Showing 1–50 of 124 results for author: Kuang, Z

ShiJianBench: From Dialogue to Decision for Long-Horizon Evaluation of Investment Advisors

Authors: Jie Gong, Maowei Jiang, Zhiwei Liu, Yang Qiao, Wenxi Wu, Mengxi Xiao, Enze Zhang, Ziyan Kuang, Yankai Chen, Caishuang Huang, Meng Zhou, Xiku Du, Xue Liu, Guojun Xiong, Min Peng, Qianqian Xie, Sophia Ananiadou

Abstract: Conversational investment advisors influence not only what users know, but also how they make subsequent decisions as market conditions evolve. Existing evaluations primarily assess response quality or observed outcomes, leaving the long-horizon pathway from advisor language to investor behavior difficult to audit. We introduce ShiJianBench, an offline framework for evaluating conversational inves… ▽ More Conversational investment advisors influence not only what users know, but also how they make subsequent decisions as market conditions evolve. Existing evaluations primarily assess response quality or observed outcomes, leaving the long-horizon pathway from advisor language to investor behavior difficult to audit. We introduce ShiJianBench, an offline framework for evaluating conversational investment advisors through matched investor trajectories under fixed historical market feedback. At its core is a multi-agent investor simulator with explicit evolving state variables, motive-driven deliberation, long-term memory, and dialogue-grounded updates. The simulator is calibrated against aggregate behavioral patterns from 7,199 real users, and advisor policies are evaluated using separate investor-side, service-side, and content-side metrics under a hard compliance gate. Experiments on Chinese fund-market traces from 2021 to 2026 identify a stable leading group of LLM advisors that combines substantially stronger personalized content with competitive investor-side trajectory outcomes. These results reveal a systematic distinction between producing a high-quality response and delivering an effective long-horizon intervention, motivating trajectory-aware evaluation of conversational advisors. △ Less

Submitted 2 August, 2026; originally announced August 2026.

An AI Approach to Verified Production Cryptographic Libraries

Authors: Chuyue Sun, Su Fong, Zhiyi Kuang, Yizheng Jiao, Nina Narodytska, Haoze Wu, David L. Dill, Clark Barrett

Abstract: Cryptographic code is critical infrastructure that must be correct, yet formally verifying production libraries remains difficult. Existing language-model proof systems solve isolated obligations with specifications and premises already given, leaving production-library verification unresolved. We present CryptoProver, an AI-based system that synthesizes internal specifications and Verus-checked… ▽ More Cryptographic code is critical infrastructure that must be correct, yet formally verifying production libraries remains difficult. Existing language-model proof systems solve isolated obligations with specifications and premises already given, leaving production-library verification unresolved. We present CryptoProver, an AI-based system that synthesizes internal specifications and Verus-checked proofs from high-level API contracts. Without changing executable code, CryptoProver constructs a new independent proof of curve25519-dalek and verifies RustCrypto's previously unverified chacha20 implementation against an RFC 8439 specification. These cryptographic lineages underpin deployed systems including Signal and Shadowsocks; Signal has an estimated 218M global downloads. The independent, human-led curve25519-dalek verification was developed publicly over eight months by five main contributors. Given the API contracts and a fixed trusted library of field specifications, arithmetic facts, axioms, and vstd, CryptoProver synthesizes the internal specifications and proofs in 11.4 hours with USD 466.99 in recorded API cost. CryptoProver follows a trust-first design principle: mechanical gates reject specification weakening, invented axioms, and cross-module breakage, while isolation blocks reference proof retrieval, including from git history. △ Less

Submitted 1 August, 2026; originally announced August 2026.

Blood Pressure Estimation from PPG: A Comparative Study of Direct and ECG-Mediated Deep Learning Pipelines

Authors: Bo Wu, Haoling Wang, Zhuodiao Kuang, Kateryna Shapovalenko

Abstract: Continuous cuffless blood pressure (BP) monitoring is essential for connected health systems and wearable devices, enabling early detection, longitudinal tracking, and personalized management of cardiovascular disease. Many prior approaches attempt to estimate BP indirectly by reconstructing electrocardiography (ECG) from photoplethysmography (PPG), assuming ECG provides a stronger physiological l… ▽ More Continuous cuffless blood pressure (BP) monitoring is essential for connected health systems and wearable devices, enabling early detection, longitudinal tracking, and personalized management of cardiovascular disease. Many prior approaches attempt to estimate BP indirectly by reconstructing electrocardiography (ECG) from photoplethysmography (PPG), assuming ECG provides a stronger physiological link to BP. However, ECG sensing is less accessible in wearable settings and may introduce unnecessary complexity. In this work, we first perform a large-scale physiological correlation analysis on the MIMIC-III waveform database, revealing that PPG exhibits substantially stronger coupling with arterial blood pressure (ABP) ($|r|=0.247$, $p<0.001$) than ECG does ($r=0.018$, $p=0.187$), challenging the assumption that ECG provides a superior intermediate representation. Motivated by this insight, we conduct a systematic comparison between direct PPG-to-BP prediction and ECG-mediated pipelines using multiple state-of-the-art deep learning models. Across 1.74M segments from 3,127 patients, direct PPG-to-BP prediction achieves British Hypertension Society Grade A performance ($\mathrm{MAE}_{\mathrm{SBP}} = 4.82 mmHg$, $\mathrm{MAE}_{\mathrm{DBP}} = 4.31 mmHg$), outperforming all ECG-mediated approaches, which achieve only Grade B accuracy. Our findings suggest that accurate continuous BP monitoring can be achieved directly from wearable PPG signals, enabling simpler, more efficient pipelines for real-world connected health systems. △ Less

Submitted 25 July, 2026; originally announced July 2026.

AlphaOracle: Oracle bone script decipherment via human-workflow-inspired deep learning

Authors: Yuliang Liu, Haisu Guan, Pengjie Wang, Xinyu Wang, Jinpeng Wan, Kaile Zhang, Handong Zheng, Xingchen Liu, Zhebin Kuang, Huanxin Yang, Bang Li, Yongge Liu, Lianwen Jin, Xiang Bai

Abstract: Approximately 3,000 of the 4,500 oracle bone script (OBS) characters remain undeciphered due to fragmentary inscriptions and sparse evidence. Current AI approaches fail to replicate expert workflows that integrate form analysis, contextual semantics, and philological reasoning. We introduce AlphaOracle, a human-workflow-inspired framework that systematizes OBS decipherment using the largest digiti… ▽ More Approximately 3,000 of the 4,500 oracle bone script (OBS) characters remain undeciphered due to fragmentary inscriptions and sparse evidence. Current AI approaches fail to replicate expert workflows that integrate form analysis, contextual semantics, and philological reasoning. We introduce AlphaOracle, a human-workflow-inspired framework that systematizes OBS decipherment using the largest digitized corpus to date. Its multi-stage pipeline comprises: (i) rubbing parsing; (ii) radical-based morphological analysis with diachronic modeling; (iii) contextual retrieval with semantic alignment; and (iv) philological validation against classical sources. Each stage generates explicit, confidence-weighted evidence chains, culminating in interpretable reports for scholarly verification. Across multiple test characters, AlphaOracle's readings strongly agreed with expert interpretations. In a study of 86 domain specialists, it reduced analysis time by 64% and 79% of participants rated it highly useful. Notably, AlphaOracle resolves the character "Lao" as a toponymic or clan designation, offering concrete revisions to Shang administrative and social interpretations. These results suggest that computational methods aligned with philological practice can facilitate OBS research and provide a conceptual reference for studies of other undeciphered scripts. △ Less

Submitted 20 July, 2026; originally announced July 2026.

Comments: Accepted by The Innovation 2026

Journal ref: The Innovation 7(11), 101462, 2026

Empirical Grounding Improves the Realism of LLM Agents Simulating Human Behavior During Disruptions

Authors: Chen Xia, Zexi Kuang, Yuqing Hu

Abstract: Large language model (LLM) agents offer a generative approach to simulating human behavior under conditions that may have few or no direct historical analogues, a common challenge in disaster and infrastructure-disruption planning. However, this generative capacity creates a validity problem: individually plausible agent reasoning may fail to reproduce empirical population behavior. We evaluate wh… ▽ More Large language model (LLM) agents offer a generative approach to simulating human behavior under conditions that may have few or no direct historical analogues, a common challenge in disaster and infrastructure-disruption planning. However, this generative capacity creates a validity problem: individually plausible agent reasoning may fail to reproduce empirical population behavior. We evaluate whether empirical grounding improves the statistical realism of LLM-agent simulations during disruptions. Specifically, we develop an empirically grounded LLM-agent framework that embeds demographic profiles from the American Community Survey, baseline routines from the American Time Use Survey, and urban spatial context into agent initialization, memory, decision prompts, and activity execution. An independent household survey conducted during the July 2024 Philadelphia heatwave is reserved as an external validation benchmark. Compared with an ungrounded LLM-agent baseline, the grounded model improved reconstruction of normal daily routines, increasing mean correlation with empirical activity profiles from 0.528 to 0.912 and reducing mean squared error from 0.066 to 0.008. Under heatwave conditions, the grounded model better reproduced survey-derived activity profiles, increasing mean correlation from 0.349 to 0.836 and reducing mean squared error from 0.098 to 0.012. The grounded model captured 46.4% of observed heatwave response amplitude, compared with 20.6% for the ungrounded baseline. These findings show that empirical grounding can make LLM agents more statistically credible simulators of population behavior while revealing remaining gaps in modeling human adaptation during disruptions. △ Less

Submitted 19 July, 2026; originally announced July 2026.

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

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

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

Submitted 4 July, 2026; originally announced July 2026.

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

Equation Asymmetry: An Algebraic Framework for Unifying Secrecy and Covertness in Information-Theoretic Security

Authors: Wang Hao, Zhang Kuang

Abstract: This paper studies the algebraic structure underlying a broad class of information-theoretic security problems. We define the equation asymmetry degree (EAD) as $Φ= (n - r)/n$, where $n$ is the signal embedding dimension and $r$ is the effective rank of the adversary's observation matrix. This single parameter is shown to simultaneously govern both secrecy (measured by equivocation $H(M|Y_E)$) and… ▽ More This paper studies the algebraic structure underlying a broad class of information-theoretic security problems. We define the equation asymmetry degree (EAD) as $Φ= (n - r)/n$, where $n$ is the signal embedding dimension and $r$ is the effective rank of the adversary's observation matrix. This single parameter is shown to simultaneously govern both secrecy (measured by equivocation $H(M|Y_E)$) and covertness (measured by detection error probability $P_e$). On finite fields $\mathbb{F}_q$, we establish the equivocation lower bound $H(M|Y_E) = \min(k, n - r_E) \log q$ with exact probabilistic conditions (Theorem~1), the secrecy capacity $C_s = (n - r_E) \log q$ with complete achievability and converse proofs (Theorem~2), and a strong converse (Theorem~8). In the continuous Gaussian regime, we derive a differential-entropy equivocation bound (Lemma~1), the high-SNR secrecy capacity asymptotics (Lemma~2), and a 2-Wasserstein distance covertness condition $W_2 \approx \sqrt{r_W} \cdot P / (2Nσ) \to 0$ (Theorem~5'). The EAD-SDoF equivalence $d_s = n \cdot Φ$ is established (Theorem~7). Both $η_s$ and $η_c$ are shown to be monotone functions of $Φ$ (Theorem~6), with a Pearson correlation of $0.997$ in continuous-domain experiments. Seven existing security schemes -- matrix embedding, MIMO wiretap, secure network coding, FRFT multi-angle transmission, traffic steganography, group-key secure summation, and MDS secure summation -- are unified under the common form $C_s = (n - r) \log q$. Post-quantum security follows from the information-theoretic hardness of underdetermined linear systems (Theorem~9). All numerical experiments are reproducible with open-source code. △ Less

Submitted 8 June, 2026; originally announced June 2026.

Effective Multi-sensor Conditioning for Street-view Novel-view Synthesis

Authors: Zhengfei Kuang, Adam Sun, Liyuan Zhu, Tong Wu, Shengqu Cai, Jonathan Tremblay, Iro Armeni, Ehsan Adeli, Lior Yariv, Gordon Wetzstein

Abstract: Modern vehicle platforms are equipped with a rich sensor suite, including LiDAR, calibrated multi-camera rigs, and accurate ego-motion, that in principle offers strong signal for re-rendering a driving scene from novel viewpoints. A growing line of recent work leverages video diffusion models for this task, using their generative priors to synthesize plausible novel views from sparse vehicle obser… ▽ More Modern vehicle platforms are equipped with a rich sensor suite, including LiDAR, calibrated multi-camera rigs, and accurate ego-motion, that in principle offers strong signal for re-rendering a driving scene from novel viewpoints. A growing line of recent work leverages video diffusion models for this task, using their generative priors to synthesize plausible novel views from sparse vehicle observations. In practice, however, existing methods exploit only a fragment of this signal, and their quality tends to degrade as the target trajectory departs from the recorded driving path. We argue that this is fundamentally a multi-sensor fusion problem: sparse LiDAR reprojections supply accurate but incomplete metric geometry, surround-view reference imagery supplies dense appearance but no metric depth, and camera poses tie the two together across views. We introduce StreetNVS, a video diffusion framework that jointly conditions on all three signals through a Reference-Enhanced Camera Attention module based on a relative ray-level positional encoding. We develop a two-stage curriculum training strategy that gradually exposes the model to increasingly sparse LiDAR. On the Waymo Open Dataset, StreetNVS substantially outperforms state-of-the-art baselines under sparse LiDAR conditioning, matches methods that rely on 10-100 times denser point clouds. We further show capabilities of synthesizing coherent videos along extreme out-of-trajectory paths such as elevation, lane-shift, pullback, and rotation. Our website: https://streetnvs.github.io △ Less

Submitted 18 August, 2026; v1 submitted 31 May, 2026; originally announced June 2026.

UniT: Unified Geometry Learning with Group Autoregressive Transformer

Authors: Haotian Wang, Yusong Huang, Zhaonian Kuang, Hongliang Lu, Xinhu Zheng, Meng Yang, Gang Hua

Abstract: Recent feed-forward models have significantly advanced geometry perception for inferring dense 3D structure from sensor observations. However, its essential capabilities remain fragmented across multiple incompatible paradigms, including online perception, offline reconstruction, multi-modal integration, long-horizon scalability, and metric-scale estimation. We present UniT, a unified model built… ▽ More Recent feed-forward models have significantly advanced geometry perception for inferring dense 3D structure from sensor observations. However, its essential capabilities remain fragmented across multiple incompatible paradigms, including online perception, offline reconstruction, multi-modal integration, long-horizon scalability, and metric-scale estimation. We present UniT, a unified model built upon a novel Group Autoregressive Transformer, which reformulates these seemingly disparate capabilities within a single framework. The key idea is to treat groups of sensor observations as the basic autoregressive units and predict the corresponding point maps in an anchor-free and scale-adaptive manner. More specifically, diverse view configurations in both online and offline settings are naturally unified within a single group autoregression process. By varying the group size, online mode operates over multiple autoregressive steps with single-frame groups, whereas offline mode aggregates a multi-frame group in a single forward pass. Meanwhile, a queue-style KV caching mechanism ensures bounded autoregressive memory over long horizons. This is enabled by reducing long-range dependencies on early frames through anchor-free relational modeling, thereby allowing outdated memory to be discarded on the fly. To improve metric-scale generalization across scenes, a scale-adaptive geometry loss is further introduced within this framework. It couples relative geometric constraints with a partial absolute scale term, implicitly regularizing global scale and inducing a progressive transition from scale-invariant geometry to metric-scale solutions. Together with a dedicated modal attention module for integrating auxiliary modalities, UniT achieves state-of-the-art performance in unified geometry perception, as validated on ten benchmarks spanning seven representative tasks. △ Less

Submitted 20 May, 2026; originally announced May 2026.

Comments: Submitted to IEEE T-PAMI

GeoFlow: Enforcing Implicit Geometric Consistency in Video Generation

Authors: Jan Ackermann, Shengqu Cai, Boyang Deng, Zhengfei Kuang, Songyou Peng, Gordon Wetzstein

Abstract: Generating geometrically consistent videos remains an open challenge: text-to-video diffusion models trained on web-scale data treat geometry only implicitly, leading to object deformation, texture drift, and non-rigid backgrounds under camera motion. Existing solutions either improve consistency as a byproduct, apply only to static scenes or realign the latent space of the model completely. We in… ▽ More Generating geometrically consistent videos remains an open challenge: text-to-video diffusion models trained on web-scale data treat geometry only implicitly, leading to object deformation, texture drift, and non-rigid backgrounds under camera motion. Existing solutions either improve consistency as a byproduct, apply only to static scenes or realign the latent space of the model completely. We introduce a geometry-consistency reward that directly measures whether motion in a generated video is compatible with a coherent scene. Our key insight is that in physically consistent videos, background motion should be explainable by rigid camera-induced flow, while independently moving objects should preserve appearance identity along motion trajectories. We operationalize this using optical flow, depth--pose predictions, and feature-based correspondence to separate rigid and dynamic regions and evaluate their respective consistency. Integrating this reward with reinforcement fine-tuning transforms geometric consistency from an emergent property into an explicit optimization objective for video generators. The approach is model agnostic and applies to diverse dynamic scenes containing both camera and object motion. Experiments show substantial reductions in temporal geometric artifacts over strong baselines while preserving perceptual quality. Code and model weights are published. △ Less

Submitted 18 May, 2026; originally announced May 2026.

Comments: Project Page: https://geometryflow.github.io/

MemOVCD: Training-Free Open-Vocabulary Change Detection via Cross-Temporal Memory Reasoning and Global-Local Adaptive Rectification

Authors: Zuzheng Kuang, Honghao Chang, Boqiang Liang, Haoqian Wang, Lijun He, Fan Li, Haixia Bi

Abstract: Open-vocabulary change detection aims to identify semantic changes in bi-temporal remote sensing images without predefined categories. Recent methods combine foundation models such as SAM, DINO and CLIP, but typically process each timestamp independently or interact only at the final comparison stage. Such paradigms suffer from insufficient temporal coupling during semantic reasoning, which limits… ▽ More Open-vocabulary change detection aims to identify semantic changes in bi-temporal remote sensing images without predefined categories. Recent methods combine foundation models such as SAM, DINO and CLIP, but typically process each timestamp independently or interact only at the final comparison stage. Such paradigms suffer from insufficient temporal coupling during semantic reasoning, which limits their ability to distinguish genuine semantic changes from non-semantic appearance discrepancies. In addition, patch-dominant inference on high-resolution images often weakens global semantic continuity and produces fragmented change regions. To address these issues, we propose MemOVCD, a training-free open-vocabulary change detection framework based on cross-temporal memory reasoning and global-local adaptive rectification. Specifically, we reformulate bi-temporal change detection as a two-frame tracking problem and introduce weighted bidirectional propagation to aggregate semantic evidence from both temporal directions. To stabilize memory propagation across large temporal gaps, we construct histogram-aligned transition frames to smooth abrupt appearance changes. Moreover, a global-local adaptive rectification strategy adaptively fuses local and global-view predictions, improving spatial consistency while preserving fine-grained details. Experiments on five benchmarks demonstrate that MemOVCD achieves favorable performance on two change detection tasks, validating its effectiveness and generalization under diverse open-vocabulary settings. △ Less

Submitted 29 April, 2026; originally announced April 2026.

TexOCR: Advancing Document OCR Models for Compilable Page-to-LaTeX Reconstruction

Authors: Chengye Wang, Lin Fu, Zexi Kuang, Yilun Zhao

Abstract: Existing document OCR largely targets plain text or Markdown, discarding the structural and executable properties that make LaTeX essential for scientific publishing. We study page-level reconstruction of scientific PDFs into compilable LaTeX and introduce TexOCR-Bench, a benchmark, and TexOCR-Train, a large-scale training corpus, for this task. TexOCR-Bench features a multi-dimensional evaluation… ▽ More Existing document OCR largely targets plain text or Markdown, discarding the structural and executable properties that make LaTeX essential for scientific publishing. We study page-level reconstruction of scientific PDFs into compilable LaTeX and introduce TexOCR-Bench, a benchmark, and TexOCR-Train, a large-scale training corpus, for this task. TexOCR-Bench features a multi-dimensional evaluation suite that jointly assesses transcription fidelity, structural faithfulness, and end-to-end compilability. Leveraging TexOCR-Train, we train a 2B-parameter model, TexOCR, using supervised fine-tuning (SFT) and reinforcement learning (RL) with verifiable rewards derived from LaTeX unit tests that directly enforce compilability and referential integrity. Experiments across 21 frontier models on TexOCR-Bench show that existing systems frequently violate key document invariants, including consistent section structure, correct float placement, and valid label-reference links, which undermines compilation reliability and downstream usability. Our analysis further reveals that RL with verifiable rewards yields consistent improvements over SFT alone, particularly on structural and compilation metrics. △ Less

Submitted 23 April, 2026; originally announced April 2026.

Comments: Accepted by ACL 2026 Main

Subject-Aware Multi-Granularity Alignment for Zero-Shot EEG-to-Image Retrieval

Authors: Lin Jiang, Qingshan She, Jiale Xu, Haiqi Xu, Duanpo Wu, Zhenzhong Kuang

Abstract: Decoding visual content from electroencephalography (EEG) is important for understanding neural visual representations and developing non-invasive brain-computer interfaces. Existing approaches mainly improve EEG representation learning and cross-modal alignment while treating pretrained visual representations as fixed supervision targets. However, pretrained vision models organize information hie… ▽ More Decoding visual content from electroencephalography (EEG) is important for understanding neural visual representations and developing non-invasive brain-computer interfaces. Existing approaches mainly improve EEG representation learning and cross-modal alignment while treating pretrained visual representations as fixed supervision targets. However, pretrained vision models organize information hierarchically, with different depths encoding complementary structural and semantic information, and the visual granularity most compatible with EEG may vary across subjects. To address this issue, we propose Subject-Aware Multi-Granularity Alignment (SAMGA), which makes visual-target construction an explicit part of EEG-visual alignment. SAMGA constructs adaptive visual supervision from multiple intermediate representations and models EEG-compatible visual granularity through a global granularity prior with subject-dependent residual calibration, enabling subject-aware training and subject-agnostic inference. Based on the resulting adaptive target, a coarse-to-fine alignment strategy first organizes global cross-modal geometry and then refines instance-level retrieval discrimination. On THINGS-EEG, SAMGA improves Top-1 retrieval accuracy over the strongest competing method by 8.7 percentage points under intra-subject evaluation and 12.0 percentage points under leave-one-subject-out evaluation. These results support a broader view of neural-visual alignment, in which performance depends not only on how neural representations are mapped, but also on what visual representations define their supervision. △ Less

Submitted 13 August, 2026; v1 submitted 20 April, 2026; originally announced April 2026.

GeoLink: A 3D-Aware Framework Towards Better Generalization in Cross-View Geo-Localization

Authors: Hongyang Zhang, Yinhao Liu, Haitao Zhang, Zhongyi Wen, Zhenyu Kuang, Shuxian Liang, Xiansheng Hua

Abstract: Generalizable cross-view geo-localization aims to match the same location across views in unseen regions and conditions without GPS supervision. Its core difficulty lies in severe semantic inconsistency caused by viewpoint variation and poor generalization under domain shift. Existing methods mainly rely on 2D correspondence, but they are easily distracted by redundant shared information across vi… ▽ More Generalizable cross-view geo-localization aims to match the same location across views in unseen regions and conditions without GPS supervision. Its core difficulty lies in severe semantic inconsistency caused by viewpoint variation and poor generalization under domain shift. Existing methods mainly rely on 2D correspondence, but they are easily distracted by redundant shared information across views, leading to less transferable representations. To address this, we propose GeoLink, a 3D-aware semantic-consistent framework for Generalizable cross-view geo-localization. Specifically, we offline reconstruct scene point clouds from multi-view drone images using VGGT, providing stable structural priors. Based on these 3D anchors, we improve 2D representation learning in two complementary ways. A Geometric-aware Semantic Refinement module mitigates potentially redundant and view-biased dependencies in 2D features under 3D guidance. In addition, a Unified View Relation Distillation module transfers 3D structural relations to 2D features, improving cross-view alignment while preserving a 2D-only inference pipeline. Extensive experiments on multiple benchmarks show that GeoLink consistently outperforms state-of-the-art methods and achieves superior generalization across unseen domains and diverse weather environments. △ Less

Submitted 10 August, 2026; v1 submitted 14 April, 2026; originally announced April 2026.

SiMing-Bench: Evaluating Procedural Correctness from Continuous Interactions in Clinical Skill Videos

Authors: Xiyang Huang, Jiawei Lin, Keying Wu, Jiaxin Huang, Kailai Yang, Renxiong Wei, Cheng zeng, Jiayi Xiang, Ziyan Kuang, Min Peng, Qianqian Xie, Sophia Ananiadou

Abstract: Current video benchmarks for multimodal large language models (MLLMs) focus on event recognition, temporal ordering, and long-context recall, but overlook a harder capability required for expert procedural judgment: tracking how ongoing interactions update the procedural state and thereby determine the correctness of later actions. We introduce SiMing-Bench, the first benchmark for evaluating this… ▽ More Current video benchmarks for multimodal large language models (MLLMs) focus on event recognition, temporal ordering, and long-context recall, but overlook a harder capability required for expert procedural judgment: tracking how ongoing interactions update the procedural state and thereby determine the correctness of later actions. We introduce SiMing-Bench, the first benchmark for evaluating this capability from full-length clinical skill videos. It targets rubric-grounded process-level judgment of whether interaction-driven state updates preserve procedural correctness across an entire workflow. SiMing-Bench is instantiated with SiMing-Score, a physician-annotated dataset of real clinical skill examination videos spanning cardiopulmonary resuscitation, automated external defibrillator operation, and bag-mask ventilation, each paired with a standardized step-wise rubric and dual-expert labels. Across diverse open- and closed-source MLLMs, we observe consistently weak agreement with physician judgments. Moreover, weak performance on rubric-defined intermediate steps persists even when overall procedure-level correlation appears acceptable, suggesting that coarse global assessment substantially overestimates current models' procedural judgment ability. Additional analyses with binary step judgment and step-aligned clips indicate that the bottleneck is not merely fine-grained scoring or temporal localization, but modeling how continuous interactions update procedural state over time. △ Less

Submitted 10 April, 2026; originally announced April 2026.

Understanding the Challenges in Iterative Generative Optimization with LLMs

Authors: Allen Nie, Xavier Daull, Zhiyi Kuang, Abhinav Akkiraju, Anish Chaudhuri, Max Piasevoli, Ryan Rong, YuCheng Yuan, Prerit Choudhary, Shannon Xiao, Rasool Fakoor, Adith Swaminathan, Ching-An Cheng

Abstract: Generative optimization uses large language models (LLMs) to iteratively improve artifacts (such as code, workflows or prompts) using execution feedback. It is a promising approach to building self-improving agents, yet in practice remains brittle: despite active research, only 9% of surveyed agents used any automated optimization. We argue that this brittleness arises because, to set up a learnin… ▽ More Generative optimization uses large language models (LLMs) to iteratively improve artifacts (such as code, workflows or prompts) using execution feedback. It is a promising approach to building self-improving agents, yet in practice remains brittle: despite active research, only 9% of surveyed agents used any automated optimization. We argue that this brittleness arises because, to set up a learning loop, an engineer must make ``hidden'' design choices: What can the optimizer edit and what is the "right" learning evidence to provide at each update? We investigate three factors that affect most applications: the starting artifact, the credit horizon for execution traces, and batching trials and errors into learning evidence. Through case studies in MLAgentBench, Atari, and BigBench Extra Hard, we find that these design decisions can determine whether generative optimization succeeds, yet they are rarely made explicit in prior work. Different starting artifacts determine which solutions are reachable in MLAgentBench, truncated traces can still improve Atari agents, and larger minibatches do not monotonically improve generalization on BBEH. We conclude that the lack of a simple, universal way to set up learning loops across domains is a major hurdle for productionization and adoption. We provide practical guidance for making these choices. △ Less

Submitted 26 May, 2026; v1 submitted 25 March, 2026; originally announced March 2026.

Comments: 39 pages, 17 figures

Synergistic Perception and Generative Recomposition: A Multi-Agent Orchestration for Expert-Level Building Inspection

Authors: Hui Zhong, Yichun Gao, Luyan Liu, Xusen Guo, Zhaonian Kuang, Qiming Zhang, Xinhu Zheng

Abstract: Building facade defect inspection is fundamental to structural health monitoring and sustainable urban maintenance, yet it remains a formidable challenge due to extreme geometric variability, low contrast against complex backgrounds, and the inherent complexity of composite defects (e.g., cracks co-occurring with spalling). Such characteristics lead to severe pixel imbalance and feature ambiguity,… ▽ More Building facade defect inspection is fundamental to structural health monitoring and sustainable urban maintenance, yet it remains a formidable challenge due to extreme geometric variability, low contrast against complex backgrounds, and the inherent complexity of composite defects (e.g., cracks co-occurring with spalling). Such characteristics lead to severe pixel imbalance and feature ambiguity, which, coupled with the critical scarcity of high-quality pixel-level annotations, hinder the generalization of existing detection and segmentation models. To address gaps, we propose \textit{FacadeFixer}, a unified multi-agent framework that treats defect perception as a collaborative reasoning task rather than isolated recognition. Specifically,\textit{FacadeFixer} orchestrates specialized agents for detection and segmentation to handle multi-type defect interference, working in tandem with a generative agent to enable semantic recomposition. This process decouples intricate defects from noisy backgrounds and realistically synthesizes them onto diverse clean textures, generating high-fidelity augmented data with precise expert-level masks. To support this, we introduce a comprehensive multi-task dataset covering six primary facade categories with pixel-level annotations. Extensive experiments demonstrate that \textit{FacadeFixer} significantly outperforms state-of-the-art (SOTA) baselines. Specifically, it excels in capturing pixel-level structural anomalies and highlights generative synthesis as a robust solution to data scarcity in infrastructure inspection. Our code and dataset will be made publicly available. △ Less

Submitted 4 May, 2026; v1 submitted 20 March, 2026; originally announced March 2026.

Comments: We are withdrawing this article because we recently identified a major methodological error regarding the multi-agent orchestration setup described in Section 4.2. This issue significantly impacts the final conclusions drawn in the paper. We sincerely apologize for any confusion this may have caused

CamLit: Unified Video Diffusion with Explicit Camera and Lighting Control

Authors: Zhiyi Kuang, Chengan He, Egor Zakharov, Yuxuan Xue, Shunsuke Saito, Olivier Maury, Timur Bagautdinov, Youyi Zheng, Giljoo Nam

Abstract: We present CamLit, the first unified video diffusion model that jointly performs novel view synthesis (NVS) and relighting from a single input image. Given one reference image, a user-defined camera trajectory, and an environment map, CamLit synthesizes a video of the scene from new viewpoints under the specified illumination. Within a single generative process, our model produces temporally coher… ▽ More We present CamLit, the first unified video diffusion model that jointly performs novel view synthesis (NVS) and relighting from a single input image. Given one reference image, a user-defined camera trajectory, and an environment map, CamLit synthesizes a video of the scene from new viewpoints under the specified illumination. Within a single generative process, our model produces temporally coherent and spatially aligned outputs, including relit novel-view frames and corresponding albedo frames, enabling high-quality control of both camera pose and lighting. Qualitative and quantitative experiments demonstrate that CamLit achieves high-fidelity outputs on par with state-of-the-art methods in both novel view synthesis and relighting, without sacrificing visual quality in either task. We show that a single generative model can effectively integrate camera and lighting control, simplifying the video generation pipeline while maintaining competitive performance and consistent realism. △ Less

Submitted 15 March, 2026; originally announced March 2026.

Comments: 11 pages, 6 figures

RayD3D: Distilling Depth Knowledge Along the Ray for Robust Multi-View 3D Object Detection

Authors: Rui Ding, Zhaonian Kuang, Zongwei Zhou, Meng Yang, Xinhu Zheng, Gang Hua

Abstract: Multi-view 3D detection with bird's eye view (BEV) is crucial for autonomous driving and robotics, but its robustness in real-world is limited as it struggles to predict accurate depth values. A mainstream solution, cross-modal distillation, transfers depth information from LiDAR to camera models but also unintentionally transfers depth-irrelevant information (e.g. LiDAR density). To mitigate this… ▽ More Multi-view 3D detection with bird's eye view (BEV) is crucial for autonomous driving and robotics, but its robustness in real-world is limited as it struggles to predict accurate depth values. A mainstream solution, cross-modal distillation, transfers depth information from LiDAR to camera models but also unintentionally transfers depth-irrelevant information (e.g. LiDAR density). To mitigate this issue, we propose RayD3D, which transfers crucial depth knowledge along the ray: a line projecting from the camera to true location of an object. It is based on the fundamental imaging principle that predicted location of this object can only vary along this ray, which is finally determined by predicted depth value. Therefore, distilling along the ray enables more effective depth information transfer. More specifically, we design two ray-based distillation modules. Ray-based Contrastive Distillation (RCD) incorporates contrastive learning into distillation by sampling along the ray to learn how LiDAR accurately locates objects. Ray-based Weighted Distillation (RWD) adaptively adjusts distillation weight based on the ray to minimize the interference of depth-irrelevant information in LiDAR. For validation, we widely apply RayD3D into three representative types of BEV-based models, including BEVDet, BEVDepth4D, and BEVFormer. Our method is trained on clean NuScenes, and tested on both clean NuScenes and RoboBEV with a variety types of data corruptions. Our method significantly improves the robustness of all the three base models in all scenarios without increasing inference costs, and achieves the best when compared to recently released multi-view and distillation models. △ Less

Submitted 8 March, 2026; originally announced March 2026.

Multi-Modal Decouple and Recouple Network for Robust 3D Object Detection

Authors: Rui Ding, Zhaonian Kuang, Yuzhe Ji, Meng Yang, Xinhu Zheng, Gang Hua

Abstract: Multi-modal 3D object detection with bird's eye view (BEV) has achieved desired advances on benchmarks. Nonetheless, the accuracy may drop significantly in the real world due to data corruption such as sensor configurations for LiDAR and scene conditions for camera. One design bottleneck of previous models resides in the tightly coupling of multi-modal BEV features during fusion, which may degrade… ▽ More Multi-modal 3D object detection with bird's eye view (BEV) has achieved desired advances on benchmarks. Nonetheless, the accuracy may drop significantly in the real world due to data corruption such as sensor configurations for LiDAR and scene conditions for camera. One design bottleneck of previous models resides in the tightly coupling of multi-modal BEV features during fusion, which may degrade the overall system performance if one modality or both is corrupted. To mitigate, we propose a Multi-Modal Decouple and Recouple Network for robust 3D object detection under data corruption. Different modalities commonly share some high-level invariant features. We observe that these invariant features across modalities do not always fail simultaneously, because different types of data corruption affect each modality in distinct ways.These invariant features can be recovered across modalities for robust fusion under data corruption.To this end, we explicitly decouple Camera/LiDAR BEV features into modality-invariant and modality-specific parts. It allows invariant features to compensate each other while mitigates the negative impact of a corrupted modality on the other.We then recouple these features into three experts to handle different types of data corruption, respectively, i.e., LiDAR, camera, and both.For each expert, we use modality-invariant features as robust information, while modality-specific features serve as a complement.Finally, we adaptively fuse the three experts to exact robust features for 3D object detection. For validation, we collect a benchmark with a large quantity of data corruption for LiDAR, camera, and both based on nuScenes. Our model is trained on clean nuScenes and tested on all types of data corruption. Our model consistently achieves the best accuracy on both corrupted and clean data compared to recent models. △ Less

Submitted 8 March, 2026; originally announced March 2026.

CoIn3D: Revisiting Configuration-Invariant Multi-Camera 3D Object Detection

Authors: Zhaonian Kuang, Rui Ding, Haotian Wang, Xinhu Zheng, Meng Yang, Gang Hua

Abstract: Multi-camera 3D object detection (MC3D) has attracted increasing attention with the growing deployment of multi-sensor physical agents, such as robots and autonomous vehicles. However, MC3D models still struggle to generalize to unseen platforms with new multi-camera configurations. Current solutions simply employ a meta-camera for unified representation but lack comprehensive consideration. In th… ▽ More Multi-camera 3D object detection (MC3D) has attracted increasing attention with the growing deployment of multi-sensor physical agents, such as robots and autonomous vehicles. However, MC3D models still struggle to generalize to unseen platforms with new multi-camera configurations. Current solutions simply employ a meta-camera for unified representation but lack comprehensive consideration. In this paper, we revisit this issue and identify that the devil lies in spatial prior discrepancies across source and target configurations, including different intrinsics, extrinsics, and array layouts. To address this, we propose CoIn3D, a generalizable MC3D framework that enables strong transferability from source configurations to unseen target ones. CoIn3D explicitly incorporates all identified spatial priors into both feature embedding and image observation through spatial-aware feature modulation (SFM) and camera-aware data augmentation (CDA), respectively. SFM enriches feature space by integrating four spatial representations, such as focal length, ground depth, ground gradient, and Plücker coordinate. CDA improves observation diversity under various configurations via a training-free dynamic novel-view image synthesis scheme. Extensive experiments demonstrate that CoIn3D achieves strong cross-configuration performance on landmark datasets such as NuScenes, Waymo, and Lyft, under three dominant MC3D paradigms represented by BEVDepth, BEVFormer, and PETR. △ Less

Submitted 26 March, 2026; v1 submitted 5 March, 2026; originally announced March 2026.

Comments: Accepted to CVPR 2026 main track

FedAFD: Multimodal Federated Learning via Adversarial Fusion and Distillation

Authors: Min Tan, Junchao Ma, Yinfu Feng, Jiajun Ding, Wenwen Pan, Tingting Han, Qian Zheng, Zhenzhong Kuang, Zhou Yu

Abstract: Multimodal Federated Learning (MFL) enables clients with heterogeneous data modalities to collaboratively train models without sharing raw data, offering a privacy-preserving framework that leverages complementary cross-modal information. However, existing methods often overlook personalized client performance and struggle with modality/task discrepancies, as well as model heterogeneity. To addres… ▽ More Multimodal Federated Learning (MFL) enables clients with heterogeneous data modalities to collaboratively train models without sharing raw data, offering a privacy-preserving framework that leverages complementary cross-modal information. However, existing methods often overlook personalized client performance and struggle with modality/task discrepancies, as well as model heterogeneity. To address these challenges, we propose FedAFD, a unified MFL framework that enhances client and server learning. On the client side, we introduce a bi-level adversarial alignment strategy to align local and global representations within and across modalities, mitigating modality and task gaps. We further design a granularity-aware fusion module to integrate global knowledge into the personalized features adaptively. On the server side, to handle model heterogeneity, we propose a similarity-guided ensemble distillation mechanism that aggregates client representations on shared public data based on feature similarity and distills the fused knowledge into the global model. Extensive experiments conducted under both IID and non-IID settings demonstrate that FedAFD achieves superior performance and efficiency for both the client and the server. △ Less

Submitted 5 March, 2026; originally announced March 2026.

Comments: Accepted by CVPR 2026

An LLM-Assisted Toolkit for Inspectable Multimodal Emotion Data Annotation

Authors: Zheyuan Kuang, Weiwei Jiang, Nicholas Koemel, Matthew Ahmadi, Emmanuel Stamatakis, Benjamin Tag, Anusha Withana, Zhanna Sarsenbayeva

Abstract: Multimodal Emotion Recognition (MER) increasingly depends on fine grained, evidence grounded annotations, yet inspection and label construction are hard to scale when cues are dynamic and misaligned across modalities. We present an LLM-assisted toolkit that supports multimodal emotion data annotation through an inspectable, event centered workflow. The toolkit preprocesses and aligns heterogeneous… ▽ More Multimodal Emotion Recognition (MER) increasingly depends on fine grained, evidence grounded annotations, yet inspection and label construction are hard to scale when cues are dynamic and misaligned across modalities. We present an LLM-assisted toolkit that supports multimodal emotion data annotation through an inspectable, event centered workflow. The toolkit preprocesses and aligns heterogeneous recordings, visualizes all modalities on an interactive shared timeline, and renders structured signals as video tracks for cross modal consistency checks. It then detects candidate events and packages synchronized keyframes and time windows as event packets with traceable pointers to the source data. Finally, the toolkit integrates an LLM with modality specific tools and prompt templates to draft structured annotations for analyst verification and editing. We demonstrate the workflow on multimodal VR emotion recordings with representative examples. △ Less

Submitted 2 March, 2026; originally announced March 2026.

Comments: 5 pages, 1 figure

Understanding the Effects of Interaction on Emotional Experiences in VR

Authors: Zheyuan Kuang, Tinghui Li, Weiwei Jiang, Sven Mayer, Flora Salim, Benjamin Tag, Anusha Withana, Zhanna Sarsenbayeva

Abstract: Virtual reality has been effectively used for eliciting emotions, yet most research focuses on the intensity of affective responses rather than on how interaction influences those experiences. To address this gap, we advance a validated VR emotion-elicitation dataset through two key extensions. First, we add a new high-arousal, high-valence scene and validate its effectiveness in a within-subject… ▽ More Virtual reality has been effectively used for eliciting emotions, yet most research focuses on the intensity of affective responses rather than on how interaction influences those experiences. To address this gap, we advance a validated VR emotion-elicitation dataset through two key extensions. First, we add a new high-arousal, high-valence scene and validate its effectiveness in a within-subject study (N=24). Second, we incorporate interactive elements into each scene, creating both interactive and non-interactive versions to examine the impact of interaction on emotional responses. We evaluate interaction through a multimodal approach combining subjective ratings and physiological signals to capture both conscious and unconscious affective responses. Our evaluation study (N=84) shows that interaction not only amplifies emotions but modulates them in context, supporting coping in negative scenes and enhancing enjoyment in positive scenes. These findings highlight the potential of scene-tailored interaction for different applications, where regulating emotions is as important as eliciting them. △ Less

Submitted 2 March, 2026; originally announced March 2026.

Comments: 20 pages, 17 figures, to be published in the Proceedings of the 2026 ACM CHI Conference on Human Factors in Computing Systems

SR3R: Rethinking Super-Resolution 3D Reconstruction With Feed-Forward Gaussian Splatting

Authors: Xiang Feng, Xiangbo Wang, Tieshi Zhong, Chengkai Wang, Yiting Zhao, Tianxiang Xu, Zhenzhong Kuang, Feiwei Qin, Xuefei Yin, Yanming Zhu

Abstract: 3D super-resolution (3DSR) aims to reconstruct high-resolution (HR) 3D scenes from low-resolution (LR) multi-view images. Existing methods rely on dense LR inputs and per-scene optimization, which restricts the high-frequency priors for constructing HR 3D Gaussian Splatting (3DGS) to those inherited from pretrained 2D super-resolution (2DSR) models. This severely limits reconstruction fidelity, cr… ▽ More 3D super-resolution (3DSR) aims to reconstruct high-resolution (HR) 3D scenes from low-resolution (LR) multi-view images. Existing methods rely on dense LR inputs and per-scene optimization, which restricts the high-frequency priors for constructing HR 3D Gaussian Splatting (3DGS) to those inherited from pretrained 2D super-resolution (2DSR) models. This severely limits reconstruction fidelity, cross-scene generalization, and real-time usability. We propose to reformulate 3DSR as a direct feed-forward mapping from sparse LR views to HR 3DGS representations, enabling the model to autonomously learn 3D-specific high-frequency geometry and appearance from large-scale, multi-scene data. This fundamentally changes how 3DSR acquires high-frequency knowledge and enables robust generalization to unseen scenes. Specifically, we introduce SR3R, a feed-forward framework that directly predicts HR 3DGS representations from sparse LR views via the learned mapping network. To further enhance reconstruction fidelity, we introduce Gaussian offset learning and feature refinement, which stabilize reconstruction and sharpen high-frequency details. SR3R is plug-and-play and can be paired with any feed-forward 3DGS reconstruction backbone: the backbone provides an LR 3DGS scaffold, and SR3R upscales it to an HR 3DGS. Extensive experiments across three 3D benchmarks demonstrate that SR3R surpasses state-of-the-art (SOTA) 3DSR methods and achieves strong zero-shot generalization, even outperforming SOTA per-scene optimization methods on unseen scenes. △ Less

Submitted 27 February, 2026; originally announced February 2026.

Comments: CVPR 2026

Object-Scene-Camera Decomposition and Recomposition for Data-Efficient Monocular 3D Object Detection

Authors: Zhaonian Kuang, Rui Ding, Meng Yang, Xinhu Zheng, Gang Hua

Abstract: Monocular 3D object detection (M3OD) is intrinsically ill-posed, hence training a high-performance deep learning based M3OD model requires a humongous amount of labeled data with complicated visual variation from diverse scenes, variety of objects and camera poses.However, we observe that, due to strong human bias, the three independent entities, i.e., object, scene, and camera pose, are always ti… ▽ More Monocular 3D object detection (M3OD) is intrinsically ill-posed, hence training a high-performance deep learning based M3OD model requires a humongous amount of labeled data with complicated visual variation from diverse scenes, variety of objects and camera poses.However, we observe that, due to strong human bias, the three independent entities, i.e., object, scene, and camera pose, are always tightly entangled when an image is captured to construct training data. More specifically, specific 3D objects are always captured in particular scenes with fixed camera poses, and hence lacks necessary diversity. Such tight entanglement induces the challenging issues of insufficient utilization and overfitting to uniform training data. To mitigate this, we propose an online object-scene-camera decomposition and recomposition data manipulation scheme to more efficiently exploit the training data. We first fully decompose training images into textured 3D object point models and background scenes in an efficient computation and storage manner. We then continuously recompose new training images in each epoch by inserting the 3D objects into the freespace of the background scenes, and rendering them with perturbed camera poses from textured 3D point representation. In this way, the refreshed training data in all epochs can cover the full spectrum of independent object, scene, and camera pose combinations. This scheme can serve as a plug-and-play component to boost M3OD models, working flexibly with both fully and sparsely supervised settings. In the sparsely-supervised setting, objects closest to the ego-camera for all instances are sparsely annotated. We then can flexibly increase the annotated objects to control annotation cost. For validation, our method is widely applied to five representative M3OD models and evaluated on both the KITTI and the more complicated Waymo datasets. △ Less

Submitted 24 February, 2026; originally announced February 2026.

Comments: IJCV

Journal ref: Int J Comput Vis 134, 155 (2026)

MEnvAgent: Scalable Polyglot Environment Construction for Verifiable Software Engineering

Authors: Chuanzhe Guo, Jingjing Wu, Sijun He, Yang Chen, Zhaoqi Kuang, Shilong Fan, Bingjin Chen, Siqi Bao, Jing Liu, Hua Wu, Qingfu Zhu, Wanxiang Che, Haifeng Wang

Abstract: The evolution of Large Language Model (LLM) agents for software engineering (SWE) is constrained by the scarcity of verifiable datasets, a bottleneck stemming from the complexity of constructing executable environments across diverse languages. To address this, we introduce MEnvAgent, a Multi-language framework for automated Environment construction that facilitates scalable generation of verifiab… ▽ More The evolution of Large Language Model (LLM) agents for software engineering (SWE) is constrained by the scarcity of verifiable datasets, a bottleneck stemming from the complexity of constructing executable environments across diverse languages. To address this, we introduce MEnvAgent, a Multi-language framework for automated Environment construction that facilitates scalable generation of verifiable task instances. MEnvAgent employs a multi-agent Planning-Execution-Verification architecture to autonomously resolve construction failures and integrates a novel Environment Reuse Mechanism that reduces computational overhead by incrementally patching historical environments. Evaluations on MEnvBench, a new benchmark comprising 1,000 tasks across 10 languages, demonstrate that MEnvAgent outperforms baselines, improving Fail-to-Pass (F2P) rates by 8.6% while reducing time costs by 43%. Additionally, we demonstrate the utility of MEnvAgent by constructing MEnvData-SWE, the largest open-source polyglot dataset of realistic verifiable Docker environments to date, alongside solution trajectories that enable consistent performance gains on SWE tasks across a wide range of models. Our code, benchmark, and dataset are available at https://github.com/ernie-research/MEnvAgent. △ Less

Submitted 6 June, 2026; v1 submitted 30 January, 2026; originally announced January 2026.

Comments: Accepted as a Spotlight Paper at ICML 2026

Envision: Embodied Visual Planning via Goal-Imagery Video Diffusion

Authors: Yuming Gu, Yizhi Wang, Yining Hong, Yipeng Gao, Hao Jiang, Angtian Wang, Bo Liu, Nathaniel S. Dennler, Zhengfei Kuang, Hao Li, Gordon Wetzstein, Chongyang Ma

Abstract: Embodied visual planning aims to enable manipulation tasks by imagining how a scene evolves toward a desired goal and using the imagined trajectories to guide actions. Video diffusion models, through their image-to-video generation capability, provide a promising foundation for such visual imagination. However, existing approaches are largely forward predictive, generating trajectories conditioned… ▽ More Embodied visual planning aims to enable manipulation tasks by imagining how a scene evolves toward a desired goal and using the imagined trajectories to guide actions. Video diffusion models, through their image-to-video generation capability, provide a promising foundation for such visual imagination. However, existing approaches are largely forward predictive, generating trajectories conditioned on the initial observation without explicit goal modeling, thus often leading to spatial drift and goal misalignment. To address these challenges, we propose Envision, a diffusion-based framework that performs visual planning for embodied agents. By explicitly constraining the generation with a goal image, our method enforces physical plausibility and goal consistency throughout the generated trajectory. Specifically, Envision operates in two stages. First, a Goal Imagery Model identifies task-relevant regions, performs region-aware cross attention between the scene and the instruction, and synthesizes a coherent goal image that captures the desired outcome. Then, an Env-Goal Video Model, built upon a first-and-last-frame-conditioned video diffusion model (FL2V), interpolates between the initial observation and the goal image, producing smooth and physically plausible video trajectories that connect the start and goal states. Experiments on object manipulation and image editing benchmarks demonstrate that Envision achieves superior goal alignment, spatial consistency, and object preservation compared to baselines. The resulting visual plans can directly support downstream robotic planning and control, providing reliable guidance for embodied agents. △ Less

Submitted 27 December, 2025; originally announced December 2025.

Comments: Page: https://envision-paper.github.io

VULCAN: Tool-Augmented Multi Agents for Iterative 3D Object Arrangement

Authors: Zhengfei Kuang, Rui Lin, Long Zhao, Gordon Wetzstein, Saining Xie, Sanghyun Woo

Abstract: Despite the remarkable progress of Multimodal Large Language Models (MLLMs) in 2D vision-language tasks, their application to complex 3D scene manipulation remains underexplored. In this paper, we bridge this critical gap by tackling three key challenges in 3D object arrangement task using MLLMs. First, to address the weak visual grounding of MLLMs, which struggle to link programmatic edits with p… ▽ More Despite the remarkable progress of Multimodal Large Language Models (MLLMs) in 2D vision-language tasks, their application to complex 3D scene manipulation remains underexplored. In this paper, we bridge this critical gap by tackling three key challenges in 3D object arrangement task using MLLMs. First, to address the weak visual grounding of MLLMs, which struggle to link programmatic edits with precise 3D outcomes, we introduce an MCP-based API. This shifts the interaction from brittle raw code manipulation to more robust, function-level updates. Second, we augment the MLLM's 3D scene understanding with a suite of specialized visual tools to analyze scene state, gather spatial information, and validate action outcomes. This perceptual feedback loop is critical for closing the gap between language-based updates and precise 3D-aware manipulation. Third, to manage the iterative, error-prone updates, we propose a collaborative multi-agent framework with designated roles for planning, execution, and verification. This decomposition allows the system to robustly handle multi-step instructions and recover from intermediate errors. We demonstrate the effectiveness of our approach on a diverse set of 25 complex object arrangement tasks, where it significantly outperforms existing baselines. Website: vulcan-3d.github.io △ Less

Submitted 6 January, 2026; v1 submitted 26 December, 2025; originally announced December 2025.

Optimal 3D Directional WPT Charging via UAV for 3D Wireless Rechargeable Sensor Networks

Authors: Zhenguo Gao, Hui Li, Yiqin Chen, Qingyu Gao, Zhufang Kuang, Shih-Hau Fang, Hsiao-Chun Wu

Abstract: The high mobility and flexible deployment capability of UAVs make them an impressive option for charging nodes in Wireless Rechargeable Sensor Networks (WRSNs) using Directional Wireless Power Transfer (WPT) technology. However, existing studies largely focus on 2D-WRSNs, lacking designs catering to real 3D-WRSNs. The spatial distribution characteristics of nodes in a 3D-WRSN further increase the… ▽ More The high mobility and flexible deployment capability of UAVs make them an impressive option for charging nodes in Wireless Rechargeable Sensor Networks (WRSNs) using Directional Wireless Power Transfer (WPT) technology. However, existing studies largely focus on 2D-WRSNs, lacking designs catering to real 3D-WRSNs. The spatial distribution characteristics of nodes in a 3D-WRSN further increase the complexity of the charging scheduling task, thus requiring a systematic framework to solve this problem. In this paper, we investigated the Directional UAV Charging Scheduling problem for 3D-WRSNs (DCS-3D) and established its NP-hard property, and then proposed a three-step framework named as directional charging scheduling algorithm using Functional Equivalent (FuncEqv) direction set and Lin-Kernighan heuristic (LKH) for 3D-WRSNs (FELKH-3D) to solve it. In FELKH-3D, the challenge of infinite charging direction space is solved by designing an algorithm generating a minimum-size direction set guaranteed to be FuncEqv to the infinite set of whole sphere surface, and the optimaility of the method was proved.To determine the optimal charging tour for the UAV, the LKH algorithm is employed.Simulation experiments demonstrated the superiority of FELKH-3D over other classical algorithms. △ Less

Submitted 22 December, 2025; originally announced December 2025.

Commercial Vehicle Braking Optimization: A Robust SIFT-Trajectory Approach

Authors: Zhe Li, Kun Cheng, Hanyue Mo, Jintao Lu, Ziwen Kuang, Jianwen Ye, Lixu Xu, Xinya Meng, Jiahui Zhao, Shengda Ji, Shuyuan Liu, Mengyu Wang

Abstract: A vision-based trajectory analysis solution is proposed to address the "zero-speed braking" issue caused by inaccurate Controller Area Network (CAN) signals in commercial vehicle Automatic Emergency Braking (AEB) systems during low-speed operation. The algorithm utilizes the NVIDIA Jetson AGX Xavier platform to process sequential video frames from a blind spot camera, employing self-adaptive Contr… ▽ More A vision-based trajectory analysis solution is proposed to address the "zero-speed braking" issue caused by inaccurate Controller Area Network (CAN) signals in commercial vehicle Automatic Emergency Braking (AEB) systems during low-speed operation. The algorithm utilizes the NVIDIA Jetson AGX Xavier platform to process sequential video frames from a blind spot camera, employing self-adaptive Contrast Limited Adaptive Histogram Equalization (CLAHE)-enhanced Scale-Invariant Feature Transform (SIFT) feature extraction and K-Nearest Neighbors (KNN)-Random Sample Consensus (RANSAC) matching. This allows for precise classification of the vehicle's motion state (static, vibration, moving). Key innovations include 1) multiframe trajectory displacement statistics (5-frame sliding window), 2) a dual-threshold state decision matrix, and 3) OBD-II driven dynamic Region of Interest (ROI) configuration. The system effectively suppresses environmental interference and false detection of dynamic objects, directly addressing the challenge of low-speed false activation in commercial vehicle safety systems. Evaluation in a real-world dataset (32,454 video segments from 1,852 vehicles) demonstrates an F1-score of 99.96% for static detection, 97.78% for moving state recognition, and a processing delay of 14.2 milliseconds (resolution 704x576). The deployment on-site shows an 89% reduction in false braking events, a 100% success rate in emergency braking, and a fault rate below 5%. △ Less

Submitted 21 December, 2025; originally announced December 2025.

Comments: 5 figures,16 pages

MiraMind: Benchmarking Reliable Mental Health Reasoning beyond Answer Accuracy

Authors: Mengxi Xiao, Kailai Yang, Pengde Zhao, Enze Zhang, Ziyan Kuang, Zhiwei Liu, Weiguang Han, Shu Liao, Lianting Huang, Guojun Xiong, Victor Gutierrez Basulto, Jinpeng Hu, Min Peng, Qianqian Xie, Sophia Ananiadou

Abstract: Mental-health reasoning with large language models (LLMs) is an evidence-constrained judgment problem: models must transform limited, subjective, and often ambiguous evidence into interpretations, decisions, or claims whose specificity, certainty, severity, and actionability remain warranted. Existing benchmarks mainly evaluate specific clinical roles or final answers, leaving the reliability of e… ▽ More Mental-health reasoning with large language models (LLMs) is an evidence-constrained judgment problem: models must transform limited, subjective, and often ambiguous evidence into interpretations, decisions, or claims whose specificity, certainty, severity, and actionability remain warranted. Existing benchmarks mainly evaluate specific clinical roles or final answers, leaving the reliability of explicit reasoning trajectories under-specified. We introduce MiraMind, a unified benchmark spanning six task families and 13 datasets across appraisal, diagnosis, intervention, abstraction, and verification. MiraMind evaluates both task outcomes and the trajectories connecting evidence to judgment through usability, logical structure, and informational contribution. Evaluating 20 LLMs reveals a restraint gap, in which the specificity or certainty of model judgments exceeds what limited evidence supports. We further train Mindora, an 8B model that targets evidence-to-judgment transitions through hard-case supervision, structured trajectory rewriting, and consistency-aware optimization. Mindora achieves the best average rank on MiraMind, improves over its backbone across all six task families, and produces more balanced reasoning trajectories. These results show that MiraMind can expose shared restraint failures and evaluate whether targeted post-training improves evidence-constrained mental-health reasoning. △ Less

Submitted 9 August, 2026; v1 submitted 10 December, 2025; originally announced December 2025.

ACM Class: I.2.7

BulletTime: Decoupled Control of Time and Camera Pose for Video Generation

Authors: Yiming Wang, Qihang Zhang, Shengqu Cai, Tong Wu, Jan Ackermann, Zhengfei Kuang, Yang Zheng, Frano Rajič, Siyu Tang, Gordon Wetzstein

Abstract: Emerging video diffusion models achieve high visual fidelity but fundamentally couple scene dynamics with camera motion, limiting their ability to provide precise spatial and temporal control. We introduce a 4D-controllable video diffusion framework that explicitly decouples scene dynamics from camera pose, enabling fine-grained manipulation of both scene dynamics and camera viewpoint. Our framewo… ▽ More Emerging video diffusion models achieve high visual fidelity but fundamentally couple scene dynamics with camera motion, limiting their ability to provide precise spatial and temporal control. We introduce a 4D-controllable video diffusion framework that explicitly decouples scene dynamics from camera pose, enabling fine-grained manipulation of both scene dynamics and camera viewpoint. Our framework takes continuous world-time sequences and camera trajectories as conditioning inputs, injecting them into the video diffusion model through a 4D positional encoding in the attention layer and adaptive normalizations for feature modulation. To train this model, we curate a unique dataset in which temporal and camera variations are independently parameterized; this dataset will be made public. Experiments show that our model achieves robust real-world 4D control across diverse timing patterns and camera trajectories, while preserving high generation quality and outperforming prior work in controllability. See our website for video results: https://19reborn.github.io/Bullet4D/ △ Less

Submitted 4 December, 2025; originally announced December 2025.

Comments: Project Page: https://19reborn.github.io/Bullet4D/

IE-SRGS: An Internal-External Knowledge Fusion Framework for High-Fidelity 3D Gaussian Splatting Super-Resolution

Authors: Xiang Feng, Tieshi Zhong, Shuo Chang, Weiliu Wang, Chengkai Wang, Yifei Chen, Yuhe Wang, Zhenzhong Kuang, Xuefei Yin, Yanming Zhu

Abstract: Reconstructing high-resolution (HR) 3D Gaussian Splatting (3DGS) models from low-resolution (LR) inputs remains challenging due to the lack of fine-grained textures and geometry. Existing methods typically rely on pre-trained 2D super-resolution (2DSR) models to enhance textures, but suffer from 3D Gaussian ambiguity arising from cross-view inconsistencies and domain gaps inherent in 2DSR models.… ▽ More Reconstructing high-resolution (HR) 3D Gaussian Splatting (3DGS) models from low-resolution (LR) inputs remains challenging due to the lack of fine-grained textures and geometry. Existing methods typically rely on pre-trained 2D super-resolution (2DSR) models to enhance textures, but suffer from 3D Gaussian ambiguity arising from cross-view inconsistencies and domain gaps inherent in 2DSR models. We propose IE-SRGS, a novel 3DGS SR paradigm that addresses this issue by jointly leveraging the complementary strengths of external 2DSR priors and internal 3DGS features. Specifically, we use 2DSR and depth estimation models to generate HR images and depth maps as external knowledge, and employ multi-scale 3DGS models to produce cross-view consistent, domain-adaptive counterparts as internal knowledge. A mask-guided fusion strategy is introduced to integrate these two sources and synergistically exploit their complementary strengths, effectively guiding the 3D Gaussian optimization toward high-fidelity reconstruction. Extensive experiments on both synthetic and real-world benchmarks show that IE-SRGS consistently outperforms state-of-the-art methods in both quantitative accuracy and visual fidelity. △ Less

Submitted 27 November, 2025; originally announced November 2025.

Comments: AAAI 2026

DITING: A Multi-Agent Evaluation Framework for Benchmarking Web Novel Translation

Authors: Enze Zhang, Jiaying Wang, Mengxi Xiao, Jifei Liu, Ziyan Kuang, Rui Dong, Eric Dong, Sophia Ananiadou, Min Peng, Qianqian Xie

Abstract: Large language models (LLMs) have substantially advanced machine translation (MT), yet their effectiveness in translating web novels remains unclear. Existing benchmarks rely on surface-level metrics that fail to capture the distinctive traits of this genre. To address these gaps, we introduce DITING, the first comprehensive evaluation framework for web novel translation, assessing narrative and c… ▽ More Large language models (LLMs) have substantially advanced machine translation (MT), yet their effectiveness in translating web novels remains unclear. Existing benchmarks rely on surface-level metrics that fail to capture the distinctive traits of this genre. To address these gaps, we introduce DITING, the first comprehensive evaluation framework for web novel translation, assessing narrative and cultural fidelity across six dimensions: idiom translation, lexical ambiguity, terminology localization, tense consistency, zero-pronoun resolution, and cultural safety, supported by over 18K expert-annotated Chinese-English sentence pairs. We further propose AgentEval, a reasoning-driven multi-agent evaluation framework that simulates expert deliberation to assess translation quality beyond lexical overlap, achieving the highest correlation with human judgments among seven tested automatic metrics. To enable metric comparison, we develop MetricAlign, a meta-evaluation dataset of 300 sentence pairs annotated with error labels and scalar quality scores. Comprehensive evaluation of fourteen open, closed, and commercial models reveals that Chinese-trained LLMs surpass larger foreign counterparts, and that DeepSeek-V3 delivers the most faithful and stylistically coherent translations. Our work establishes a new paradigm for exploring LLM-based web novel translation and provides public resources to advance future research. △ Less

Submitted 13 October, 2025; v1 submitted 10 October, 2025; originally announced October 2025.

SkyLink: Unifying Street-Satellite Geo-Localization via UAV-Mediated 3D Scene Alignment

Authors: Hongyang Zhang, Yinhao Liu, Zhenyu Kuang

Abstract: Cross-view geo-localization aims at establishing location correspondences between different viewpoints. Existing approaches typically learn cross-view correlations through direct feature similarity matching, often overlooking semantic degradation caused by extreme viewpoint disparities. To address this unique problem, we focus on robust feature retrieval under viewpoint variation and propose the n… ▽ More Cross-view geo-localization aims at establishing location correspondences between different viewpoints. Existing approaches typically learn cross-view correlations through direct feature similarity matching, often overlooking semantic degradation caused by extreme viewpoint disparities. To address this unique problem, we focus on robust feature retrieval under viewpoint variation and propose the novel SkyLink method. We firstly utilize the Google Retrieval Enhancement Module to perform data enhancement on street images, which mitigates the occlusion of the key target due to restricted street viewpoints. The Patch-Aware Feature Aggregation module is further adopted to emphasize multiple local feature aggregations to ensure the consistent feature extraction across viewpoints. Meanwhile, we integrate the 3D scene information constructed from multi-scale UAV images as a bridge between street and satellite viewpoints, and perform feature alignment through self-supervised and cross-view contrastive learning. Experimental results demonstrate robustness and generalization across diverse urban scenarios, which achieve 25.75$\%$ Recall@1 accuracy on University-1652 in the UAVM2025 Challenge. Code will be released at https://github.com/HRT00/CVGL-3D. △ Less

Submitted 29 September, 2025; originally announced September 2025.

ChessArena: A Chess Testbed for Evaluating Strategic Reasoning Capabilities of Large Language Models

Authors: Jincheng Liu, Sijun He, Jingjing Wu, Xiangsen Wang, Yang Chen, Zhaoqi Kuang, Siqi Bao, Yuan Yao

Abstract: Recent large language models (LLMs) have shown strong reasoning capabilities. However, a critical question remains: do these models possess genuine strategic reasoning, or do they primarily excel at pattern recognition? To address this, we present ChessArena, a chess-based testbed for evaluating LLMs. Chess demands strategic reasoning, precise rule adherence, and the ability to track complex game… ▽ More Recent large language models (LLMs) have shown strong reasoning capabilities. However, a critical question remains: do these models possess genuine strategic reasoning, or do they primarily excel at pattern recognition? To address this, we present ChessArena, a chess-based testbed for evaluating LLMs. Chess demands strategic reasoning, precise rule adherence, and the ability to track complex game states. ChessArena is a competitive framework where LLMs play against each other under four play modes. We evaluate 13 LLMs across over 800 games, testing basic understanding, move selection, and puzzle solving. Results reveal significant shortcomings: no model beats Maia-1100 (human amateur level), and some lose to random play. We also present a strong baseline: our fine-tuned Qwen3-8B substantially improves performance, approaching much larger state-of-the-art reasoning models. △ Less

Submitted 23 April, 2026; v1 submitted 28 September, 2025; originally announced September 2025.

Learning Game-Playing Agents with Generative Code Optimization

Authors: Zhiyi Kuang, Ryan Rong, YuCheng Yuan, Allen Nie

Abstract: We present a generative optimization approach for learning game-playing agents, where policies are represented as Python programs and refined using large language models (LLMs). Our method treats decision-making policies as self-evolving code, with current observation as input and an in-game action as output, enabling agents to self-improve through execution traces and natural language feedback wi… ▽ More We present a generative optimization approach for learning game-playing agents, where policies are represented as Python programs and refined using large language models (LLMs). Our method treats decision-making policies as self-evolving code, with current observation as input and an in-game action as output, enabling agents to self-improve through execution traces and natural language feedback with minimal human intervention. Applied to Atari games, our game-playing Python program achieves performance competitive with deep reinforcement learning (RL) baselines while using significantly less training time and much fewer environment interactions. This work highlights the promise of programmatic policy representations for building efficient, adaptable agents capable of complex, long-horizon reasoning. △ Less

Submitted 26 August, 2025; originally announced August 2025.

Comments: ICML 2025 Workshop on Programmatic Representations for Agent Learning, Vancouver, Canada

AWM-Fuse: Multi-Modality Image Fusion for Adverse Weather via Global and Local Text Perception

Authors: Xilai Li, Huichun Liu, Xiaosong Li, Tao Ye, Zhenyu Kuang, Huafeng Li

Abstract: Multi-modality image fusion (MMIF) in adverse weather aims to address the loss of visual information caused by weather-related degradations, providing clearer scene representations. Although less studies have attempted to incorporate textual information to improve semantic perception, they often lack effective categorization and thorough analysis of textual content. In response, we propose AWM-Fus… ▽ More Multi-modality image fusion (MMIF) in adverse weather aims to address the loss of visual information caused by weather-related degradations, providing clearer scene representations. Although less studies have attempted to incorporate textual information to improve semantic perception, they often lack effective categorization and thorough analysis of textual content. In response, we propose AWM-Fuse, a novel fusion method for adverse weather conditions, designed to handle multiple degradations through global and local text perception within a unified, shared weight architecture. In particular, a global feature perception module leverages BLIP-produced captions to extract overall scene features and identify primary degradation types, thus promoting generalization across various adverse weather conditions. Complementing this, the local module employs detailed scene descriptions produced by ChatGPT to concentrate on specific degradation effects through concrete textual cues, thereby capturing finer details. Furthermore, textual descriptions are used to constrain the generation of fusion images, effectively steering the network learning process toward better alignment with real semantic labels, thereby promoting the learning of more meaningful visual features. Extensive experiments demonstrate that AWM-Fuse outperforms current state-of-the-art methods in complex weather conditions and downstream tasks. Our code is available at https://github.com/Feecuin/AWM-Fuse. △ Less

Submitted 22 August, 2025; originally announced August 2025.

From Scores to Skills: A Cognitive Diagnosis Framework for Evaluating Financial Large Language Models

Authors: Ziyan Kuang, Feiyu Zhu, Maowei Jiang, Yanzhao Lai, Zelin Wang, Zhitong Wang, Meikang Qiu, Jiajia Huang, Min Peng, Qianqian Xie, Sophia Ananiadou

Abstract: Large Language Models (LLMs) have shown promise for financial applications, yet their suitability for this high-stakes domain remains largely unproven due to inadequacies in existing benchmarks. Existing benchmarks solely rely on score-level evaluation, summarizing performance with a single score that obscures the nuanced understanding of what models truly know and their precise limitations. They… ▽ More Large Language Models (LLMs) have shown promise for financial applications, yet their suitability for this high-stakes domain remains largely unproven due to inadequacies in existing benchmarks. Existing benchmarks solely rely on score-level evaluation, summarizing performance with a single score that obscures the nuanced understanding of what models truly know and their precise limitations. They also rely on datasets that cover only a narrow subset of financial concepts, while overlooking other essentials for real-world applications. To address these gaps, we introduce FinCDM, the first cognitive diagnosis evaluation framework tailored for financial LLMs, enabling the evaluation of LLMs at the knowledge-skill level, identifying what financial skills and knowledge they have or lack based on their response patterns across skill-tagged tasks, rather than a single aggregated number. We construct CPA-KQA, the first cognitively informed financial evaluation dataset derived from the Certified Public Accountant (CPA) examination, with comprehensive coverage of real-world accounting and financial skills. It is rigorously annotated by domain experts, who author, validate, and annotate questions with high inter-annotator agreement and fine-grained knowledge labels. Our extensive experiments on 30 proprietary, open-source, and domain-specific LLMs show that FinCDM reveals hidden knowledge gaps, identifies under-tested areas such as tax and regulatory reasoning overlooked by traditional benchmarks, and uncovers behavioral clusters among models. FinCDM introduces a new paradigm for financial LLM evaluation by enabling interpretable, skill-aware diagnosis that supports more trustworthy and targeted model development, and all datasets and evaluation scripts will be publicly released to support further research. △ Less

Submitted 23 August, 2025; v1 submitted 18 August, 2025; originally announced August 2025.

MambaTrans: Multimodal Fusion Image Translation via Large Language Model Priors for Downstream Visual Tasks

Authors: Yushen Xu, Xiaosong Li, Zhenyu Kuang, Xiaoqi Cheng, Haishu Tan, Huafeng Li

Abstract: The goal of multimodal image fusion is to integrate complementary information from infrared and visible images, generating multimodal fused images for downstream tasks. Existing downstream pre-training models are typically trained on visible images. However, the significant pixel distribution differences between visible and multimodal fusion images can degrade downstream task performance, sometime… ▽ More The goal of multimodal image fusion is to integrate complementary information from infrared and visible images, generating multimodal fused images for downstream tasks. Existing downstream pre-training models are typically trained on visible images. However, the significant pixel distribution differences between visible and multimodal fusion images can degrade downstream task performance, sometimes even below that of using only visible images. This paper explores adapting multimodal fused images with significant modality differences to object detection and semantic segmentation models trained on visible images. To address this, we propose MambaTrans, a novel multimodal fusion image modality translator. MambaTrans uses descriptions from a multimodal large language model and masks from semantic segmentation models as input. Its core component, the Multi-Model State Space Block, combines mask-image-text cross-attention and a 3D-Selective Scan Module, enhancing pure visual capabilities. By leveraging object detection prior knowledge, MambaTrans minimizes detection loss during training and captures long-term dependencies among text, masks, and images. This enables favorable results in pre-trained models without adjusting their parameters. Experiments on public datasets show that MambaTrans effectively improves multimodal image performance in downstream tasks. △ Less

Submitted 11 August, 2025; originally announced August 2025.

Graph-R1: Towards Agentic GraphRAG Framework via End-to-end Reinforcement Learning

Authors: Haoran Luo, Haihong E, Guanting Chen, Qika Lin, Yikai Guo, Fangzhi Xu, Zemin Kuang, Meina Song, Xiaobao Wu, Yifan Zhu, Luu Anh Tuan

Abstract: Retrieval-Augmented Generation (RAG) mitigates hallucination in LLMs by incorporating external knowledge, but relies on chunk-based retrieval that lacks structural semantics. GraphRAG methods improve RAG by modeling knowledge as entity-relation graphs, but still face challenges in high construction cost, fixed one-time retrieval, and reliance on long-context reasoning and prompt design. To address… ▽ More Retrieval-Augmented Generation (RAG) mitigates hallucination in LLMs by incorporating external knowledge, but relies on chunk-based retrieval that lacks structural semantics. GraphRAG methods improve RAG by modeling knowledge as entity-relation graphs, but still face challenges in high construction cost, fixed one-time retrieval, and reliance on long-context reasoning and prompt design. To address these challenges, we propose Graph-R1, the first agentic GraphRAG framework via end-to-end reinforcement learning (RL). It introduces lightweight knowledge hypergraph construction, models retrieval as a multi-turn agent-environment interaction, and optimizes the agent process via an end-to-end reward mechanism. Experiments on standard RAG datasets show that Graph-R1 outperforms traditional GraphRAG and RL-enhanced RAG methods in reasoning accuracy, retrieval efficiency, and generation quality. Our software and data are publicly available at https://github.com/LHRLAB/Graph-R1. △ Less

Submitted 2 June, 2026; v1 submitted 29 July, 2025; originally announced July 2025.

Comments: Accepted by ICML 2026 main conference

Journal ref: ICML 2026

SciVer: Evaluating Foundation Models for Multimodal Scientific Claim Verification

Authors: Chengye Wang, Yifei Shen, Zexi Kuang, Arman Cohan, Yilun Zhao

Abstract: We introduce SciVer, the first benchmark specifically designed to evaluate the ability of foundation models to verify claims within a multimodal scientific context. SciVer consists of 3,000 expert-annotated examples over 1,113 scientific papers, covering four subsets, each representing a common reasoning type in multimodal scientific claim verification. To enable fine-grained evaluation, each exam… ▽ More We introduce SciVer, the first benchmark specifically designed to evaluate the ability of foundation models to verify claims within a multimodal scientific context. SciVer consists of 3,000 expert-annotated examples over 1,113 scientific papers, covering four subsets, each representing a common reasoning type in multimodal scientific claim verification. To enable fine-grained evaluation, each example includes expert-annotated supporting evidence. We assess the performance of 21 state-of-the-art multimodal foundation models, including o4-mini, Gemini-2.5-Flash, Llama-3.2-Vision, and Qwen2.5-VL. Our experiment reveals a substantial performance gap between these models and human experts on SciVer. Through an in-depth analysis of retrieval-augmented generation (RAG), and human-conducted error evaluations, we identify critical limitations in current open-source models, offering key insights to advance models' comprehension and reasoning in multimodal scientific literature tasks. △ Less

Submitted 18 June, 2025; originally announced June 2025.

Can LLMs Generate High-Quality Test Cases for Algorithm Problems? TestCase-Eval: A Systematic Evaluation of Fault Coverage and Exposure

Authors: Zheyuan Yang, Zexi Kuang, Xue Xia, Yilun Zhao

Abstract: We introduce TestCase-Eval, a new benchmark for systematic evaluation of LLMs in test-case generation. TestCase-Eval includes 500 algorithm problems and 100,000 human-crafted solutions from the Codeforces platform. It focuses on two pivotal tasks: (1) Fault Coverage, which measures how well LLM-generated test sets probe diverse input scenarios and cover a wide range of potential failure modes. (2)… ▽ More We introduce TestCase-Eval, a new benchmark for systematic evaluation of LLMs in test-case generation. TestCase-Eval includes 500 algorithm problems and 100,000 human-crafted solutions from the Codeforces platform. It focuses on two pivotal tasks: (1) Fault Coverage, which measures how well LLM-generated test sets probe diverse input scenarios and cover a wide range of potential failure modes. (2) Fault Exposure, which evaluates whether LLMs can craft a tailored test input that reveals a specific incorrect code implementation. We provide a comprehensive assessment of 19 state-of-the-art open-source and proprietary LLMs on TestCase-Eval, offering insights into their strengths and limitations in generating effective test cases for algorithm problems. △ Less

Submitted 13 June, 2025; originally announced June 2025.

Comments: ACL 2025

ECP-Mamba: An Efficient Multi-scale Self-supervised Contrastive Learning Method with State Space Model for PolSAR Image Classification

Authors: Zuzheng Kuang, Haixia Bi, Chen Xu, Jian Sun

Abstract: Recently, polarimetric synthetic aperture radar (PolSAR) image classification has been greatly promoted by deep neural networks. However,current deep learning-based PolSAR classification methods encounter difficulties due to its dependence on extensive labeled data and the computational inefficiency of architectures like Transformers. This paper presents ECP-Mamba, an efficient framework integrati… ▽ More Recently, polarimetric synthetic aperture radar (PolSAR) image classification has been greatly promoted by deep neural networks. However,current deep learning-based PolSAR classification methods encounter difficulties due to its dependence on extensive labeled data and the computational inefficiency of architectures like Transformers. This paper presents ECP-Mamba, an efficient framework integrating multi-scale self-supervised contrastive learning with a state space model (SSM) backbone. Specifically, ECP-Mamba addresses annotation scarcity through a multi-scale predictive pretext task based on local-to-global feature correspondences, which uses a simplified self-distillation paradigm without negative sample pairs. To enhance computational efficiency,the Mamba architecture (a selective SSM) is first tailored for pixel-wise PolSAR classification task by designing a spiral scan strategy. This strategy prioritizes causally relevant features near the central pixel, leveraging the localized nature of pixel-wise classification tasks. Additionally, the lightweight Cross Mamba module is proposed to facilitates complementary multi-scale feature interaction with minimal overhead. Extensive experiments across four benchmark datasets demonstrate ECP-Mamba's effectiveness in balancing high accuracy with resource efficiency. On the Flevoland 1989 dataset, ECP-Mamba achieves state-of-the-art performance with an overall accuracy of 99.70%, average accuracy of 99.64% and Kappa coefficient of 99.62e-2. Our code will be available at https://github.com/HaixiaBi1982/ECP_Mamba. △ Less

Submitted 1 June, 2025; originally announced June 2025.

HyperGraphRAG: Retrieval-Augmented Generation via Hypergraph-Structured Knowledge Representation

Authors: Haoran Luo, Haihong E, Guanting Chen, Yandan Zheng, Xiaobao Wu, Yikai Guo, Qika Lin, Yu Feng, Zemin Kuang, Meina Song, Yifan Zhu, Luu Anh Tuan

Abstract: Standard Retrieval-Augmented Generation (RAG) relies on chunk-based retrieval, whereas GraphRAG advances this approach by graph-based knowledge representation. However, existing graph-based RAG approaches are constrained by binary relations, as each edge in an ordinary graph connects only two entities, limiting their ability to represent the n-ary relations (n >= 2) in real-world knowledge. In thi… ▽ More Standard Retrieval-Augmented Generation (RAG) relies on chunk-based retrieval, whereas GraphRAG advances this approach by graph-based knowledge representation. However, existing graph-based RAG approaches are constrained by binary relations, as each edge in an ordinary graph connects only two entities, limiting their ability to represent the n-ary relations (n >= 2) in real-world knowledge. In this work, we propose HyperGraphRAG, a novel hypergraph-based RAG method that represents n-ary relational facts via hyperedges, and consists of knowledge hypergraph construction, retrieval, and generation. Experiments across medicine, agriculture, computer science, and law demonstrate that HyperGraphRAG outperforms both standard RAG and previous graph-based RAG methods in answer accuracy, retrieval efficiency, and generation quality. Our data and code are publicly available at https://github.com/LHRLAB/HyperGraphRAG. △ Less

Submitted 21 October, 2025; v1 submitted 27 March, 2025; originally announced March 2025.

Comments: Accepted by NeurIPS 2025 main conference

Journal ref: NeurIPS 2025

Fair Federated Medical Image Classification Against Quality Shift via Inter-Client Progressive State Matching

Authors: Nannan Wu, Zhuo Kuang, Zengqiang Yan, Ping Wang, Li Yu

Abstract: Despite the potential of federated learning in medical applications, inconsistent imaging quality across institutions-stemming from lower-quality data from a minority of clients-biases federated models toward more common high-quality images. This raises significant fairness concerns. Existing fair federated learning methods have demonstrated some effectiveness in solving this problem by aligning a… ▽ More Despite the potential of federated learning in medical applications, inconsistent imaging quality across institutions-stemming from lower-quality data from a minority of clients-biases federated models toward more common high-quality images. This raises significant fairness concerns. Existing fair federated learning methods have demonstrated some effectiveness in solving this problem by aligning a single 0th- or 1st-order state of convergence (e.g., training loss or sharpness). However, we argue in this work that fairness based on such a single state is still not an adequate surrogate for fairness during testing, as these single metrics fail to fully capture the convergence characteristics, making them suboptimal for guiding fair learning. To address this limitation, we develop a generalized framework. Specifically, we propose assessing convergence using multiple states, defined as sharpness or perturbed loss computed at varying search distances. Building on this comprehensive assessment, we propose promoting fairness for these states across clients to achieve our ultimate fairness objective. This is accomplished through the proposed method, FedISM+. In FedISM+, the search distance evolves over time, progressively focusing on different states. We then incorporate two components in local training and global aggregation to ensure cross-client fairness for each state. This gradually makes convergence equitable for all states, thereby improving fairness during testing. Our empirical evaluations, performed on the well-known RSNA ICH and ISIC 2019 datasets, demonstrate the superiority of FedISM+ over existing state-of-the-art methods for fair federated learning. The code is available at https://github.com/wnn2000/FFL4MIA. △ Less

Submitted 12 March, 2025; originally announced March 2025.

Comments: Preprint

ILETIA: An AI-enhanced method for individualized trigger-oocyte pickup interval estimation of progestin-primed ovarian stimulation protocol

Authors: Binjian Wu, Qian Li, Zhe Kuang, Hongyuan Gao, Xinyi Liu, Haiyan Guo, Qiuju Chen, Xinyi Liu, Yangruizhe Jiang, Yuqi Zhang, Jinyin Zha, Mingyu Li, Qiuhan Ren, Sishuo Feng, Haicang Zhang, Xuefeng Lu, Jian Zhang

Abstract: In vitro fertilization-embryo transfer (IVF-ET) stands as one of the most prevalent treatments for infertility. During an IVF-ET cycle, the time interval between trigger shot and oocyte pickup (OPU) is a pivotal period for follicular maturation, which determines mature oocytes yields and impacts the success of subsequent procedures. However, accurately predicting this interval is severely hindered… ▽ More In vitro fertilization-embryo transfer (IVF-ET) stands as one of the most prevalent treatments for infertility. During an IVF-ET cycle, the time interval between trigger shot and oocyte pickup (OPU) is a pivotal period for follicular maturation, which determines mature oocytes yields and impacts the success of subsequent procedures. However, accurately predicting this interval is severely hindered by the variability of clinicians'experience that often leads to suboptimal oocyte retrieval rate. To address this challenge, we propose ILETIA, the first machine learning-based method that could predict the optimal trigger-OPU interval for patients receiving progestin-primed ovarian stimulation (PPOS) protocol. Specifically, ILETIA leverages a Transformer to learn representations from clinical tabular data, and then employs gradient-boosted trees for interval prediction. For model training and evaluating, we compiled a dataset PPOS-DS of nearly ten thousand patients receiving PPOS protocol, the largest such dataset to our knowledge. Experimental results demonstrate that our method achieves strong performance (AUROC = 0.889), outperforming both clinicians and other widely used computational models. Moreover, ILETIA also supports premature ovulation risk prediction in a specific OPU time (AUROC = 0.838). Collectively, by enabling more precise and individualized decisions, ILETIA has the potential to improve clinical outcomes and lay the foundation for future IVF-ET research. △ Less

Submitted 25 January, 2025; originally announced January 2025.

X-Dyna: Expressive Dynamic Human Image Animation

Authors: Di Chang, Hongyi Xu, You Xie, Yipeng Gao, Zhengfei Kuang, Shengqu Cai, Chenxu Zhang, Guoxian Song, Chao Wang, Yichun Shi, Zeyuan Chen, Shijie Zhou, Linjie Luo, Gordon Wetzstein, Mohammad Soleymani

Abstract: We introduce X-Dyna, a novel zero-shot, diffusion-based pipeline for animating a single human image using facial expressions and body movements derived from a driving video, that generates realistic, context-aware dynamics for both the subject and the surrounding environment. Building on prior approaches centered on human pose control, X-Dyna addresses key shortcomings causing the loss of dynamic… ▽ More We introduce X-Dyna, a novel zero-shot, diffusion-based pipeline for animating a single human image using facial expressions and body movements derived from a driving video, that generates realistic, context-aware dynamics for both the subject and the surrounding environment. Building on prior approaches centered on human pose control, X-Dyna addresses key shortcomings causing the loss of dynamic details, enhancing the lifelike qualities of human video animations. At the core of our approach is the Dynamics-Adapter, a lightweight module that effectively integrates reference appearance context into the spatial attentions of the diffusion backbone while preserving the capacity of motion modules in synthesizing fluid and intricate dynamic details. Beyond body pose control, we connect a local control module with our model to capture identity-disentangled facial expressions, facilitating accurate expression transfer for enhanced realism in animated scenes. Together, these components form a unified framework capable of learning physical human motion and natural scene dynamics from a diverse blend of human and scene videos. Comprehensive qualitative and quantitative evaluations demonstrate that X-Dyna outperforms state-of-the-art methods, creating highly lifelike and expressive animations. The code is available at https://github.com/bytedance/X-Dyna. △ Less

Submitted 20 January, 2025; v1 submitted 17 January, 2025; originally announced January 2025.

Comments: Project page:https://x-dyna.github.io/xdyna.github.io/ Code:https://github.com/bytedance/X-Dyna Model:https://huggingface.co/Boese0601/X-Dyna

OCRBench v2: An Improved Benchmark for Evaluating Large Multimodal Models on Visual Text Localization and Reasoning

Authors: Ling Fu, Zhebin Kuang, Jiajun Song, Mingxin Huang, Biao Yang, Yuzhe Li, Linghao Zhu, Qidi Luo, Xinyu Wang, Hao Lu, Zhang Li, Guozhi Tang, Bin Shan, Chunhui Lin, Qi Liu, Binghong Wu, Hao Feng, Hao Liu, Can Huang, Jingqun Tang, Wei Chen, Lianwen Jin, Yuliang Liu, Xiang Bai

Abstract: Scoring the Optical Character Recognition (OCR) capabilities of Large Multimodal Models (LMMs) has witnessed growing interest. Existing benchmarks have highlighted the impressive performance of LMMs in text recognition; however, their abilities in certain challenging tasks, such as text localization, handwritten content extraction, and logical reasoning, remain underexplored. To bridge this gap, w… ▽ More Scoring the Optical Character Recognition (OCR) capabilities of Large Multimodal Models (LMMs) has witnessed growing interest. Existing benchmarks have highlighted the impressive performance of LMMs in text recognition; however, their abilities in certain challenging tasks, such as text localization, handwritten content extraction, and logical reasoning, remain underexplored. To bridge this gap, we introduce OCRBench v2, a large-scale bilingual text-centric benchmark with currently the most comprehensive set of tasks (4x more tasks than the previous multi-scene benchmark OCRBench), the widest coverage of scenarios (31 diverse scenarios), and thorough evaluation metrics, with 10,000 human-verified question-answering pairs and a high proportion of difficult samples. Moreover, we construct a private test set with 1,500 manually annotated images. The consistent evaluation trends observed across both public and private test sets validate the OCRBench v2's reliability. After carefully benchmarking state-of-the-art LMMs, we find that most LMMs score below 50 (100 in total) and suffer from five-type limitations, including less frequently encountered text recognition, fine-grained perception, layout perception, complex element parsing, and logical reasoning. The project website is at: https://99franklin.github.io/ocrbench_v2/ △ Less

Submitted 4 June, 2025; v1 submitted 31 December, 2024; originally announced January 2025.