Artificial Intelligence
AdaGuard: An Adaptive Guard Model with User-defined Policies
Yunhao Feng, Yifan Ding, Yuxiang Xie et al. (+4 more)
Guard models support the safe deployment of
language model
agents, but fixed risk taxonomies limit their ability to accommodate requirements that vary across applications and tasks. Under user-defined policies, detecting violations requires interpreting both the applicable rules and the
agent's behavior, since identical actions can receive different judgments under different policies. To support learning this capability, we introduce AdaptiveSafety, a
dataset of 10,939 training examples and 1,000 test examples covering policies with 1–100 rules. The
dataset combines trajectories from multiple sources with policy and behavioral counterfactuals, pairing each example with an explanation and the complete set of violated rules. These counterfactuals expose changes that alter compliance, while structural augmentations provide supervision for consistency under rule reordering and identifier remapping. Building on this supervision, we propose SafePO, a reinforcement learning
algorithm for refining violation identification while balancing explanatory
reasoning and final verdicts. SafePO uses structured rewards to assess prediction correctness, retains group-relative advantages at the response level, and employs a separately trained value model to modulate
token weights within explanation and verdict regions. Separate normalization controls their relative contribution to training despite differences in length. Through supervised initialization followed by SafePO, we develop AdaGuard, a family of 0.6B, 4B, and 8B guard models that assess
agent trajectories under policies supplied at inference time. Our 4B model achieves binary accuracies of 89.30% on AdaptiveSafety and 71.82% on DynaBench. The project repository is available at https://github.com/Yunhao-Feng/AdaGuard
Read the full abstract ▾
All authors Yunhao Feng, Yifan Ding, Yuxiang Xie, Zheng Li, Mingrui Lao, Zeyuan Wang, Yanming Guo
Computation and Language
Coherence-Aware Distributional Evaluation of Open-Ended Text Generation
Jinnuo Liu, Junhao Zhu, Weifeng Jiang et al. (+2 more)
Existing open-ended generation metrics measure likelihood, lexical diversity, or distributional similarity in generic representation space, yet can miss fundamental dimensions of quality. A prominent blind spot is global coherence: a generated passage may be locally fluent while remaining globally contradictory, causally inconsistent, or topically disconnected. We identify representation as a central bottleneck in detecting these failures and introduce CHORD (Coherence-aware Hidden-state Open-generation Reference Distance), a coherence-sensitive distributional metric. CHORD encodes generated and human-written corpora in the hidden-state space of a frozen LLM using a coherence-eliciting
prompt, and compares the resulting distributions using RBF-MMD. To test coherence sensitivity and selectivity, we construct a counterfactual evaluation suite pairing graded coherence-degrading perturbations with meaning-preserving controls. CHORD selectively detects relation, discourse, structural, and mixture failures that perplexity,
entropy, MAUVE, FBD, and MMD-based baselines either miss or cannot separate from benign rewriting. Factorial ablations show that representation is the primary source of coherence sensitivity, while RBF-MMD improves sample efficiency. Larger backbones capture finer-grained distinctions, but coherence prompting improves selectivity only when the backbone can follow the
prompt. On unconditional generation and prefix continuation, CHORD yields model rankings that strongly align with human judgments of whether outputs make sense and appear human-written. Together, these results establish representation design as central to reliable distributional evaluation. Code: https://github.com/MAPS-research/CHORD. Experiments: https://github.com/MAPS-research/CHORD-Experiment.
Read the full abstract ▾
All authors Jinnuo Liu, Junhao Zhu, Weifeng Jiang, Haoming Liu, Hongyi Wen
Computer Vision
DecFlowEdit: Self-Localized Flow-based Image Editing via Guidance Decoupling
Zheyuan Zhan, Can Wang, Jiawei Chen et al. (+4 more)
Flow-based image editing (FlowEdit) enables inversion-free semantic changes through the difference between source and target velocities. In this paper, we observe that FlowEdit's default classifier-free guidance (CFG) configuration, with asymmetric source and target scales, causes substantial background leakage. Matching these guidance scales, for example by removing CFG, improves edit-relevant localization but severely degrades editability. To get the best of both worlds, we propose DecFlowEdit, which decouples the optimal guidance scales for localization and for editing in flow-based generative models. In particular, DecFlowEdit first extracts an edit-relevant prior by temporally aggregating velocity differences evaluated without CFG, and then uses this prior to reweight the original updates under default CFG. Our method remains training-free and inversion-free, requiring neither external spatial masks nor
attention manipulation. Experiments on PIE-Bench across FLUX, SD3, and SD3.5 show that DecFlowEdit improves background preservation, reducing structure distance by approximately 61 to 73 percent and background LPIPS by 68 to 80 percent relative to FlowEdit at comparable editing fidelity.
Read the full abstract ▾
All authors Zheyuan Zhan, Can Wang, Jiawei Chen, Chun Chen, Siwei Lyu, Zeyu Zheng, Defang Chen
Discrete Mathematics
Computing the Helly Number, Radon Number and Rank in Cycle Convexity
Revathy S.Nair, Bijo S. Anand, Ullas Chandran S.V. et al. (+2 more)
In this paper, we investigate three fundamental convexity parameters of graphs under cycle convexity, namely the Helly number, Radon number, and rank. We first study the computational complexity of these parameters. For each of these parameters, we consider the associated threshold decision problem of determining whether the parameter of a given graph is at least a prescribed integer. We establish that all three problems are $\NP$-hard and $\W[1]$-hard when parameterized by the threshold. Moreover, we strengthen these results by showing that the $\NP$-hardness persists even when the input is restricted to planar graphs of maximum degree at most $6$. We also focus on the structural properties of connected graphs corresponding to extremal values of these parameters. In particular, we characterize the graph classes for which the three parameters attain the values $n-1$ and $n-2$, where $n$ is the order of $G$.
Read the full abstract ▾
All authors Revathy S.Nair, Bijo S. Anand, Ullas Chandran S.V., Julliano R. Nascimento, Arun Anil
Computer Vision
SegBanana: Steering Unified Multimodal Models into Medical Segmenters
Xiaoye Liang, Ye Yan, Mingze Yin et al. (+5 more)
Medical image
segmentation remains challenging in practical deployment, as models often struggle to generalize beyond the distributions covered by their training data and high-quality pixel-level annotations are typically unavailable for adaptation. Inspired by the cross-task transferability of large
language models, we investigate whether unified
multimodal models (UMMs) can transfer their pretrained visual understanding,
reasoning, and generation capabilities to medical image
segmentation without task-specific post-training. By recasting
segmentation as structured visual generation, we find that frontier UMMs (e.g., Nano Banana) already exhibit basic
segmentation capabilities across diverse clinical scenarios, but still struggle with challenging tasks requiring specialized anatomical or domain-specific knowledge. We further show that these limitations can be effectively mitigated by incorporating visual anatomical knowledge from in-context exemplars, expanding candidate solutions through repeated sampling, and refining suboptimal predictions via targeted editing.Motivated by these observations, we propose SegBanana, to our knowledge, the first agentic visual generation framework for training-free medical image
segmentation. SegBanana builds on a frozen UMM as the core generative model, augmented with Anatomy-Aware Knowledge
Retrieval and Comparative Quality Critique to unlock its potential
segmentation capability. A State-Aware
Multimodal Controller maintains structured state and iteratively orchestrates these tools, repeatedly refining intermediate predictions toward higher-quality masks. Across eight medical
segmentation datasets, SegBanana achieves an average mDice of 77.45%, outperforming representative generalist (SAM3 and SegGPT) and medical-specific (BiomedParse and MedSAM3) baselines by at least 14.93 points, while remaining robust to out-of-domain visual supports.
Read the full abstract ▾
All authors Xiaoye Liang, Ye Yan, Mingze Yin, Shikun Feng, Mai Xu, Haiguang Liu, Lai Jiang, Yiheng Zhu
Computation and Language
MAS-OPD: On-Policy Distillation for Multi-agent Systems
Qiyong Zhong, Mao Zheng, Mingyang Song et al. (+5 more)
Multi-
agent systems (MAS) split a task across specialized roles and are promising on complex tasks, yet a prevailing approach relies on inference-time orchestration alone. General-purpose APIs are costly and hard to customize, while small models with role
prompts rarely develop stable role competence or reliable collaboration, so post-training a MAS jointly is central. Most attempts use reinforcement learning, whose team-level reward leaves undetermined which step of which
agent brought about the outcome, while local rewards need redesigning per task. On-policy
distillation (OPD) gives
token-level teacher supervision on trajectories the student samples, a denser signal needing no local reward, yet is underexplored for the interdependent
agents of a MAS. Two difficulties arise: building complementary specialization from a judgement of which role a behavior belongs to while preserving the knowledge all roles need, and turning cross-
agent collaborative information into supervision OPD can exploit. We present MAS-OPD, where Role-Advantage Specialization defines the role advantage as the difference between the teacher signals under target and non-target role conditions, and Privileged Attribution for Coordination attributes an interaction conflict to its source and supplies it to the teacher alone as privileged information. Extensive experiments on code and mathematics
benchmarks show that MAS-OPD attains the highest mean score at both student scales and leads the
agents to develop clearer role specialization and more effective collaborative behavior.
Read the full abstract ▾
All authors Qiyong Zhong, Mao Zheng, Mingyang Song, Houcheng Jiang, Jiajie Su, Huwei Ji, Li Zhang, Junfeng Fang
Robotics
GAE: General Action Expert for Real-Time Humanoid Teleoperation
Yuefan Wang, Huaicheng Zhou, Xiao He et al. (+6 more)
Humanoid avatars extend human physical presence beyond the body, enabling people to participate in social, service, and labor activities through remotely operated
robots. This requires teleoperation systems capable of realizing diverse and dynamic whole-body behaviors while maintaining responsive human-
robot synchronization. We present General Action Expert(GAE), a unified learning framework for general-purpose, low-latency
humanoid whole-body teleoperation. To cover diverse human behaviors, GAE builds a large-scale human motion
dataset from heterogeneous sources, including
videos, animations, and motion capture, followed by standardization and augmentation. GAE then addresses the noise and embodiment mismatch in human motions with a two-stage training paradigm: a privileged generator policy first tracks human motion references in simulation and rolls out feasible
humanoid trajectories; a deployable executor policy then learns to track these generated trajectories under curriculum domain randomization. For responsive human-
robot synchronization, GAE introduces a latency-conditioned anticipation mechanism that adaptively compensates for end-to-end delay during real-time teleoperation. Simulation and real-world experiments on Unitree G1 and Westlake O1
robots demonstrate that GAE enables
humanoids to smoothly mirror diverse, agile, and expressive human behaviors. Project website: https://wangyf0928.github.io/gae-wlrobotics/
Read the full abstract ▾
All authors Yuefan Wang, Huaicheng Zhou, Xiao He, Zhijie He, Mingchuan Yang, Huayi Zhang, Li Chai, Jinxin Liu, Donglin Wang
Computer Vision
Trustworthy synthetic visual media: Evidence across the media lifecycle
Zexi Jia, Zhiqiang Yuan, Jie Zhou et al. (+1 more)
Images and
videos have long helped people understand what happened and how a work came into being. Generative systems complicate that role. Realistic media can now be produced and revised without leaving a stable history, so appearance no longer reveals whether a scene was captured, synthesized, or altered along the way. Trust must instead come from evidence that explains the path an asset has taken and the circumstances in which it was used. Some of this evidence can be recovered from the media, while some must be recorded during production and preserved as the asset circulates. This review brings those approaches together and asks when their claims remain meaningful after ordinary processing or deliberate manipulation. We argue that trustworthy media do not depend on one universal marker of authenticity. The evidence must suit the question at hand, reach the person making the judgment, and remain open to correction when better information emerges. The larger goal is to keep the history of media intelligible even as the media itself continues to change.
Read the full abstract ▾
All authors Zexi Jia, Zhiqiang Yuan, Jie Zhou, Jinchao Zhang
Computer Vision
ReGDiff: Guided Diffusion in Regulated Latent Space for Exploring Metamaterial Voxel Geometry
Wangzhi Zhan, Jianpeng Chen, Dongqi Fu et al. (+1 more)
Metamaterials are artificially engineered structures whose mechanical and physical behaviors are strongly shaped by geometry rather than composition. Voxel representation provides a unified format for metamaterial geometry generation, as it can express diverse classes such as truss, shell, and porous structures within a single cubic discretization. However, voxel-based generation faces a plausibility-novelty trade-off: staying close to known geometries helps preserve
geometric regularities, while moving away from them is necessary for novelty but may produce degenerate geometries. To address this challenge, we propose REGDIFF, a generative framework that couples voxel representation with latent space regulation and guided
diffusion. REGDIFF introduces a repel-and-sink (RAS) mechanism to smooth the latent distribution of plausible geometries, and short-range repulsion (SRR) guidance to discourage generation overly close to known samples while maintaining
geometric plausibility. We further contribute a voxel-based
benchmark covering truss- and shell-type metamaterial geometries, together with an evaluation module for
geometric plausibility, novelty, and diversity. Experiments show that REGDIFF outperforms voxel-based generative baselines, achieving +8.9% in
geometric plausibility, +46.4% in novelty, and +128.6% in diversity on average across two
datasets. These results suggest that REGDIFF is a strong geometry candidate generator for downstream evaluation. Our code is provided at https://github.com/wzhan24/ReGDiff.
Read the full abstract ▾
All authors Wangzhi Zhan, Jianpeng Chen, Dongqi Fu, Dawei Zhou
Information Retrieval
Measuring and Mitigating Identity-Cue Preference Drift in LLM-based Recommender Systems
Zhuoxiong Gan and Qiang Dong
In large
language model-based recommender systems, identity cues embedded in
prompts can steer recommendations toward group-level patterns even when the underlying behavioral evidence remains unchanged. We introduce PromptShift, an interpretable, training-free framework for quantifying and mitigating such identity-cue preference drift. We define Drift as the divergence, in both item membership and ranking order, between a recommendation list generated under an identity-cued
prompt and the reference list produced from the same user's interaction history alone. SliceShift then measures the extent to which a cued list gravitates, relative to the history-only reference, toward items that are more popular within the cued slice than among the global user population. Beyond conventional accuracy, we propose DifHitRate, a difficulty-weighted hit metric that credits only relevant items, assigning higher credit to hits that are less popular within the cued slice and ranked higher in the list. All components are supported by an identity-slice-by-item table constructed from positive interactions, which further enables an adaptive post-hoc reranking strategy: the reranker interpolates between the original LLM ranking and inverse slice-popularity, with personalized interpolation weight. Experiments on two
datasets with three LLMs show that identity-cued
prompts incur higher mean Drift than identity-free paraphrase controls, an effect beyond generic wording sensitivity, and that SliceShift is positive across all six
dataset-model settings. PromptShift consistently reduces both Drift and SliceShift, lowering macro-mean SliceShift by 62.42%, while improving DifHitRate, HitRate and MRR. These results demonstrate that identity-cue preference drift can be measured and mitigated without any model training, albeit with a modest, metric-dependent utility cost.
Read the full abstract ▾
Machine Learning
SleuthBench: Benchmarking Statistical LLM Evaluation Using Tabular Hidden Signals
Jingyun Jia, Antoine Remond-Tiedrez, Aaron Alvarez et al. (+3 more)
Evaluating statistical discovery by large
language model (LLM)
agents requires verifiable analytical ground truth. Establishing such ground truth for real-world
datasets is costly, and prior knowledge of public
datasets can influence
agent responses. We introduce SLEUTHBENCH, a
benchmark that addresses both problems by injecting controlled data-quality problems and feature effects into public tabular
datasets: the injected pattern determines the answer, so reference answers are computed automatically and memorized knowledge of the original table is insufficient, while the table keeps its background structure. The injected patterns are modeled on phenomena reported in real data analyses. The
benchmark defines 17 question templates in two families: data-quality questions and feature-contribution questions. We evaluate six state-of-the-art LLMs that analyze the data using a Python coding tool, on data-science and business phrasings of 70 validated
dataset-template combinations, yielding 1680 graded responses in total. The models detect data-quality problems reliably (83.8% accuracy) but recover feature contributions poorly (41.9%). Finding how features shape the target requires searching over both candidate variables and analytical procedures. To address this issue, we propose the Empirical Layer, a set of precomputed statistical artifacts comprising summaries, fitted feature and interaction effects, and
dataset descriptions, which exposes candidate patterns for direct inspection. Access to these artifacts raises feature-contribution accuracy from 41.9% to 68.0%.
Read the full abstract ▾
All authors Jingyun Jia, Antoine Remond-Tiedrez, Aaron Alvarez, Joshua Shunk, Rich Caruana, Ben Lengerich
Artificial Intelligence
When Does Selection Replace Extraction? A Pre-Registered Test of Agent Memory with a Typed Decision Model
Rishabh Sharma and Rishika Lall
Does conversational
memory need LLM-extracted facts, or is selecting the right raw turns enough? Published results disagree. Extraction-based systems report gains from distilled facts. Recent studies find raw history with good ranking does as well, but disagree about whether ranking matters. We ran a pre-registered study on held-out LoCoMo conversations and LongMemEval. At a tight budget on LoCoMo, raw turns selected by a single call to Jev, a typed decision model, are non-inferior to an LLM-extraction
memory (one-sided 95% bound -3.0 points against a -5-point margin). Blind human grading narrows the margin but does not change the result. Raw turns cost 3,061 times less to write, and the result holds with a second answer model. Within this study, reranking's gain shrinks as the budget grows. It adds 17.4 points on LoCoMo and 9.1 on LongMemEval when three of 30 candidates are kept. At generous budgets it adds 1.5 and 1.1, and extraction systems are more accurate. This suggests why published results disagree. At matched context, Jev selects as accurately as an LLM reranker (non-inferiority bound -2.0) at a third of the latency, and more accurately than a multi-call graph traversal. Reranking lowers correct abstention. Plans, code and graded answers are released.
Read the full abstract ▾
Computation and Language
USA: Update-aware SAM for Cross-domain On-Policy Disitllation of Language Agents
Qiyong Zhong, Mao Zheng, Mingyang Song et al. (+6 more)
On-policy
distillation instils multi-turn agentic
reasoning through dense
token-level supervision on the student's own trajectories, but a single domain saturates early, so further supervision has to be drawn from other domains. Multi-domain data mixing is the most direct way of incorporating them, at the cost of conflicts between their data distributions and of retraining the entire model whenever one domain is revised. Model merging avoids both by distilling every domain independently and fusing the resulting task vectors afterwards. We find instead that the benefit polarizes across domain pairs: on those exhibiting negative transfer, every merging operator we evaluate falls below the single-domain reference. We attribute this to cross-domain update coupling, where a substantial fraction of coordinates is updated comparably by both domains and a merge can therefore displace them by as much as their own updates. To overcome this limitation, we propose USA, which converts per-parameter update magnitudes measured during a brief warm-up into per-coordinate perturbation radii, reducing
curvature precisely on the coordinates that carry most of the merging displacement. Experiments across mathematics, science and code at two student scales show USA strongest in all six transfer directions, ahead of the single-domain reference by more than four points on average, and reverse the negative transfer of the conflicting pairs.
Read the full abstract ▾
All authors Qiyong Zhong, Mao Zheng, Mingyang Song, Huwei Ji, Houcheng Jiang, Jiajie Su, Li Zhang, Gengsheng Li, Junfeng Fang
Computer Vision
Uncovering Ordinal-Matching Bias in Audio-Visual LLMs
Jihoo Jung, Youngjoon Jang, Hyebin Cho et al. (+2 more)
This work aims to improve how audio-visual large
language models (AVLLMs) associate
speech with the correct visible speaker in multi-speaker scenes. We find that current AVLLMs frequently fail at this task, and analyze the nature of these failures. To this end, we construct a synthetic diagnostic
dataset in which multiple visible speakers each utter a single word. Analysis on this corpus reveals a consistent error pattern across three recent open-source AVLLMs: models attribute utterances by simply matching the order of spoken sentences with the left-to-right, top-to-bottom arrangement of visible faces, rather than relying on audio-visual cues such as lip synchronization. We term this behavior ordinal-matching bias. We further show that this bias can be substantially mitigated through a simple remedy, Ordinal-Decoupled Fine-Tuning (OD-FT), in which models are fine-tuned on synthetic
videos where spatial positions of speakers and speaking order are independently randomized. Despite using only 400 synthetic training
videos, OD-FT not only suppresses ordinal-matching bias but also improves audio-visual understanding on real-world
videos, yielding average gains of 8.27% for Qwen2.5-Omni and 2.57% for
video-SALMONN2+ across three audio-visual
benchmarks.
Read the full abstract ▾
All authors Jihoo Jung, Youngjoon Jang, Hyebin Cho, Suho Yoo, Joon Son Chung
Robotics
Proprioceptive Force Estimation for Quadruped Locomotion and Human-Robot Interaction
Run Wang, Xu Yang, Alapati Tuerxun et al. (+1 more)
Payload forces must be accommodated during locomotion, while leash forces can specify desired motion. We investigate whether a shared three-dimensional force estimate in newtons, inferred from proprioceptive history under sustained loading, can support both tasks. An
estimator and locomotion policy are jointly trained with supervised force and velocity outputs and learned latent context. The estimated force conditions locomotion and additionally generates planar-velocity and yaw-rate commands for leash guidance through an analytical map. In sustained-force simulation sweeps, temporal means of componentwise force root mean square error range from 1.44 to 2.83\,N. Compared with a domain-randomized baseline, the framework reduces velocity-tracking and base-orientation error scores by 21.6% and 46.5%, respectively, and increases mean survival from 68.29% to 94.60% in separate sustained-force tests. Unitree Go1 experiments demonstrate stationary vertical and horizontal force estimation, locomotion with an 8.5\,kg payload whose weight exceeds the 70\,N training force limit, and leash guidance using the same force-estimation interface.
Read the full abstract ▾
All authors Run Wang, Xu Yang, Alapati Tuerxun, Yilin Mo
Computer Vision
WorldWeave: Growing Persistent Geometric Worlds for Video Generation
Yifan Huang, Lifan Jiang, Qingyue Hao et al. (+5 more)
Despite rapid progress, world models still lack explicit, persistent structural
memory, making it difficult to preserve consistent world structure during continual scene expansion and cross-view revisits. To address this limitation, we present WorldWeave, a world generation framework that decouples world-state maintenance from visual rendering. Specifically, WorldWeave combines continual elevation-map generation with
agent-guided scene organization and stitching to build an expandable explicit 3D world that incrementally extends structural
memory while preserving existing structure. First, its terrain module uses
diffusion-based image outpainting to generate continuous metric elevation maps under neighborhood conditioning and boundary constraints. Next, an
agent integrates user intent, terrain evidence, and cross-region connectivity constraints to construct scenes through hierarchical semantic planning, deterministic geometry compilation, and local revision. Finally, during visual generation, planned camera trajectories query world geometry through a read-only interface, producing depth sequences that guide
video synthesis without writing the generated results back into the world state. As a result, structural
memory remains independent of short-window
video generation, enabling continual expansion without predefined map boundaries and providing a consistent
geometric basis for observations across trajectories and repeated visits.
Read the full abstract ▾
All authors Yifan Huang, Lifan Jiang, Qingyue Hao, Cheng Chen, Boxi Wu, Xiaoxue Ren, Xiaofei He, Dehai Zhao
Robotics
mmHRI: Towards Privacy-Preserving Human-Robot Interaction with Millimeter-Wave Radar
Junqiao Fan, Yuxuan Hu, Bofan Lyu et al. (+7 more)
Assistive
robots increasingly operate in many human-centered environments and perform various human-
robot interaction (HRI) tasks, such as object delivery. However, most existing HRI systems rely on RGB cameras that continuously observe humans to respond to non-verbal commands, such as hand gestures. This raises
privacy concerns in
privacy- critical environments, such as hospital wards or restaurants, where direct camera observation of humans is restricted. To develop
privacy-preserving HRI, we leverage millimeter-wave (mmWave) radar, which can sense human motion through
privacy barriers without identifiable imagery. We propose mmHRI, the first multi-modal
robot manipulation framework that achieves mmWave radar-guided
privacy-preserving HRI. mmHRI introduces two key designs to mitigate the sparsity and temporal inconsistency of radar data in cluttered
robot manipulation environments. First, we propose a dual-stream architecture that jointly learns from unfiltered raw radar tensors and radar point clouds to estimate both human actions and 3D poses. To mitigate signal inconsistency, mmHRI further incorporates a
memory-based state-space model (MSSM) that retains historical radar features to reduce abrupt changes in pose/action. These estimated human states are then converted into structured textual
robot instructions, which control a
vision-
language-action (VLA) policy for closed-loop
robot manipulation and human-aware reactions. Our evaluation covers human action recognition and closed-loop delivery and
retrieval. In the
privacy-preserving curtain setting, mmHRI achieves 85.09% action-recognition accuracy, outperforming existing radar-based alternatives.
Robot trials further demonstrate successful delivery and
retrieval under visual occlusion, with stable task performance across unseen subjects, clutter configurations, and environments.
Read the full abstract ▾
All authors Junqiao Fan, Yuxuan Hu, Bofan Lyu, Yanshuo Lu, Pengfei Liu, Jiarui Zhang, Fangqiang Ding, Lihua Xie, Gen Li, Jianfei Yang
Machine Learning
Loop Dropout: Regularizing Shared Updates in Looped Language Models
Zirui Zhu, Hailun Xu, Xuanlei Zhao et al. (+5 more)
Looped
language models separate computational depth from parameter count by repeatedly applying the same
transformer block. Adapting these models requires a shared update that remains effective as hidden states evolve throughout the recurrent computation. Our empirical analysis reveals a pronounced late-loop bias in standard low-rank adaptation (LoRA): the shared update provides limited adaptation at early loop positions. This imbalance motivates training shared updates under varying combinations of their applications. Randomly omitting adapter applications alone, however, does not improve task performance; it reduces expected update strength during training while leaving inference unchanged. We introduce Loop Dropout, which couples
stochastic masking of adapter applications with inverse-survival rescaling to preserve expected update strength and promote effective adaptation across loops. Extensive experiments demonstrate improved mathematical
reasoning across model sizes, adapter ranks and training recipes, with benefits extending to general instruction tuning and code generation. Loop Dropout outperforms existing LoRA variants and adapter regularizers, while further analysis shows stronger early-loop adaptation. Every backbone loop remains active, and inference applies the adapter at all loops using standard LoRA without additional trainable parameters or inference computation. Code is available at https://github.com/NUS-HPC-AI-Lab/loop-dropout .
Read the full abstract ▾
All authors Zirui Zhu, Hailun Xu, Xuanlei Zhao, Yong Liu, Yingxuan Ren, Kanchan Sarkar, Kun Xu, Yang You
Computation and Language · Cross-listed from eess.AS
Explainable and Generalisable LLM-based Cognitive Decline Detection with Spontaneous Speech
Ziyun Cui, Wen Wu, Chuan Shi et al. (+10 more)
Alzheimer's disease (AD) and mild cognitive impairment (MCI), which may precede AD, manifest early through subtle linguistic and acoustic alterations. Traditional diagnostics, however, are often resource-intensive and lack scalability for mass screening. To address these challenges, we introduce a novel bilingual
speech large
language model framework for automated, explainable cognitive screening. Unlike conventional pipelines that rely on error-prone automatic
speech recognition, our system directly processes raw
speech to learn joint acoustic-semantic representations, preserving critical prosodic cues often lost in transcription. Utilising our newly collected PUTH-AD
dataset alongside multiple open-source corpora, we implemented a multi-task learning objective that simultaneously performs cognitive status classification and generates clinician-understandable natural
language explanations. Our system achieved the highest average accuracy and AUROC across six
dataset/task conditions, comparing three representative baselines. The system demonstrated cross-task transfer to held-out PUTH-AD task subsets, maintaining classification accuracy on an entirely unseen cognitive task without task-specific fine-tuning. Furthermore, clinician evaluation confirms that the generated explanations are both clinically relevant and largely consistent with the underlying
speech evidence, supporting their potential utility in clinical interpretation. This study provides a scalable, objective, and explainable framework for
speech-based cognitive screening, combining cognitive status classification with natural
language explanations that clinicians can assess and verify, bridging the gap between advanced AI and clinical utility.
Read the full abstract ▾
All authors Ziyun Cui, Wen Wu, Chuan Shi, Shuguang Yang, Xueying Gui, Yan Zheng, Qiong Yang, Haiyan Zhao, Wei-Qiang Zhang, Ji Wu, Yelei Li, Nan Li, Chao Zhang
Artificial Intelligence
Same Winners, Different Success Rates: Evaluating How LLM Agents Recover from Failures
Dong Xu, Zhangfan Yang, Jiantao Wu et al. (+5 more)
Evaluating how LLM
agents recover from mid-task failures is central to deploying reliable agentic systems. Existing checkpoint-based
benchmarks measure recovery by comparing which action is selected as best across independent runs, a quantity known as set agreement. However, set agreement is a purely ordinal measure that records which action wins without reflecting the absolute level of performance. When all actions fail, they tie at zero reward, and independent runs produce the same tied set with high probability, creating an illusion of stability that masks near-zero recovery success. We formalize this limitation through a set-path
symmetry result, proving that for equal-cost Bernoulli actions the success probabilities (0.9, 0.8) and (0.2, 0.1) yield identical best-action-set distributions at every sample size. No procedure based solely on which action wins can distinguish these two regimes. We further prove that certifying exact population ties is impossible in finite time, and that the assignment of outcomes to checkpoints carries information beyond marginal outcome distributions. The pooled success probability is the missing scalar that resolves the ordinal ambiguity. Experiments on 864 frozen RecoveryBench episodes and two planning cohorts totaling 3,456 responses confirm the theoretical predictions. Agreement and held-out quality can move in opposite directions, and permuting checkpoint-to-action bindings changes 8 to 13 percent of
cell-level conclusions. Based on these findings, we propose reporting four diagnostic quantities (agreement, all-zero fraction, held-out success, and pooled success) that expose this failure mode with no additional data collection.
Read the full abstract ▾
All authors Dong Xu, Zhangfan Yang, Jiantao Wu, Shipeng Zhang, Zexuan Zhu, Jiangqiang Li, Jun Zhang, Junkai Ji
Artificial Intelligence
GlyphBench: A Playground for Language-Model Reinforcement Learning
Roger Creus Castanyer, Marc-Alexandre Côté, Matthew James Sargent et al. (+3 more)
We introduce GlyphBench, an environment suite for reinforcement learning (RL) post-training of
language-model
agents, with over 360 tasks spanning diverse games. GlyphBench renders spatial observations as two-dimensional Unicode grids and connects training, evaluation, and trajectory replay through a unified interface designed to support efficient and reproducible research. We use GlyphBench to study how observation interfaces,
reasoning effort, and
agent harnesses affect performance, and how RL configurations shape learning dynamics. Our results show that glyph observations outperform native text and pixels in our Craftax experiments, with further gains on several BALROG environments. RL on 100 GlyphBench tasks improves Qwen3.5-4B on held-out
Reasoning Gym problems, reaching 63.48% accuracy and outperforming the base model, a math-trained baseline, and a code-trained baseline. These experiments provide empirical evidence that
reasoning gains from gameplay can yield stronger transfer than math or code. Together, these results highlight GlyphBench's value as a testbed for systematic research on how
language-model
agents learn, interact, and generalize.
Read the full abstract ▾
All authors Roger Creus Castanyer, Marc-Alexandre Côté, Matthew James Sargent, Augustine N. Mavor-Parker, Glen Berseth, Pablo Samuel Castro
Information Theory · Cross-listed from quant-ph
Monotonicity of the Rényi channel capacity under non-signaling assisted channel simulation
Tina Shariat, Matt Hoogsteder-Riera and Marco Tomamichel
We give an alternative variational proof that non-signaling correlations shared between the sender and receiver cannot increase the Rényi capacity of a classical channel. This monotonicity follows from the data-processing inequality for channel divergences established by Oufkir, Yao and Berta. Our approach works directly with the information-radius representation of the capacity. The constraints defining non-signaling boxes are linear in the box, while the Rényi capacity depends on the channel nonlinearly. We use a Legendre transform to express the capacity as an extremum over a family of affine functionals in which the channel appears linearly. Combining this method with separate arguments for the Shannon capacity and endpoint orders, we recover monotonicity at every order from zero to infinity. We then derive consequences for standard coding exponent functions: non-signaling simulation cannot increase the random-coding or sphere-packing exponent below capacity, or decrease the strong-converse exponent above capacity.
Read the full abstract ▾
Machine Learning
X-MoD: Practical Scaling Laws for Sparse-Depth Routing Beyond Mixture-of-Depths
Bowen Dong, Yilong Fan, Tengyu Pan et al. (+6 more)
Mixture-of-Depths (MoD) enables conditional computation across
Transformer depth by routing only a subset of
tokens through selected layers, but its original one-sparse–one-dense alternation tightly couples total capacity to active capacity and limits sparse-depth scaling. We introduce X-MoD, a scalable sparse-depth architecture that decouples
token sparsity from anchor stride, allowing total parameter count to grow while keeping active-equivalent capacity nearly fixed. To make deep sparse routing trainable, X-MoD combines dense anchors with variance-scaled layer-wise gating and depth-wise
token balancing. To make this regime analyzable and usable, we formulate sparse-depth routing as a conditional architecture-design problem: given compute, context length, and active-equivalent backbone size, how should the routing configuration be chosen? We develop a practical scaling-law framework by fitting X-MoD relative to FLOP-matched dense baselines, yielding an interpretable law that decomposes performance into sparse-capacity gain, sparse-context correction, and anchor-stride interaction. The law predicts validation loss across routing configurations and reveals how context length, model scale, and anchor stride shape sparse-depth performance. We validate the architecture and law through pretraining sweeps, held-out scaling-law prediction, ablations, downstream evaluations, and comparisons with Dense, MoD, and representative MoE baselines.
Read the full abstract ▾
All authors Bowen Dong, Yilong Fan, Tengyu Pan, Yike Zhang, Zhenyu Li, Zijian Zhang, Xuewei Li, Mei Yu, Jianyong Wang
Artificial Intelligence
Behavior-Grounded Semantic Enrichment for Financial Fraud Modeling and Reasoning
Linbo Shao, Huilin He, Yating Lou et al. (+1 more)
In financial fraud detection, rich semantic context can provide important evidence for transaction behavior modeling and fraud
reasoning. However, public real-world financial
datasets often lack rich semantics due to
privacy constraints. Consequently, synthetic
datasets incorporate generated semantics, but at the cost of behavioral realism; textual descriptions for contextual
reasoning remain scarce. We address this gap through a semantic enrichment framework grounded in original transaction behavior to simulate
multimodal financial data. We (1) propose a multi-
agent semantic enrichment framework that generates interpretable financial semantics grounded in transaction behavior through role-specialized
agents and consistency refinement, and (2) newly contribute a valuable
multimodal financial fraud
dataset, MS-FFSD, enriched with structured semantics and textual semantics while preserving real-data-grounded transaction behavior. Furthermore, we systematically analyze the quality and utility of semantic enrichment. Results demonstrate statistical fidelity and framework generalizability, while showing that richer semantics benefit fraud modeling and context-aware LLM
reasoning. Overall, this work advances
multimodal financial fraud research and bridges emerging LLM and multi-
agent capabilities with operational anti-fraud practice. The framework and
dataset are released at https://github.com/AI4Risk/MS-FFSD.
Read the full abstract ▾
All authors Linbo Shao, Huilin He, Yating Lou, Dawei Cheng
Robotics
RLE-Bench: A Qualifying Exam for Coding Agents as Robot Learning Engineers
Haitong Ma, Chenxiao Gao, Rushi Qiang et al. (+2 more)
Coding
agents are beginning to move beyond purely digital tasks to tackle physical-world challenges, particularly in robotics. Existing robotics
benchmarks, however, primarily focus on the performance of individual artifacts, such as policies or controllers, offering limited coverage of coding
agents' broader engineering capabilities. Real-world robotics extends beyond control:
agents must build, integrate, diagnose, and improve heterogeneous artifacts under resource constraints and reason from
multimodal feedback. To evaluate these broader capabilities, we introduce RLE-Bench, a
benchmark of
robot-learning tasks spanning four representative robotics development workflows: interactive control, policy learning, perception and estimation, and mechanical design. We use diverse task-specific metrics to evaluate the artifacts submitted by the coding
agents, from the success rate the
agents achieved to the policy
agents trained, the harness
agent built, and the mechanical structures the
agent designed. We aggregate these metrics into an overall RLE Index and report workflow-specific capability profiles, enabling systematic comparison of coding
agents' capabilities across multiple capability dimensions. Beyond performance ranks, we also conduct in-depth case studies examining
agent behavior on representative tasks, highlighting both current capabilities and limitations, and pointing to the opportunities robotics tasks have to offer for future
agent training.
Read the full abstract ▾
All authors Haitong Ma, Chenxiao Gao, Rushi Qiang, Bo Dai, Na Li
Numerical Analysis
Immersogeometric fluid-structure interaction modeling of the human mitral valve and transcatheter edge-to-edge repair
Keon Ho Kim, Ashton M. Corpuz, Monu Jaiswal et al. (+2 more)
Transcatheter edge-to-edge repair (TEER) treats mitral regurgitation with a clip that grasps and coapts the mitral valve (MV) leaflets. Predicting its outcome requires resolving the interaction between the blood flow and the complete MV apparatus, in which highly deformable leaflets and chordae tendineae undergo large deformation and complex contact. In this work, we introduce an immersogeometric fluid-structure interaction (FSI) formulation in which the coupling among the valvular components, the surrounding flow, and the repair is enforced weakly, without requiring conforming discretizations. A fully parameterized reconstruction pipeline is developed to convert segmented images into analysis-ready isogeometric meshes. The leaflets and chordae are modeled as Kirchhoff-Love shells and cables, and the shell formulation is modified to include an in-vivo prestrain tensor that accounts for the loading already present in the imaged geometry. We propose a subcell approach that improves the quadrature for the interfacial constraint and eliminates the over-refinement otherwise needed to prevent unphysical leakage through the leaflets. The MV repair is modeled by a penalty formulation that enforces leaflet coaptation over a prescribed region, allowing clip number and placement to be varied without remeshing or modeling the device. A lumped-parameter model at the
outflow tract of the left heart imposes a flow-dependent pressure rather than a prescribed waveform, so that the split between forward and regurgitant flow matches physiological behavior. We validate the reconstruction against the segmented imaging data, demonstrate mesh convergence of the proposed formulations, and apply the framework to a prolapsed MV and its single- and double-clip repairs to quantify pre- and post-operative changes in valve dynamics and left-heart hemodynamics.
Read the full abstract ▾
All authors Keon Ho Kim, Ashton M. Corpuz, Monu Jaiswal, Michael S. Sacks, Ming-Chen Hsu
Machine Learning · Cross-listed from stat.ML
Functional Autoencoders for Amplitude-Phase Representation Learning
Peida Wu, Xinyang Xiong and Pengcheng Zeng
Functional data are intrinsically infinite-dimensional, and often exhibit phase variation, where corresponding events occur at different times across observations. Existing linear dimension reduction methods struggle with nonlinear amplitude variation, while functional autoencoders without an explicit warp entangle temporal misalignment with shape. We propose the Amplitude–Phase Functional Autoencoders (AP-FAE), an unsupervised framework for functional data that spans both univariate and multivariate cases, with emphasis on the multivariate setting, and factorizes the latent space into separate amplitude and phase
embeddings derived from all channels. A smooth functional decoder reconstructs channel-specific amplitude functions in canonical time, and a shared monotone, endpoint-preserving warp captures phase variation. We prove a bound linking amplitude recovery to registration, reconstruction, and noise errors, and validate it numerically. Across synthetic data and six real-world
benchmarks, AP-FAE outperforms state-of-the-art baselines on most clustering and
alignment metrics and on all reconstruction metrics. Clustering with amplitude
embeddings alone consistently surpasses joint amplitude–phase clustering, confirming the benefit of explicit disentanglement. Code is available at https://anonymous.4open.science/r/APFAE-418C/}{https://anonymous.4open.science/r/APFAE-418C/.
Read the full abstract ▾
Computer Vision
WorldGuide: Learning Success-Failure Boundaries in Latent World Models for Vision-Language-Action Policies
Lin Liu, Lu Zhang, Ziying Song et al. (+6 more)
Latent world models offer a promising way to improve
Vision-
Language-Action policies by capturing the consequences of actions. However, models trained primarily on expert demonstrations have limited exposure to failure outcomes and may struggle to distinguish visually similar successful and failed interactions. We propose WorldGuide, a framework that learns these distinctions in latent space and uses them to guide policy training. WorldGuide combines predictive pretraining on successful and failed trajectories with contrastive learning on matched success–failure pairs. The learned predictor then provides a differentiable reward to guide joint optimization of the policy and visual encoder. The predictor is discarded after training, so deployment requires no additional world-model inference. Extensive experiments show that WorldGuide substantially improves VLA reliability and achieves state of the art performance on LIBERO 100 and SimplerEnv, reaching 96.8% and 72.0%, respectively. Code will be publicly available.
Read the full abstract ▾
All authors Lin Liu, Lu Zhang, Ziying Song, Wu Yang, Yuzheng Zhuang, Yunzhi Zhuge, Shuai Tao, Wulong Liu, Huchuan Lu
Machine Learning
Learning to Optimize through Solver-Grounded Self-Play
Xia Jiang, Yaoxin Wu, Chenyu Zhou et al. (+3 more)
Optimization modeling is central to many decision-making scenarios, but traditionally requires extensive domain expertise. While Large
Language Models (LLMs) have shown promise in automating this process, current training paradigms mainly rely on human-annotated or teacher-generated
datasets. This dependence introduces a Generalization Ceiling, where models overfit to narrow data distributions, and Capability Anchoring, where models'
reasoning is bounded by annotator proficiency and teacher model capability. In response, we propose OPT-Zero, the first fully self-play training framework for optimization modeling that requires zero external training data. OPT-Zero employs a single LLM in a dual-role closed loop: a Proposer that synthesizes increasingly challenging optimization problems alongside their mathematical formulations and solving code, and a Solver that attempts to resolve the problems given only natural-
language problem descriptions. Grounded in execution feedback from external optimization solvers, we alternately train both roles using reinforcement learning. This process fosters an auto-curriculum in which the Proposer and Solver co-evolve: generating harder valid problems by the Proposer seamlessly enhances the structural
reasoning ability of the Solver. Extensive results indicate that with zero curated data, OPT-Zero matches state-of-the-art data-dependent methods while exhibiting substantially stronger generalizability, establishing self-play training as a highly scalable paradigm for advancing LLM
reasoning in modeling and solving optimization problems.
Read the full abstract ▾
All authors Xia Jiang, Yaoxin Wu, Chenyu Zhou, Mengzhu Xu, Wim P.M. Nuijten, Yingqian Zhang
Robotics
WB-WAM: Heterogeneous Body-Hand Pre-training for Humanoid Loco-Manipulation
Chuan Qin, Shaoting Zhu, Siyuan Luo et al. (+4 more)
Humanoid loco-manipulation demands coordinated body and hand behavior, while conventional
robot pre-training data provide limited coverage of such whole-body motion. We present WB-WAM, a World Action Model that incorporates explicit whole-body action supervision into generative
video pre-training. A shared physical action space integrates body, root, and dexterous hand annotations from heterogeneous sources, enabling joint
video and action learning from 1880.2 hours of partially annotated
video and motion data. The resulting priors are refined through PICO mid-training and adapted to
robot tasks with auxiliary forward kinematics supervision. We construct WB-
Datasets to support these stages with retargeted
egocentric human demonstrations and
robot trajectories, allowing task-aligned human motion to supplement limited
robot data. Evaluations in simulation demonstrate strong whole-body task performance with 81.9% in HumanoidArena, while real-world experiments further validate WB-WAM with 84.0% mean success across five tasks. Moreover, task-aligned PICO mid-training improves downstream task performance while reducing the need for real-
robot demonstrations. These results support heterogeneous whole-body pre-training and human motion transfer as a practical route to data-efficient
humanoid loco-manipulation.
Read the full abstract ▾
All authors Chuan Qin, Shaoting Zhu, Siyuan Luo, Siqiao Huang, Hongyu Zhao, Shanaka Baduge, Hang Zhao