a large-scale empirical study to examine how first-issue characteristics affect newcomer retention. We combine predictive analysis, interpretability techniques, and causal inference to estimate the causal effects of issue characteristics on retention outcomes. The prediction task supports the interpretation and shows that interaction-related characteristics exhibit stronger associations with retention than intrinsic issue attributes. The causal analysis further reveals that issues reported by moderately experienced contributors, accompanied by moderate discussion intensity and participation from project members, and neutral or slightly negative comment sentiment, have higher retention potential. These findings provide actionable insights for OSS maintainers on designing issue management practices that better support long-term newcomer retention. △ Less

Submitted 28 March, 2026; originally announced March 2026.

Label What Matters: Modality-Balanced and Difficulty-Aware Multimodal Active Learning

Authors: Yuqiao Zeng, Xu Wang, Tengfei Liang, Yiqing Hao, Yi Jin, Hui Yu

Abstract: Multimodal learning integrates complementary information from different modalities such as image, text, and audio to improve model performance, but its success relies on large-scale labeled data, which is costly to obtain. Active learning (AL) mitigates this challenge by selectively annotating informative samples. In multimodal settings, many approaches implicitly assume that modality importance i… ▽ More Multimodal learning integrates complementary information from different modalities such as image, text, and audio to improve model performance, but its success relies on large-scale labeled data, which is costly to obtain. Active learning (AL) mitigates this challenge by selectively annotating informative samples. In multimodal settings, many approaches implicitly assume that modality importance is stable across rounds and keep selection rules fixed at the fusion stage, which leaves them insensitive to the dynamic nature of multimodal learning, where the relative value of modalities and the difficulty of instances shift as training proceeds. To address this issue, we propose RL-MBA, a reinforcement-learning framework for modality-balanced, difficulty-aware multimodal active learning. RL-MBA models sample selection as a Markov Decision Process, where the policy adapts to modality contributions, uncertainty, and diversity, and the reward encourages accuracy gains and balance. Two key components drive this adaptability: (1) Adaptive Modality Contribution Balancing (AMCB), which dynamically adjusts modality weights via reinforcement feedback, and (2) Evidential Fusion for DifficultyAware Policy Adjustment (EFDA), which estimates sample difficulty via uncertainty-based evidential fusion to prioritize informative samples. Experiments on Food101, KineticsSound, and VGGSound demonstrate that RL-MBA consistently outperforms strong baselines, improving both classification accuracy and modality fairness under limited labeling budgets. △ Less

Submitted 26 March, 2026; originally announced March 2026.

Backward Arcs in Hamilton Oriented Cycles and Paths in Directed Graphs with Independence Number Two

Authors: S. Gerke, Q. Guo, G. Gutin, Y. Hao, W. Veeranonchai, A. Yeo

Abstract: In a digraph $D=(V,A)$, an oriented path is a sequence $P=x_1x_2\dots x_p$ of distinct vertices such that either $x_ix_{i+1}\in A$ or $x_{i+1}x_{i}\in A$ or both for every $i\in [p-1]$. If $x_ix_{i+1}\in A$ in $P$, then $x_ix_{i+1}$ is a forward arc of $P$; otherwise, $x_{i+1}x_{i}$ is a backward arc. The independence number $α(D)$ is the maximum integer $p$ such that $D$ has a set of $p$ vertices… ▽ More In a digraph $D=(V,A)$, an oriented path is a sequence $P=x_1x_2\dots x_p$ of distinct vertices such that either $x_ix_{i+1}\in A$ or $x_{i+1}x_{i}\in A$ or both for every $i\in [p-1]$. If $x_ix_{i+1}\in A$ in $P$, then $x_ix_{i+1}$ is a forward arc of $P$; otherwise, $x_{i+1}x_{i}$ is a backward arc. The independence number $α(D)$ is the maximum integer $p$ such that $D$ has a set of $p$ vertices where there is no arc between any pair of vertices. A digraph is $k$-connected if its underlying undirected graph is $k$-connected. Freschi and Lo (JCT-B 2024) proved that every $n$-vertex oriented graph with minimum degree $δ\ge n/2$ has a Hamilton oriented cycle with at most $n-δ$ backward arcs. We prove that every 2-connected digraph $D$ with $α(D)\le 2$ has a Hamilton oriented cycle with at most five backward arcs, and every 1-connected digraph $D$ with $α(D)\le 2$ has a Hamilton oriented path with at most two backward arcs. △ Less

Submitted 24 March, 2026; originally announced March 2026.

The Interspeech 2026 Audio Encoder Capability Challenge for Large Audio Language Models

Authors: Heinrich Dinkel, Jiahao Zhou, Guanbo Wang, Yadong Niu, Junbo Zhang, Yufeng Hao, Ying Liu, Ke Li, Wenwu Wang, Zhiyong Wu, Jian Luan

Abstract: This paper presents the Interspeech 2026 Audio Encoder Capability Challenge, a benchmark specifically designed to evaluate and advance the performance of pre-trained audio encoders as front-end modules for Large Audio Language Models (LALMs). While LALMs have shown remarkable understanding of complex acoustic scenes, their performance depends on the semantic richness of the underlying audio encode… ▽ More This paper presents the Interspeech 2026 Audio Encoder Capability Challenge, a benchmark specifically designed to evaluate and advance the performance of pre-trained audio encoders as front-end modules for Large Audio Language Models (LALMs). While LALMs have shown remarkable understanding of complex acoustic scenes, their performance depends on the semantic richness of the underlying audio encoder representations. This challenge addresses the integration gap by providing a unified generative evaluation framework, XARES-LLM, which assesses submitted encoders across a diverse suite of downstream classification and generation tasks. By decoupling encoder development from LLM fine-tuning, the challenge establishes a standardized protocol for general-purpose audio representations that can effectively be used for the next generation of multimodal language models. △ Less

Submitted 23 March, 2026; originally announced March 2026.

Comments: Interspeech 2026 Challenge

PIVM: Diffusion-Based Prior-Integrated Variation Modeling for Anatomically Precise Abdominal CT Synthesis

Authors: Dinglun He, Baoming Zhang, Xu Wang, Yao Hao, Deshan Yang, Ye Duan

Abstract: Abdominal CT data are limited by high annotation costs and privacy constraints, which hinder the development of robust segmentation and diagnostic models. We present a Prior-Integrated Variation Modeling (PIVM) framework, a diffusion-based method for anatomically accurate CT image synthesis. Instead of generating full images from noise, PIVM predicts voxel-wise intensity variations relative to org… ▽ More Abdominal CT data are limited by high annotation costs and privacy constraints, which hinder the development of robust segmentation and diagnostic models. We present a Prior-Integrated Variation Modeling (PIVM) framework, a diffusion-based method for anatomically accurate CT image synthesis. Instead of generating full images from noise, PIVM predicts voxel-wise intensity variations relative to organ-specific intensity priors derived from segmentation labels. These priors and labels jointly guide the diffusion process, ensuring spatial alignment and realistic organ boundaries. Unlike latent-space diffusion models, our approach operates directly in image space while preserving the full Hounsfield Unit (HU) range, capturing fine anatomical textures without smoothing. Source code is available at https://github.com/BZNR3/PIVM. △ Less

Submitted 23 March, 2026; originally announced March 2026.

Comments: Accepted at the IEEE International Symposium on Biomedical Imaging (ISBI) 2026 (Oral). Equal contribution by the first three authors

User Preference Modeling for Conversational LLM Agents: Weak Rewards from Retrieval-Augmented Interaction

Authors: Yuren Hao, Shuhaib Mehri, ChengXiang Zhai, Dilek Hakkani-Tür

Abstract: Large language models are increasingly used as personal assistants, yet most lack a persistent user model, forcing users to repeatedly restate preferences across sessions. We propose Vector-Adapted Retrieval Scoring (VARS), a pipeline-agnostic, frozen-backbone framework that represents each user with long-term and short-term vectors in a shared preference space and uses these vectors to bias retri… ▽ More Large language models are increasingly used as personal assistants, yet most lack a persistent user model, forcing users to repeatedly restate preferences across sessions. We propose Vector-Adapted Retrieval Scoring (VARS), a pipeline-agnostic, frozen-backbone framework that represents each user with long-term and short-term vectors in a shared preference space and uses these vectors to bias retrieval scoring over structured preference memory. The vectors are updated online from weak scalar rewards from users' feedback, enabling personalization without per-user fine-tuning. We evaluate on \textsc{MultiSessionCollab}, an online multi-session collaboration benchmark with rich user preference profiles, across math and code tasks. Under frozen backbones, the main benefit of user-aware retrieval is improved interaction efficiency rather than large gains in raw task accuracy: our full VARS agent achieves the strongest overall performance, matches a strong Reflection baseline in task success, and reduces timeout rate and user effort. The learned long-term vectors also align with cross-user preference overlap, while short-term vectors capture session-specific adaptation, supporting the interpretability of the dual-vector design. Code, model, and data are available at https://github.com/YurenHao0426/VARS. △ Less

Submitted 21 March, 2026; originally announced March 2026.

Comments: 21 pages including appendices

ACM Class: I.2.7

Reasoning Traces Shape Outputs but Models Won't Say So

Authors: Yijie Hao, Lingjie Chen, Ali Emami, Joyce Ho

Abstract: Can we trust the reasoning traces that large reasoning models (LRMs) produce? We investigate whether these traces faithfully reflect what drives model outputs, and whether models will honestly report their influence. We introduce Thought Injection, a method that injects synthetic reasoning snippets into a model's <think> trace, then measures whether the model follows the injected reasoning and ack… ▽ More Can we trust the reasoning traces that large reasoning models (LRMs) produce? We investigate whether these traces faithfully reflect what drives model outputs, and whether models will honestly report their influence. We introduce Thought Injection, a method that injects synthetic reasoning snippets into a model's <think> trace, then measures whether the model follows the injected reasoning and acknowledges doing so. Across 45,000 samples from three LRMs, we find that injected hints reliably alter outputs, confirming that reasoning traces causally shape model behavior. However, when asked to explain their changed answers, models overwhelmingly refuse to disclose the influence: overall non-disclosure exceeds 90% for extreme hints across 30,000 follow-up samples. Instead of acknowledging the injected reasoning, models fabricate aligned-appearing but unrelated explanations. Activation analysis reveals that sycophancy- and deception-related directions are strongly activated during these fabrications, suggesting systematic patterns rather than incidental failures. Our findings reveal a gap between the reasoning LRMs follow and the reasoning they report, raising concern that aligned-appearing explanations may not be equivalent to genuine alignment. △ Less

Submitted 20 March, 2026; originally announced March 2026.

Adaptive Layerwise Perturbation: Unifying Off-Policy Corrections for LLM RL

Authors: Chenlu Ye, Xuanchang Zhang, Yifan Hao, Zhou Yu, Ziji Zhang, Abhinav Gullapalli, Hao Chen, Jing Huang, Tong Zhang

Abstract: Off-policy problems such as policy staleness and training-inference mismatch, has become a major bottleneck for training stability and further exploration for LLM RL. To enhance inference efficiency, the distribution gap between the inference and updated policy grows, leading to heavy-tailed importance ratios. Heavy-tailed ratios arise when the policy is locally sharp, which further inflates sharp… ▽ More Off-policy problems such as policy staleness and training-inference mismatch, has become a major bottleneck for training stability and further exploration for LLM RL. To enhance inference efficiency, the distribution gap between the inference and updated policy grows, leading to heavy-tailed importance ratios. Heavy-tailed ratios arise when the policy is locally sharp, which further inflates sharp gradients and can push updates outside the trust region. To address this, we propose Adaptive Layerwise Perturbation(ALP) by injecting small learnable perturbations into input hidden states of each layer during updates, which is used as the numerator of the importance ratio against the unchanged inference policy in the objective. Intuitively, by adding controlled noise to intermediate representations, ALP prevents the updated policy from deviating too sharply from the inference policy, and enlarges the policy family to cover the inference policy family with mismatch noises. Hence, the flattened distribution can naturally tighten the updated and inference policy gap and reduce the tail of importance ratios, thus maintaining training stability. This is further validated empirically. Experiments on single-turn math and multi-turn tool-integrated reasoning tasks show that ALP not only improves final performance, but also avoid blow up of importance ratio tail and KL spikes during iterative training, along with boosted exploration. Ablations show that representation-level perturbations across all layers are most effective, substantially outperforming partial-layer and logits-only variants. △ Less

Submitted 19 March, 2026; originally announced March 2026.

DA-Mamba: Learning Domain-Aware State Space Model for Global-Local Alignment in Domain Adaptive Object Detection

Authors: Haochen Li, Rui Zhang, Hantao Yao, Xin Zhang, Yifan Hao, Shaohui Peng, Yongwei Zhao, Ling Li

Abstract: Domain Adaptive Object Detection (DAOD) aims to transfer detectors from a labeled source domain to an unlabeled target domain. Existing DAOD methods employ multi-granularity feature alignment to learn domain-invariant representations. However, the local connectivity of their CNN-based backbone and detection head restricts alignment to local regions, failing to extract global domain-invariant featu… ▽ More Domain Adaptive Object Detection (DAOD) aims to transfer detectors from a labeled source domain to an unlabeled target domain. Existing DAOD methods employ multi-granularity feature alignment to learn domain-invariant representations. However, the local connectivity of their CNN-based backbone and detection head restricts alignment to local regions, failing to extract global domain-invariant features. Although transformer-based DAOD methods capture global dependencies via attention mechanisms, their quadratic computational cost hinders practical deployment. To solve this, we propose DA-Mamba, a hybrid CNN-State Space Models (SSMs) architecture that combines the efficiency of CNNs with the linear-time long-range modeling capability of State Space Models (SSMs) to capture both global and local domain-invariant features. Specifically, we introduce two novel modules: Image-Aware SSM (IA-SSM) and Object-Aware SSM (OA-SSM). IA-SSM is integrated into the backbone to enhance global domain awareness, enabling image-level global and local alignment. OA-SSM is inserted into the detection head to model spatial and semantic dependencies among objects, enhancing instance-level alignment. Comprehensive experiments demonstrate that the proposed method can efficiently improve the cross-domain performance of the object detector. △ Less

Submitted 19 March, 2026; originally announced March 2026.

Comments: Accepted by CVPR 2026

GNNVerifier: Graph-based Verifier for LLM Task Planning

Authors: Yu Hao, Qiuyu Wang, Cheng Yang, Yawen Li, Zhiqiang Zhang, Chuan Shi

Abstract: Large language models (LLMs) facilitate the development of autonomous agents. As a core component of such agents, task planning aims to decompose complex natural language requests into concrete, solvable sub-tasks. Since LLM-generated plans are frequently prone to hallucinations and sensitive to long-context prom-pts, recent research has introduced plan verifiers to identify and correct potential… ▽ More Large language models (LLMs) facilitate the development of autonomous agents. As a core component of such agents, task planning aims to decompose complex natural language requests into concrete, solvable sub-tasks. Since LLM-generated plans are frequently prone to hallucinations and sensitive to long-context prom-pts, recent research has introduced plan verifiers to identify and correct potential flaws. However, most existing approaches still rely on an LLM as the verifier via additional prompting for plan review or self-reflection. LLM-based verifiers can be misled by plausible narration and struggle to detect failures caused by structural relations across steps, such as type mismatches, missing intermediates, or broken dependencies. To address these limitations, we propose a graph-based verifier for LLM task planning. Specifically, the proposed method has four major components: Firstly, we represent a plan as a directed graph with enriched attributes, where nodes denote sub-tasks and edges encode execution order and dependency constraints. Secondly, a graph neural network (GNN) then performs structural evaluation and diagnosis, producing a graph-level plausibility score for plan acceptance as well as node/edge-level risk scores to localize erroneous regions. Thirdly, we construct controllable perturbations from ground truth plan graphs, and automatically generate training data with fine-grained annotations. Finally, guided by the feedback from our GNN verifier, we enable an LLM to conduct local edits (e.g., tool replacement or insertion) to correct the plan when the graph-level score is insufficient. Extensive experiments across diverse datasets, backbone LLMs, and planners demonstrate that our GNNVerifier achieves significant gains in improving plan quality. Our data and code is available at https://github.com/BUPT-GAMMA/GNNVerifier. △ Less

Submitted 17 March, 2026; v1 submitted 15 March, 2026; originally announced March 2026.

Comments: 17pages,12figures

Think While Watching: Online Streaming Segment-Level Memory for Multi-Turn Video Reasoning in Multimodal Large Language Models

Authors: Lu Wang, Zhuoran Jin, Yupu Hao, Yubo Chen, Kang Liu, Yulong Ao, Jun Zhao

Abstract: Multimodal large language models (MLLMs) have shown strong performance on offline video understanding, but most are limited to offline inference or have weak online reasoning, making multi-turn interaction over continuously arriving video streams difficult. Existing streaming methods typically use an interleaved perception-generation paradigm, which prevents concurrent perception and generation an… ▽ More Multimodal large language models (MLLMs) have shown strong performance on offline video understanding, but most are limited to offline inference or have weak online reasoning, making multi-turn interaction over continuously arriving video streams difficult. Existing streaming methods typically use an interleaved perception-generation paradigm, which prevents concurrent perception and generation and leads to early memory decay as streams grow, hurting long-range dependency modeling. We propose Think While Watching, a memory-anchored streaming video reasoning framework that preserves continuous segment-level memory during multi-turn interaction. We build a three-stage, multi-round chain-of-thought dataset and adopt a stage-matched training strategy, while enforcing strict causality through a segment-level streaming causal mask and streaming positional encoding. During inference, we introduce an efficient pipeline that overlaps watching and thinking and adaptively selects the best attention backend. Under both single-round and multi-round streaming input protocols, our method achieves strong results. Built on Qwen3-VL, it improves single-round accuracy by 2.6% on StreamingBench and by 3.79% on OVO-Bench. In the multi-round setting, it maintains performance while reducing output tokens by 56%. Code is available at: https://github.com/wl666hhh/Think_While_Watching/ △ Less

Submitted 12 March, 2026; originally announced March 2026.

Multi-Agent Collaboration for Automated Design Exploration on High Performance Computing Systems

Authors: Harshitha Menon, Charles F. Jekel, Kevin Korner, Brian Gunnarson, Nathan K. Brown, Michael Stees, M. Giselle Fernandez-Godino, Walter Nissen, Meir H. Shachar, Dane M. Sterbentz, William J. Schill, Yue Hao, Robert Rieben, William Quadros, Steve Owen, Scott Mitchell, Ismael D. Boureima, Jonathan L. Belof

Abstract: Today's scientific challenges, from climate modeling to Inertial Confinement Fusion design to novel material design, require exploring huge design spaces. In order to enable high-impact scientific discovery, we need to scale up our ability to test hypotheses, generate results, and learn from them rapidly. We present MADA (Multi-Agent Design Assistant), a Large Language Model (LLM) powered multi-ag… ▽ More Today's scientific challenges, from climate modeling to Inertial Confinement Fusion design to novel material design, require exploring huge design spaces. In order to enable high-impact scientific discovery, we need to scale up our ability to test hypotheses, generate results, and learn from them rapidly. We present MADA (Multi-Agent Design Assistant), a Large Language Model (LLM) powered multi-agent framework that coordinates specialized agents for complex design workflows. A Job Management Agent (JMA) launches and manages ensemble simulations on HPC systems, a Geometry Agent (GA) generates meshes, and an Inverse Design Agent (IDA) proposes new designs informed by simulation outcomes. While general purpose, we focus development and validation on Richtmyer--Meshkov Instability (RMI) suppression, a critical challenge in Inertial Confinement Fusion. We evaluate on two complementary settings: running a hydrodynamics simulations on HPC systems, and using a pre-trained machine learning surrogate for rapid design exploration. Our results demonstrate that the MADA system successfully executes iterative design refinement, automatically improving designs toward optimal RMI suppression with minimal manual intervention. Our framework reduces cumbersome manual workflow setup, and enables automated design exploration at scale. More broadly, it demonstrates a reusable pattern for coupling reasoning, simulation, specialized tools, and coordinated workflows to accelerate scientific discovery. △ Less

Submitted 12 March, 2026; originally announced March 2026.

SlideSparse: Fast and Flexible (2N-2):2N Structured Sparsity

Authors: Hanyong Shao, Yingbo Hao, Ting Song, Yan Xia, Di Zhang, Shaohan Huang, Xun Wu, Songchen Xu, Le Xu, Li Dong, Zewen Chi, Yi Zou, Furu Wei

Abstract: NVIDIA's 2:4 Sparse Tensor Cores deliver 2x throughput but demand strict 50% pruning -- a ratio that collapses LLM reasoning accuracy (Qwen3: 54% to 15%). Milder $(2N-2):2N$ patterns (e.g., 6:8, 25% pruning) preserve accuracy yet receive no hardware support, falling back to dense execution without any benefit from sparsity. We present SlideSparse, the first system to unlock Sparse Tensor Core acce… ▽ More NVIDIA's 2:4 Sparse Tensor Cores deliver 2x throughput but demand strict 50% pruning -- a ratio that collapses LLM reasoning accuracy (Qwen3: 54% to 15%). Milder $(2N-2):2N$ patterns (e.g., 6:8, 25% pruning) preserve accuracy yet receive no hardware support, falling back to dense execution without any benefit from sparsity. We present SlideSparse, the first system to unlock Sparse Tensor Core acceleration for the $(2N-2):2N$ model family on commodity GPUs. Our Sliding Window Decomposition reconstructs any $(2N-2):2N$ weight block into $N-1$ overlapping 2:4-compliant windows without any accuracy loss; Activation Lifting fuses the corresponding activation rearrangement into per-token quantization at near-zero cost. Integrated into vLLM, SlideSparse is evaluated across various GPUs (A100, H100, B200, RTX 4090, RTX 5080, DGX-spark), precisions (FP4, INT8, FP8, BF16, FP16), and model families (Llama, Qwen, BitNet). On compute-bound workloads, the measured speedup ratio (1.33x) approaches the theoretical upper-bound $N/(N-1)=4/3$ at 6:8 weight sparsity in Qwen2.5-7B, establishing $(2N-2):2N$ as a practical path to accuracy-preserving LLM acceleration. Code available at https://github.com/bcacdwk/vllmbench. △ Less

Submitted 5 March, 2026; originally announced March 2026.

Critic in the Loop: A Tri-System VLA Framework for Robust Long-Horizon Manipulation

Authors: Pengfei Yi, Yingjie Ma, Wenjiang Xu, Yanan Hao, Shuai Gan, Wanting Li, Shanlin Zhong

Abstract: Balancing high-level semantic reasoning with low-level reactive control remains a core challenge in visual robotic manipulation. While Vision-Language Models (VLMs) excel at cognitive planning, their inference latency precludes real-time execution. Conversely, fast Vision-Language-Action (VLA) models often lack the semantic depth required for complex, long-horizon tasks. To bridge this gap, we int… ▽ More Balancing high-level semantic reasoning with low-level reactive control remains a core challenge in visual robotic manipulation. While Vision-Language Models (VLMs) excel at cognitive planning, their inference latency precludes real-time execution. Conversely, fast Vision-Language-Action (VLA) models often lack the semantic depth required for complex, long-horizon tasks. To bridge this gap, we introduce Critic in the Loop, an adaptive hierarchical framework driven by dynamic VLM-Expert scheduling. At its core is a bionic Tri-System architecture comprising a VLM brain for global reasoning, a VLA cerebellum for reactive execution, and a lightweight visual Critic. By continuously monitoring the workspace, the Critic dynamically routes control authority. It sustains rapid closed-loop execution via the VLA for routine subtasks, and adaptively triggers the VLM for replanning upon detecting execution anomalies such as task stagnation or failures. Furthermore, our architecture seamlessly integrates human-inspired rules to intuitively break infinite retry loops. This visually-grounded scheduling minimizes expensive VLM queries, while substantially enhancing system robustness and autonomy in out-of-distribution (OOD) scenarios. Comprehensive experiments on challenging, long-horizon manipulation benchmarks reveal that our approach achieves state-of-the-art performance. △ Less

Submitted 5 March, 2026; originally announced March 2026.

Sparse-BitNet: 1.58-bit LLMs are Naturally Friendly to Semi-Structured Sparsity

Authors: Di Zhang, Xun Wu, Shaohan Huang, Yudong Wang, Hanyong Shao, Yingbo Hao, Zewen Chi, Li Dong, Ting Song, Yan Xia, Zhifang Sui, Furu Wei

Abstract: Semi-structured N:M sparsity and low-bit quantization (e.g., 1.58-bit BitNet) are two promising approaches for improving the efficiency of large language models (LLMs), yet they have largely been studied in isolation. In this work, we investigate their interaction and show that 1.58-bit BitNet is naturally more compatible with N:M sparsity than full-precision models. To study this effect, we propo… ▽ More Semi-structured N:M sparsity and low-bit quantization (e.g., 1.58-bit BitNet) are two promising approaches for improving the efficiency of large language models (LLMs), yet they have largely been studied in isolation. In this work, we investigate their interaction and show that 1.58-bit BitNet is naturally more compatible with N:M sparsity than full-precision models. To study this effect, we propose Sparse-BitNet, a unified framework that jointly applies 1.58-bit quantization and dynamic N:M sparsification while ensuring stable training for the first time. Across multiple model scales and training regimes (sparse pretraining and dense-to-sparse schedules), 1.58-bit BitNet consistently exhibits smaller performance degradation than full-precision baselines at the same sparsity levels and can tolerate higher structured sparsity before accuracy collapse. Moreover, using our custom sparse tensor core, Sparse-BitNet achieves substantial speedups in both training and inference, reaching up to 1.30X. These results highlight that combining extremely low-bit quantization with semi-structured N:M sparsity is a promising direction for efficient LLMs. Code available at https://github.com/AAzdi/Sparse-BitNet △ Less

Submitted 5 March, 2026; originally announced March 2026.

Look Forward to Walk Backward: Efficient Terrain Memory for Backward Locomotion with Forward Vision

Authors: Shixin Luo, Songbo Li, Yuan Hao, Yaqi Wang, Jun Zheng, Jun Wu, Qiuguo Zhu

Abstract: Legged robots with egocentric forward-facing depth cameras can couple exteroception and proprioception to achieve robust forward agility on complex terrain. When these robots walk backward, the forward-only field of view provides no preview. Purely proprioceptive controllers can remain stable on moderate ground when moving backward but cannot fully exploit the robot's capabilities on complex terra… ▽ More Legged robots with egocentric forward-facing depth cameras can couple exteroception and proprioception to achieve robust forward agility on complex terrain. When these robots walk backward, the forward-only field of view provides no preview. Purely proprioceptive controllers can remain stable on moderate ground when moving backward but cannot fully exploit the robot's capabilities on complex terrain and must collide with obstacles. We present Look Forward to Walk Backward (LF2WB), an efficient terrain-memory locomotion framework that uses forward egocentric depth and proprioception to write a compact associative memory during forward motion and to retrieve it for collision-free backward locomotion without rearward vision. The memory backbone employs a delta-rule selective update that softly removes then writes the memory state along the active subspace. Training uses hardware-efficient parallel computation, and deployment runs recurrent, constant-time per-step inference with a constant-size state, making the approach suitable for onboard processors on low-cost robots. Experiments in both simulations and real-world scenarios demonstrate the effectiveness of our method, improving backward agility across complex terrains under limited sensing. △ Less

Submitted 3 March, 2026; originally announced March 2026.

Comments: Accepted for 2026 IEEE International Conference on Robotics and Automation (ICRA)

IdGlow: Dynamic Identity Modulation for Multi-Subject Generation

Authors: Honghao Cai, Xiangyuan Wang, Yunhao Bai, Tianze Zhou, Sijie Xu, Yuyang Hao, Zezhou Cui, Yuyuan Yang, Wei Zhu, Yibo Chen, Xu Tang, Yao Hu, Zhen Li

Abstract: Multi-subject image generation requires seamlessly harmonizing multiple reference identities within a coherent scene. However, existing methods relying on rigid spatial masks or localized attention often struggle with the "stability-plasticity dilemma," particularly failing in tasks that require complex structural deformations, such as identity-preserving age transformation. To address this, we pr… ▽ More Multi-subject image generation requires seamlessly harmonizing multiple reference identities within a coherent scene. However, existing methods relying on rigid spatial masks or localized attention often struggle with the "stability-plasticity dilemma," particularly failing in tasks that require complex structural deformations, such as identity-preserving age transformation. To address this, we present IdGlow, a mask-free, progressive two-stage framework built upon Flow Matching diffusion models. In the supervised fine-tuning (SFT) stage, we introduce task-adaptive timestep scheduling aligned with diffusion generative dynamics: a linear decay schedule that progressively relaxes constraints for natural group composition, and a temporal gating mechanism that concentrates identity injection within a critical semantic window, successfully preserving adult facial semantics without overriding child-like anatomical structures. To resolve attribute leakage and semantic ambiguity without explicit layout inputs, we further integrate a badcase-driven Vision-Language Model (VLM) for precise, context-aware prompt synthesis. In the second stage, we design a Fine-Grained Group-Level Direct Preference Optimization (DPO) with a weighted margin formulation to simultaneously eliminate multi-subject artifacts, elevate texture harmony, and recalibrate identity fidelity towards real-world distributions. Extensive experiments on two challenging benchmarks -- direct multi-person fusion and age-transformed group generation -- demonstrate that IdGlow fundamentally mitigates the stability-plasticity conflict, achieving a superior Pareto balance between state-of-the-art facial fidelity and commercial-grade aesthetic quality. △ Less

Submitted 28 February, 2026; originally announced March 2026.

MPU: Towards Secure and Privacy-Preserving Knowledge Unlearning for Large Language Models

Authors: Tiantong Wang, Xinyu Yan, Tiantong Wu, Yurong Hao, Yong Jiang, Fei Huang, Wei Yang Bryan Lim

Abstract: Machine unlearning for large language models often faces a privacy dilemma in which strict constraints prohibit sharing either the server's parameters or the client's forget set. To address this dual non-disclosure constraint, we propose MPU, an algorithm-agnostic privacy-preserving Multiple Perturbed Copies Unlearning framework that primarily introduces two server-side modules: Pre-Process for ra… ▽ More Machine unlearning for large language models often faces a privacy dilemma in which strict constraints prohibit sharing either the server's parameters or the client's forget set. To address this dual non-disclosure constraint, we propose MPU, an algorithm-agnostic privacy-preserving Multiple Perturbed Copies Unlearning framework that primarily introduces two server-side modules: Pre-Process for randomized copy generation and Post-Process for update aggregation. In Pre-Process, the server distributes multiple perturbed and reparameterized model instances, allowing the client to execute unlearning locally on its private forget set without accessing the server's exact original parameters. After local unlearning, the server performs Post-Process by inverting the reparameterization and aggregating updates with a harmonic denoising procedure to alleviate the impact of perturbation. Experiments with seven unlearning algorithms show that MPU achieves comparable unlearning performance to noise-free baselines, with most algorithms' average degradation well below 1% under 10% noise, and can even outperform the noise-free baseline for some algorithms under 1% noise. Code is available at https://github.com/Tristan-SHU/MPU. △ Less

Submitted 27 February, 2026; originally announced February 2026.

FAMOSE: A ReAct Approach to Automated Feature Discovery

Authors: Keith Burghardt, Jienan Liu, Sadman Sakib, Yuning Hao, Bo Li

Abstract: Feature engineering remains a critical yet challenging bottleneck in machine learning, particularly for tabular data, as identifying optimal features from an exponentially large feature space traditionally demands substantial domain expertise. To address this challenge, we introduce FAMOSE (Feature AugMentation and Optimal Selection agEnt), a novel framework that leverages the ReAct paradigm to au… ▽ More Feature engineering remains a critical yet challenging bottleneck in machine learning, particularly for tabular data, as identifying optimal features from an exponentially large feature space traditionally demands substantial domain expertise. To address this challenge, we introduce FAMOSE (Feature AugMentation and Optimal Selection agEnt), a novel framework that leverages the ReAct paradigm to autonomously explore, generate, and refine features while integrating feature selection and evaluation tools within an agent architecture. To our knowledge, FAMOSE represents the first application of an agentic ReAct framework to automated feature engineering, especially for both regression and classification tasks. Extensive experiments demonstrate that FAMOSE is at or near the state-of-the-art on classification tasks (especially tasks with more than 10K instances, where ROC-AUC increases 0.23% on average), and achieves the state-of-the-art for regression tasks by reducing RMSE by 2.0% on average, while remaining more robust to errors than other algorithms. We hypothesize that FAMOSE's strong performance is because ReAct allows the LLM context window to record (via iterative feature discovery and evaluation steps) what features did or did not work. This is similar to a few-shot prompt and guides the LLM to invent better, more innovative features. Our work offers evidence that AI agents are remarkably effective in solving problems that require highly inventive solutions, such as feature engineering. △ Less

Submitted 19 February, 2026; originally announced February 2026.

Comments: 23 pages, 6 figures

Deeper detection limits in astronomical imaging using self-supervised spatiotemporal denoising

Authors: Yuduo Guo, Hao Zhang, Mingyu Li, Fujiang Yu, Yunjing Wu, Yuhan Hao, Song Huang, Yongming Liang, Xiaojing Lin, Xinyang Li, Jiamin Wu, Zheng Cai, Qionghai Dai

Abstract: The detection limit of astronomical imaging observations is limited by several noise sources. Some of that noise is correlated between neighbouring image pixels and exposures, so in principle could be learned and corrected. We present an astronomical self-supervised transformer-based denoising algorithm (ASTERIS), that integrates spatiotemporal information across multiple exposures. Benchmarking o… ▽ More The detection limit of astronomical imaging observations is limited by several noise sources. Some of that noise is correlated between neighbouring image pixels and exposures, so in principle could be learned and corrected. We present an astronomical self-supervised transformer-based denoising algorithm (ASTERIS), that integrates spatiotemporal information across multiple exposures. Benchmarking on mock data indicates that ASTERIS improves detection limits by 1.0 magnitude at 90% completeness and purity, while preserving the point spread function and photometric accuracy. Observational validation using data from the James Webb Space Telescope (JWST) and Subaru telescope identifies previously undetectable features, including low-surface-brightness galaxy structures and gravitationally-lensed arcs. Applied to deep JWST images, ASTERIS identifies three times more redshift > 9 galaxy candidates, with rest-frame ultraviolet luminosity 1.0 magnitude fainter, than previous methods. △ Less

Submitted 19 February, 2026; originally announced February 2026.

Comments: Published in Science. This is the author's version of the work. It is posted here by permission of the AAAS for personal use, not for redistribution

ProbeLLM: Automating Principled Diagnosis of LLM Failures

Authors: Yue Huang, Zhengzhe Jiang, Yuchen Ma, Yu Jiang, Xiangqi Wang, Yujun Zhou, Yuexing Hao, Kehan Guo, Pin-Yu Chen, Stefan Feuerriegel, Xiangliang Zhang

Abstract: Understanding how and why large language models (LLMs) fail is becoming a central challenge as models rapidly evolve and static evaluations fall behind. While automated probing has been enabled by dynamic test generation, existing approaches often discover isolated failure cases, lack principled control over exploration, and provide limited insight into the underlying structure of model weaknesses… ▽ More Understanding how and why large language models (LLMs) fail is becoming a central challenge as models rapidly evolve and static evaluations fall behind. While automated probing has been enabled by dynamic test generation, existing approaches often discover isolated failure cases, lack principled control over exploration, and provide limited insight into the underlying structure of model weaknesses. We propose ProbeLLM, a benchmark-agnostic automated probing framework that elevates weakness discovery from individual failures to structured failure modes. ProbeLLM formulates probing as a hierarchical Monte Carlo Tree Search, explicitly allocating limited probing budgets between global exploration of new failure regions and local refinement of recurring error patterns. By restricting probing to verifiable test cases and leveraging tool-augmented generation and verification, ProbeLLM grounds failure discovery in reliable evidence. Discovered failures are further consolidated into interpretable failure modes via failure-aware embeddings and boundary-aware induction. Across diverse benchmarks and LLMs, ProbeLLM reveals substantially broader, cleaner, and more fine-grained failure landscapes than static benchmarks and prior automated methods, supporting a shift from case-centric evaluation toward principled weakness discovery. △ Less

Submitted 13 February, 2026; originally announced February 2026.

"Not Human, Funnier": How Machine Identity Shapes Humor Perception in Online AI Stand-up Comedy

Authors: Xuehan Huang, Canwen Wang, Yifei Hao, Daijin Yang, Ray LC

Abstract: Chatbots are increasingly applied to domains previously reserved for human actors. One such domain is comedy, whereby both the general public working with ChatGPT and research-based LLM-systems have tried their hands on making humor. In formative interviews with professional comedians and video analyses of stand-up comedy in humans, we found that human performers often use their ethnic, gender, co… ▽ More Chatbots are increasingly applied to domains previously reserved for human actors. One such domain is comedy, whereby both the general public working with ChatGPT and research-based LLM-systems have tried their hands on making humor. In formative interviews with professional comedians and video analyses of stand-up comedy in humans, we found that human performers often use their ethnic, gender, community, and demographic-based identity to enable joke-making. This suggests whether the identity of AI itself can empower AI humor generation for human audiences. We designed a machine-identity-based agent that uses its own status as AI to tell jokes in online performance format. Studies with human audiences (N=32) showed that machine-identity-based agents were seen as funnier than baseline-GPT agent. This work suggests the design of human-AI integrated systems that explicitly utilize AI as its own unique identity apart from humans. △ Less

Submitted 13 February, 2026; originally announced February 2026.

Comments: 27 pages, 5 figures. Conditionally Accepted to CHI '26

Pretraining with Token-Level Adaptive Latent Chain-of-Thought

Authors: Boyi Zeng, Yiqin Hao, He Li, Shixiang Song, Feichen Song, Zitong Wang, Siyuan Huang, Yi Xu, ZiWei He, Xinbing Wang, Zhouhan Lin

Abstract: Scaling large language models by increasing parameters and training data is increasingly constrained by limited high-quality corpora and rising communication costs. This work explores an alternative axis: increasing per-token computation without expanding parameters, by internalizing latent Chain-of-Thought (CoT) into pretraining. We propose Pretraining with Token-Level Adaptive Latent CoT (adapti… ▽ More Scaling large language models by increasing parameters and training data is increasingly constrained by limited high-quality corpora and rising communication costs. This work explores an alternative axis: increasing per-token computation without expanding parameters, by internalizing latent Chain-of-Thought (CoT) into pretraining. We propose Pretraining with Token-Level Adaptive Latent CoT (adaptive latent CoT), where the model generates a variable-length latent CoT trajectory before emitting each token -- allocating longer trajectories to difficult tokens and shorter (or even zero) trajectories to easy ones. Importantly, this behavior emerges naturally from one-stage pretraining on general text and reduces computation in both training and inference via token-wise adaptive halting. Experiments with Llama architectures show that adaptive latent CoT consistently improves language modeling perplexity and broad downstream accuracy, even with fewer training FLOPs than prior recurrent baselines. △ Less

Submitted 10 March, 2026; v1 submitted 8 February, 2026; originally announced February 2026.

Comments: 15pages

VENOMREC: Cross-Modal Interactive Poisoning for Targeted Promotion in Multimodal LLM Recommender Systems

Authors: Guowei Guan, Yurong Hao, Jiaming Zhang, Tiantong Wu, Fuyao Zhang, Tianxiang Chen, Longtao Huang, Cyril Leung, Wei Yang Bryan Lim

Abstract: Multimodal large language models (MLLMs) are pushing recommender systems (RecSys) toward content-grounded retrieval and ranking via cross-modal fusion. We find that while cross-modal consensus often mitigates conventional poisoning that manipulates interaction logs or perturbs a single modality, it also introduces a new attack surface where synchronised multimodal poisoning can reliably steer fuse… ▽ More Multimodal large language models (MLLMs) are pushing recommender systems (RecSys) toward content-grounded retrieval and ranking via cross-modal fusion. We find that while cross-modal consensus often mitigates conventional poisoning that manipulates interaction logs or perturbs a single modality, it also introduces a new attack surface where synchronised multimodal poisoning can reliably steer fused representations along stable semantic directions during fine-tuning. To characterise this threat, we formalise cross-modal interactive poisoning and propose VENOMREC, which performs Exposure Alignment to identify high-exposure regions in the joint embedding space and Cross-modal Interactive Perturbation to craft attention-guided coupled token-patch edits. Experiments on three real-world multimodal datasets demonstrate that VENOMREC consistently outperforms strong baselines, achieving 0.73 mean ER@20 and improving over the strongest baseline by +0.52 absolute ER points on average, while maintaining comparable recommendation utility. △ Less

Submitted 6 February, 2026; originally announced February 2026.

ProphetKV: User-Query-Driven Selective Recomputation for Efficient KV Cache Reuse in Retrieval-Augmented Generation

Authors: Shihao Wang, Jiahao Chen, Yanqi Pan, Hao Huang, Yichen Hao, Xiangyu Zou, Wen Xia, Wentao Zhang, Chongyang Qiu, Pengfei Wang

Abstract: The prefill stage of long-context Retrieval-Augmented Generation (RAG) is severely bottlenecked by computational overhead. To mitigate this, recent methods assemble pre-calculated KV caches of retrieved RAG documents (by a user query) and reprocess selected tokens to recover cross-attention between these pre-calculated KV caches. However, we identify a fundamental "crowding-out effect" in current… ▽ More The prefill stage of long-context Retrieval-Augmented Generation (RAG) is severely bottlenecked by computational overhead. To mitigate this, recent methods assemble pre-calculated KV caches of retrieved RAG documents (by a user query) and reprocess selected tokens to recover cross-attention between these pre-calculated KV caches. However, we identify a fundamental "crowding-out effect" in current token selection criteria: globally salient but user-query-irrelevant tokens saturate the limited recomputation budget, displacing the tokens truly essential for answering the user query and degrading inference accuracy. We propose ProphetKV, a user-query-driven KV Cache reuse method for RAG scenarios. ProphetKV dynamically prioritizes tokens based on their semantic relevance to the user query and employs a dual-stage recomputation pipeline to fuse layer-wise attention metrics into a high-utility set. By ensuring the recomputation budget is dedicated to bridging the informational gap between retrieved context and the user query, ProphetKV achieves high-fidelity attention recovery with minimal overhead. Our extensive evaluation results show that ProphetKV retains 96%-101% of full-prefill accuracy with only a 20% recomputation ratio, while achieving accuracy improvements of 8.8%-24.9% on RULER and 18.6%-50.9% on LongBench over the state-of-the-art approaches (e.g., CacheBlend, EPIC, and KVShare). △ Less

Submitted 4 February, 2026; v1 submitted 31 January, 2026; originally announced February 2026.

Physics-informed Diffusion Generation for Geomagnetic Map Interpolation

Authors: Wenda Li, Tongya Zheng, Kaixuan Chen, Shunyu Liu, Haoze Jiang, Yunzhi Hao, Rui Miao, Zujie Ren, Mingli Song, Hang Shi, Gang Chen

Abstract: Geomagnetic map interpolation aims to infer unobserved geomagnetic data at spatial points, yielding critical applications in navigation and resource exploration. However, existing methods for scattered data interpolation are not specifically designed for geomagnetic maps, which inevitably leads to suboptimal performance due to detection noise and the laws of physics. Therefore, we propose a Physic… ▽ More Geomagnetic map interpolation aims to infer unobserved geomagnetic data at spatial points, yielding critical applications in navigation and resource exploration. However, existing methods for scattered data interpolation are not specifically designed for geomagnetic maps, which inevitably leads to suboptimal performance due to detection noise and the laws of physics. Therefore, we propose a Physics-informed Diffusion Generation framework~(PDG) to interpolate incomplete geomagnetic maps. First, we design a physics-informed mask strategy to guide the diffusion generation process based on a local receptive field, effectively eliminating noise interference. Second, we impose a physics-informed constraint on the diffusion generation results following the kriging principle of geomagnetic maps, ensuring strict adherence to the laws of physics. Extensive experiments and in-depth analyses on four real-world datasets demonstrate the superiority and effectiveness of each component of PDG. △ Less

Submitted 31 January, 2026; originally announced February 2026.

Comments: 5 pages, 2 figures, IEEE ICASSP'26

Yunque DeepResearch Technical Report

Authors: Yuxuan Cai, Xinyi Lai, Peng Yuan, Weiting Liu, Huajian Li, Mingda Li, Xinghua Wang, Shengxie Zheng, Yanchao Hao, Yuyang Yin, Zheng Wei

Abstract: Deep research has emerged as a transformative capability for autonomous agents, empowering Large Language Models to navigate complex, open-ended tasks. However, realizing its full potential is hindered by critical limitations, including escalating contextual noise in long-horizon tasks, fragility leading to cascading errors, and a lack of modular extensibility. To address these challenges, we intr… ▽ More Deep research has emerged as a transformative capability for autonomous agents, empowering Large Language Models to navigate complex, open-ended tasks. However, realizing its full potential is hindered by critical limitations, including escalating contextual noise in long-horizon tasks, fragility leading to cascading errors, and a lack of modular extensibility. To address these challenges, we introduce Yunque DeepResearch, a hierarchical, modular, and robust framework. The architecture is characterized by three key components: (1) a centralized Multi-Agent Orchestration System that routes subtasks to an Atomic Capability Pool of tools and specialized sub-agents; (2) a Dynamic Context Management mechanism that structures completed sub-goals into semantic summaries to mitigate information overload; and (3) a proactive Supervisor Module that ensures resilience through active anomaly detection and context pruning. Yunque DeepResearch achieves state-of-the-art performance across a range of agentic deep research benchmarks, including GAIA, BrowseComp, BrowseComp-ZH, and Humanity's Last Exam. We open-source the framework, reproducible implementations, and application cases to empower the community. △ Less

Submitted 27 January, 2026; originally announced January 2026.

VIBEVOICE-ASR Technical Report

Authors: Zhiliang Peng, Jianwei Yu, Yaoyao Chang, Zilong Wang, Li Dong, Yingbo Hao, Yujie Tu, Chenyu Yang, Wenhui Wang, Songchen Xu, Yutao Sun, Hangbo Bao, Weijiang Xu, Yi Zhu, Zehua Wang, Ting Song, Yan Xia, Zewen Chi, Shaohan Huang, Liang Wang, Chuang Ding, Shuai Wang, Xie Chen, Furu Wei

Abstract: This report presents VibeVoice-ASR, a general-purpose speech understanding framework built upon VibeVoice, designed to address the persistent challenges of context fragmentation and multi-speaker complexity in long-form audio (e.g., meetings, podcasts) that remain despite recent advancements in short-form speech recognition. Unlike traditional pipelined approaches that rely on audio chunking, Vibe… ▽ More This report presents VibeVoice-ASR, a general-purpose speech understanding framework built upon VibeVoice, designed to address the persistent challenges of context fragmentation and multi-speaker complexity in long-form audio (e.g., meetings, podcasts) that remain despite recent advancements in short-form speech recognition. Unlike traditional pipelined approaches that rely on audio chunking, VibeVoice-ASRsupports single-pass processing for up to 60 minutes of audio. It unifies Automatic Speech Recognition, Speaker Diarization, and Timestamping into a single end-to-end generation task. In addition, VibeVoice-ASR supports over 50 languages, requires no explicit language setting, and natively handles code-switching within and across utterances. Furthermore, we introduce a prompt-based context injection mechanism that allows users to supply customized conetxt, significantly improving accuracy on domain-specific terminology and polyphonic character disambiguation. △ Less

Submitted 14 March, 2026; v1 submitted 26 January, 2026; originally announced January 2026.

FastStair: Learning to Run Up Stairs with Humanoid Robots

Authors: Yan Liu, Tao Yu, Haolin Song, Hongbo Zhu, Nianzong Hu, Yuzhi Hao, Xiuyong Yao, Xizhe Zang, Hua Chen, Jie Zhao

Abstract: Running up stairs is effortless for humans but remains extremely challenging for humanoid robots due to the simultaneous requirements of high agility and strict stability. Model-free reinforcement learning (RL) can generate dynamic locomotion, yet implicit stability rewards and heavy reliance on task-specific reward shaping tend to result in unsafe behaviors, especially on stairs; conversely, mode… ▽ More Running up stairs is effortless for humans but remains extremely challenging for humanoid robots due to the simultaneous requirements of high agility and strict stability. Model-free reinforcement learning (RL) can generate dynamic locomotion, yet implicit stability rewards and heavy reliance on task-specific reward shaping tend to result in unsafe behaviors, especially on stairs; conversely, model-based foothold planners encode contact feasibility and stability structure, but enforcing their hard constraints often induces conservative motion that limits speed. We present FastStair, a planner-guided, multi-stage learning framework that reconciles these complementary strengths to achieve fast and stable stair ascent. FastStair integrates a parallel model-based foothold planner into the RL training loop to bias exploration toward dynamically feasible contacts and to pretrain a safety-focused base policy. To mitigate planner-induced conservatism and the discrepancy between low- and high-speed action distributions, the base policy was fine-tuned into speed-specialized experts and then integrated via Low-Rank Adaptation (LoRA) to enable smooth operation across the full commanded-speed range. We deploy the resulting controller on the Oli humanoid robot, achieving stable stair ascent at commanded speeds up to 1.65 m/s and traversing a 33-step spiral staircase (17 cm rise per step) in 12 s, demonstrating robust high-speed performance on long staircases. Notably, the proposed approach served as the champion solution in the Canton Tower Robot Run Up Competition. △ Less

Submitted 15 January, 2026; originally announced January 2026.

A Grouped Sorting Queue Supporting Dynamic Updates for Timer Management in High-Speed Network Interface Cards

Authors: Zekun Wang, Binghao Yue, Weitao Pan, Jianyi Shi, Yue Hao

Abstract: With the hardware offloading of network functions, network interface cards (NICs) undertake massive stateful, high-precision, and high-throughput tasks, where timers serve as a critical enabling component. However, existing timer management schemes suffer from heavy software load, low precision, lack of hardware update support, and overflow. This paper proposes two novel operations for priority qu… ▽ More With the hardware offloading of network functions, network interface cards (NICs) undertake massive stateful, high-precision, and high-throughput tasks, where timers serve as a critical enabling component. However, existing timer management schemes suffer from heavy software load, low precision, lack of hardware update support, and overflow. This paper proposes two novel operations for priority queues--update and group sorting--to enable hardware timer management. To the best of our knowledge, this work presents the first hardware priority queue to support an update operation through the composition and propagation of basic operations to modify the priorities of elements within the queue. The group sorting mechanism ensures correct timing behavior post-overflow by establishing a group boundary priority to alter the sorting process and element insertion positions. Implemented with a hybrid architecture of a one-dimension (1D) systolic array and shift registers, our design is validated through packet-level simulations for flow table timeout management. Results demonstrate that a 4K-depth, 16-bit timer queue achieves over 500 MHz (175 Mpps, 12 ns precision) in a 28nm process and over 300 MHz (116 Mpps) on an FPGA. Critically, it reduces LUTs and FFs usage by 31% and 25%, respectively, compared to existing designs. △ Less

Submitted 13 January, 2026; originally announced January 2026.

Multiplex Thinking: Reasoning via Token-wise Branch-and-Merge

Authors: Yao Tang, Li Dong, Yaru Hao, Qingxiu Dong, Furu Wei, Jiatao Gu

Abstract: Large language models often solve complex reasoning tasks more effectively with Chain-of-Thought (CoT), but at the cost of long, low-bandwidth token sequences. Humans, by contrast, often reason softly by maintaining a distribution over plausible next steps. Motivated by this, we propose Multiplex Thinking, a stochastic soft reasoning mechanism that, at each thinking step, samples K candidate token… ▽ More Large language models often solve complex reasoning tasks more effectively with Chain-of-Thought (CoT), but at the cost of long, low-bandwidth token sequences. Humans, by contrast, often reason softly by maintaining a distribution over plausible next steps. Motivated by this, we propose Multiplex Thinking, a stochastic soft reasoning mechanism that, at each thinking step, samples K candidate tokens and aggregates their embeddings into a single continuous multiplex token. This preserves the vocabulary embedding prior and the sampling dynamics of standard discrete generation, while inducing a tractable probability distribution over multiplex rollouts. Consequently, multiplex trajectories can be directly optimized with on-policy reinforcement learning (RL). Importantly, Multiplex Thinking is self-adaptive: when the model is confident, the multiplex token is nearly discrete and behaves like standard CoT; when it is uncertain, it compactly represents multiple plausible next steps without increasing sequence length. Across challenging math reasoning benchmarks, Multiplex Thinking consistently outperforms strong discrete CoT and RL baselines from Pass@1 through Pass@1024, while producing shorter sequences. The code and checkpoints are available at https://github.com/GMLR-Penn/Multiplex-Thinking. △ Less

Submitted 13 January, 2026; originally announced January 2026.

Comments: 21 pages. Code available at https://github.com/GMLR-Penn/Multiplex-Thinking

PC2P: Multi-Agent Path Finding via Personalized-Enhanced Communication and Crowd Perception

Authors: Guotao Li, Shaoyun Xu, Yuexing Hao, Yang Wang, Yuhui Sun

Abstract: Distributed Multi-Agent Path Finding (MAPF) integrated with Multi-Agent Reinforcement Learning (MARL) has emerged as a prominent research focus, enabling real-time cooperative decision-making in partially observable environments through inter-agent communication. However, due to insufficient collaborative and perceptual capabilities, existing methods are inadequate for scaling across diverse envir… ▽ More Distributed Multi-Agent Path Finding (MAPF) integrated with Multi-Agent Reinforcement Learning (MARL) has emerged as a prominent research focus, enabling real-time cooperative decision-making in partially observable environments through inter-agent communication. However, due to insufficient collaborative and perceptual capabilities, existing methods are inadequate for scaling across diverse environmental conditions. To address these challenges, we propose PC2P, a novel distributed MAPF method derived from a Q-learning-based MARL framework. Initially, we introduce a personalized-enhanced communication mechanism based on dynamic graph topology, which ascertains the core aspects of ``who" and ``what" in interactive process through three-stage operations: selection, generation, and aggregation. Concurrently, we incorporate local crowd perception to enrich agents' heuristic observation, thereby strengthening the model's guidance for effective actions via the integration of static spatial constraints and dynamic occupancy changes. To resolve extreme deadlock issues, we propose a region-based deadlock-breaking strategy that leverages expert guidance to implement efficient coordination within confined areas. Experimental results demonstrate that PC2P achieves superior performance compared to state-of-the-art distributed MAPF methods in varied environments. Ablation studies further confirm the effectiveness of each module for overall performance. △ Less

Submitted 5 January, 2026; originally announced January 2026.

Comments: 8 pages,7 figures,3 tables,Accepted to IROS 2025

SoliReward: Mitigating Susceptibility to Reward Hacking and Annotation Noise in Video Generation Reward Models

Authors: Jiesong Lian, Ruizhe Zhong, Zixiang Zhou, Xiaoyue Mi, Long Hu, Yuan Zhou, Qinglin Lu, Yixue Hao, Junchi Yan

Abstract: Post-training alignment of video generation models with human preferences is a critical goal. Developing effective Reward Models (RMs) for this process faces significant methodological hurdles. Current data collection paradigms, reliant on in-prompt pairwise annotations, suffer from labeling noise. Concurrently, the architectural design of VLM-based RMs, particularly their output mechanisms, remai… ▽ More Post-training alignment of video generation models with human preferences is a critical goal. Developing effective Reward Models (RMs) for this process faces significant methodological hurdles. Current data collection paradigms, reliant on in-prompt pairwise annotations, suffer from labeling noise. Concurrently, the architectural design of VLM-based RMs, particularly their output mechanisms, remains underexplored. Furthermore, RM is susceptible to reward hacking in post-training. To mitigate these limitations, we propose SoliReward, a systematic framework for video RM training. Our framework first sources high-quality, cost-efficient data via single-item binary annotations, then constructs preference pairs using a cross-prompt pairing strategy. Architecturally, we employ a Hierarchical Progressive Query Attention mechanism to enhance feature aggregation. Finally, we introduce a modified BT loss that explicitly accommodates win-tie scenarios. This approach regularizes the RM's score distribution for positive samples, providing more nuanced preference signals to alleviate over-focus on a small number of top-scoring samples. Our approach is validated on benchmarks evaluating physical plausibility, subject deformity, and semantic alignment, demonstrating improvements in direct RM evaluation metrics and in the efficacy of post-training on video generation models. Code and benchmark are available at https://github.com/lian700/SoliReward. △ Less

Submitted 16 March, 2026; v1 submitted 17 December, 2025; originally announced December 2025.

Comments: 16 pages, 9 figures

How Do Graph Signals Affect Recommendation: Unveiling the Mystery of Low and High-Frequency Graph Signals

Authors: Feng Liu, Hao Cang, Huanhuan Yuan, Jiaqing Fan, Yongjing Hao, Fuzhen Zhuang, Guanfeng Liu, Pengpeng Zhao

Abstract: Spectral graph neural networks (GNNs) are highly effective in modeling graph signals, with their success in recommendation often attributed to low-pass filtering. However, recent studies highlight the importance of high-frequency signals. The role of low-frequency and high-frequency graph signals in recommendation remains unclear. This paper aims to bridge this gap by investigating the influence o… ▽ More Spectral graph neural networks (GNNs) are highly effective in modeling graph signals, with their success in recommendation often attributed to low-pass filtering. However, recent studies highlight the importance of high-frequency signals. The role of low-frequency and high-frequency graph signals in recommendation remains unclear. This paper aims to bridge this gap by investigating the influence of graph signals on recommendation performance. We theoretically prove that the effects of low-frequency and high-frequency graph signals are equivalent in recommendation tasks, as both contribute by smoothing the similarities between user-item pairs. To leverage this insight, we propose a frequency signal scaler, a plug-and-play module that adjusts the graph signal filter function to fine-tune the smoothness between user-item pairs, making it compatible with any GNN model. Additionally, we identify and prove that graph embedding-based methods cannot fully capture the characteristics of graph signals. To address this limitation, a space flip method is introduced to restore the expressive power of graph embeddings. Remarkably, we demonstrate that either low-frequency or high-frequency graph signals alone are sufficient for effective recommendations. Extensive experiments on four public datasets validate the effectiveness of our proposed methods. Code is avaliable at https://github.com/mojosey/SimGCF. △ Less

Submitted 10 December, 2025; originally announced December 2025.

Building Audio-Visual Digital Twins with Smartphones

Authors: Zitong Lan, Yiwei Tang, Yuhan Wang, Haowen Lai, Yiduo Hao, Mingmin Zhao

Abstract: Digital twins today are almost entirely visual, overlooking acoustics-a core component of spatial realism and interaction. We introduce AV-Twin, the first practical system that constructs editable audio-visual digital twins using only commodity smartphones. AV-Twin combines mobile RIR capture and a visual-assisted acoustic field model to efficiently reconstruct room acoustics. It further recovers… ▽ More Digital twins today are almost entirely visual, overlooking acoustics-a core component of spatial realism and interaction. We introduce AV-Twin, the first practical system that constructs editable audio-visual digital twins using only commodity smartphones. AV-Twin combines mobile RIR capture and a visual-assisted acoustic field model to efficiently reconstruct room acoustics. It further recovers per-surface material properties through differentiable acoustic rendering, enabling users to modify materials, geometry, and layout while automatically updating both audio and visuals. Together, these capabilities establish a practical path toward fully modifiable audio-visual digital twins for real-world environments. △ Less

Submitted 11 December, 2025; originally announced December 2025.

Comments: Under Mobisys 2026 review, single blind

Meta Lattice: Model Space Redesign for Cost-Effective Industry-Scale Ads Recommendations

Authors: Liang Luo, Yuxin Chen, Zhengyu Zhang, Mengyue Hang, Andrew Gu, Buyun Zhang, Boyang Liu, Chen Chen, Chengze Fan, Dong Liang, Fan Yang, Feifan Gu, Huayu Li, Jade Nie, Jiayi Xu, Jiyan Yang, Jongsoo Park, Laming Chen, Longhao Jin, Qianru Li, Qin Huang, Shali Jiang, Shiwen Shen, Shuaiwen Wang, Sihan Zeng , et al. (17 additional authors not shown)

Abstract: The rapidly evolving landscape of products, surfaces, policies, and regulations poses significant challenges for deploying state-of-the-art recommendation models at industry scale, primarily due to data fragmentation across domains and escalating infrastructure costs that hinder sustained quality improvements. To address this challenge, we propose Lattice, a recommendation framework centered aro… ▽ More The rapidly evolving landscape of products, surfaces, policies, and regulations poses significant challenges for deploying state-of-the-art recommendation models at industry scale, primarily due to data fragmentation across domains and escalating infrastructure costs that hinder sustained quality improvements. To address this challenge, we propose Lattice, a recommendation framework centered around model space redesign that extends Multi-Domain, Multi-Objective (MDMO) learning beyond models and learning objectives. Lattice addresses these challenges through a comprehensive model space redesign that combines cross-domain knowledge sharing, data consolidation, model unification, distillation, and system optimizations to achieve significant improvements in both quality and cost-efficiency. Our deployment of Lattice at Meta has resulted in 10% revenue-driving top-line metrics gain, 11.5% user satisfaction improvement, 6% boost in conversion rate, with 20% capacity saving. △ Less

Submitted 14 December, 2025; v1 submitted 9 December, 2025; originally announced December 2025.

Comments: Accepted to KDD 2026

3DID: Direct 3D Inverse Design for Aerodynamics with Physics-Aware Optimization

Authors: Yuze Hao, Linchao Zhu, Yi Yang

Abstract: Inverse design aims to design the input variables of a physical system to optimize a specified objective function, typically formulated as a search or optimization problem. However, in 3D domains, the design space grows exponentially, rendering exhaustive grid-based searches infeasible. Recent advances in deep learning have accelerated inverse design by providing powerful generative priors and dif… ▽ More Inverse design aims to design the input variables of a physical system to optimize a specified objective function, typically formulated as a search or optimization problem. However, in 3D domains, the design space grows exponentially, rendering exhaustive grid-based searches infeasible. Recent advances in deep learning have accelerated inverse design by providing powerful generative priors and differentiable surrogate models. Nevertheless, current methods tend to approximate the 3D design space using 2D projections or fine-tune existing 3D shapes. These approaches sacrifice volumetric detail and constrain design exploration, preventing true 3D design from scratch. In this paper, we propose a 3D Inverse Design (3DID) framework that directly navigates the 3D design space by coupling a continuous latent representation with a physics-aware optimization strategy. We first learn a unified physics-geometry embedding that compactly captures shape and physical field data in a continuous latent space. Then, we introduce a two-stage strategy to perform physics-aware optimization. In the first stage, a gradient-guided diffusion sampler explores the global latent manifold. In the second stage, an objective-driven, topology-preserving refinement further sculpts each candidate toward the target objective. This enables 3DID to generate high-fidelity 3D geometries, outperforming existing methods in both solution quality and design versatility. △ Less

Submitted 6 December, 2025; originally announced December 2025.

Comments: Accepted at NeurIPS 2025

LoFA: Learning to Predict Personalized Priors for Fast Adaptation of Visual Generative Models

Authors: Yiming Hao, Mutian Xu, Chongjie Ye, Jie Qin, Shunlin Lu, Yipeng Qin, Xiaoguang Han

Abstract: Personalizing visual generative models to meet specific user needs has gained increasing attention, yet current methods like Low-Rank Adaptation (LoRA) remain impractical due to their demand for task-specific data and lengthy optimization. While a few hypernetwork-based approaches attempt to predict adaptation weights directly, they struggle to map fine-grained user prompts to complex LoRA distrib… ▽ More Personalizing visual generative models to meet specific user needs has gained increasing attention, yet current methods like Low-Rank Adaptation (LoRA) remain impractical due to their demand for task-specific data and lengthy optimization. While a few hypernetwork-based approaches attempt to predict adaptation weights directly, they struggle to map fine-grained user prompts to complex LoRA distributions, limiting their practical applicability. To bridge this gap, we propose LoFA, a general framework that efficiently predicts personalized priors for fast model adaptation. We first identify a key property of LoRA: structured distribution patterns emerge in the relative changes between LoRA and base model parameters. Building on this, we design a two-stage hypernetwork: first predicting relative distribution patterns that capture key adaptation regions, then using these to guide final LoRA weight prediction. Extensive experiments demonstrate that our method consistently predicts high-quality personalized priors within seconds, across multiple tasks and user prompts, even outperforming conventional LoRA that requires hours of processing. Project page: https://jaeger416.github.io/lofa/. △ Less

Submitted 9 December, 2025; originally announced December 2025.

Comments: Project page: https://jaeger416.github.io/lofa/

Time-Varying Formation Tracking Control of Wheeled Mobile Robots With Region Constraint: A Generalized Udwadia-Kalaba Framework

Authors: Yijie Kang, Yuqing Hao, Qingyun Wang, Guanrong Chen

Abstract: In this article, the time-varying formation tracking control of wheeled mobile robots with region constraint is investigated from a generalized Udwadia-Kalaba framework. The communication network is modeled as a directed and weighted graph that has a spanning tree with the leader being the root. By reformulating the time-varying formation tracking control objective as an equality constrained equat… ▽ More In this article, the time-varying formation tracking control of wheeled mobile robots with region constraint is investigated from a generalized Udwadia-Kalaba framework. The communication network is modeled as a directed and weighted graph that has a spanning tree with the leader being the root. By reformulating the time-varying formation tracking control objective as an equality constrained equation and transforming the region constraint by a diffeomorphism, the time-varying formation tracking controller with the region constraint is designed under the generalized Udwadia-Kalaba framework. Compared with the existing works on time-varying formation tracking control, the region constraint is taken into account in this paper, which ensures the safety of the robots. Finally, the feasibility of the proposed control strategy is illustrated through some numerical simulations. △ Less

Submitted 26 February, 2026; v1 submitted 7 December, 2025; originally announced December 2025.

Comments: 17 pages,9 figures

Hardware-Software Collaborative Computing of Photonic Spiking Reinforcement Learning for Robotic Continuous Control

Authors: Mengting Yu, Shuiying Xiang, Changjian Xie, Yonghang Chen, Haowen Zhao, Xingxing Guo, Yahui Zhang, Yanan Han, Yue Hao

Abstract: Robotic continuous control tasks impose stringent demands on the energy efficiency and latency of computing architectures due to their high-dimensional state spaces and real-time interaction requirements. Conventional electronic computing platforms face computational bottlenecks, whereas the fusion of photonic computing and spiking reinforcement learning (RL) offers a promising alternative. Here,… ▽ More Robotic continuous control tasks impose stringent demands on the energy efficiency and latency of computing architectures due to their high-dimensional state spaces and real-time interaction requirements. Conventional electronic computing platforms face computational bottlenecks, whereas the fusion of photonic computing and spiking reinforcement learning (RL) offers a promising alternative. Here, we propose a novel computing architecture based on photonic spiking RL, which integrates the Twin Delayed Deep Deterministic policy gradient (TD3) algorithm with spiking neural network (SNN). The proposed architecture employs an optical-electronic hybrid computing paradigm wherein a silicon photonic Mach-Zehnder interferometer (MZI) chip executes linear matrix computations, while nonlinear spiking activations are performed in the electronic domain. Experimental validation on the Pendulum-v1 and HalfCheetah-v2 benchmarks demonstrates the system capability for software-hardware co-inference, achieving a control policy reward of 5831 on HalfCheetah-v2, a 23.33% reduction in convergence steps, and an action deviation below 2.2%. Notably, this work represents the first application of a programmable MZI photonic computing chip to robotic continuous control tasks, attaining an energy efficiency of 1.39 TOPS/W and an ultralow computational latency of 120 ps. Such performance underscores the promise of photonic spiking RL for real-time decision-making in autonomous and industrial robotic systems. △ Less

Submitted 29 November, 2025; originally announced December 2025.

Guiding the Inner Eye: A Framework for Hierarchical and Flexible Visual Grounded Reasoning

Authors: Zhaoyang Wei, Wenchao Ding, Yanchao Hao, Xi Chen

Abstract: Models capable of "thinking with images" by dynamically grounding their reasoning in visual evidence represent a major leap in multimodal AI. However, replicating and advancing this ability is non-trivial, with current methods often trapped between the instability of end-to-end reinforcement learning (RL) and the rigidity of supervised fine-tuning (SFT). This leads to models that either struggle t… ▽ More Models capable of "thinking with images" by dynamically grounding their reasoning in visual evidence represent a major leap in multimodal AI. However, replicating and advancing this ability is non-trivial, with current methods often trapped between the instability of end-to-end reinforcement learning (RL) and the rigidity of supervised fine-tuning (SFT). This leads to models that either struggle to learn or lack the cognitive flexibility required for complex, real-world scenes. To navigate this dilemma, we introduce GRiP (Guided Reasoning and Perception), a novel two-stage training framework that cultivates robust and flexible visual grounded reasoning by explicitly guiding the model's perceptual focus and logical pathways. GRiP's core lies in its cognitive-enhanced RL stage, which features two key innovations: (1) a Salience-Weighted IoU Reward that incentivizes the model to prioritize the localization of mission-critical objects over trivial distractors, and (2) a Multi-Heuristic Reward that encourages cognitive flexibility by rewarding diverse yet logically valid reasoning pathways. Initialized from the Qwen2.5-VL-7B model, GRiP demonstrates significant performance gains across multiple challenging benchmarks. It achieves state-of-the-art results among open-source models on the highly challenging TreeBench and V* Bench, proving its effectiveness in complex visual reasoning. Our work demonstrates that moving beyond simplistic rewards and instead guiding models with cognitively-inspired signals for what to see and how to think is crucial for unlocking the next level of multimodal intelligence. The code will be made publicly available. △ Less

Submitted 27 November, 2025; originally announced November 2025.

Comments: 9pages