Artificial Intelligence · Cross-listed from cs.SE
JevCoder: When Does Decision Delegation Pay Off in Code Repair?
Jiahao Zhang, Yifan Zhang and Yu Huang
A coding
agent repeatedly makes decisions as it repairs code. These decisions shape repair trajectories and affect the
token cost of successive model requests. We ask which repair decisions can be delegated to a low-cost probabilistic judge to reduce
token use while maintaining repair quality. The judges answer questions with probabilities over predefined answers, without generating explanations. We evaluate judgment quality across repair stages and test how individual interventions change subsequent repair work. We compare these interventions with unguided repair, which uses no judge, and with the same workflow using the repair model itself as the judge. The benefit depends on the work an intervention changes, rather than the cost of the judgment alone. Based on these studies, we propose JevCoder, a workflow built around Jev, a low-cost probabilistic judge. It combines code-snippet recommendations with action-type guidance and uses judgments to determine when to submit. We evaluate three repair models, each on the same 400 tasks: 200 from SWE-bench Verified and 200 from SWE-Bench Pro Verified. Compared to unguided repair, JevCoder reduces mean repair-model
token use by 21.6%-50.6%, with higher observed repair rates across all three models.
Read the full abstract ▾
Artificial Intelligence
L3M: A Verifiable Harness and Runtime for Self-Extending and Self-Improving Agents
Carson Eisenach, Robert Joseph George, Vincent Quenneville-Belair et al. (+2 more)
The lasting gains in AI have come from methods that absorb more computation, and
language models already scale computation within a call. Useful work, though, happens across calls: over hours or days, through tools and external systems, among many
agents and the humans they work for, and increasingly through code the
agents write themselves. Whether that work compounds is enabled by the harness around it, and the harness faces a constraint the model does not: of the three participants in such a computation, the human, the model and code, only the human's
attention does not scale, so the computation can grow only if the human is not in every loop. L3M is a runtime built on that premise, written in Lean 4 so that one
language implements the harness, states its invariants, and is the
language agents extend it in. One computation spans humans,
agents, models and code, and all of its state lives in Git: a rich state trajectory – rather than a chat conversation – is an object the runtime can resume, rewind, fork, compare and merge. Waiting, communication, delegation and undo are primitives, not patterns the model reasons through each time. When an
agent works out a procedure, it can write it once as a sealed Lean program whose type is its permission request, and the harness offers it back as a tool – machinery for the model to use. To take the human out of the loop, L3M uses reversibility to decide where that is safe: a tool's review tier follows from how far its effects can be undone, so
agents act freely where the runtime can recover. Because the machinery can change, the properties that make this safe are stated as kernel-checked
theorems about the code that runs. On Terminal-Bench, an untuned L3M reaches about two thirds of its (unsafe) shell-equipped pass rate using only safe (not needing human review) tools.
Read the full abstract ▾
All authors Carson Eisenach, Robert Joseph George, Vincent Quenneville-Belair, Kate Gwimm, Dean Foster
Artificial Intelligence
JockeyBench: Evaluating LLM Agents on Horse-Racing Decision-Making
Zhongwei Xie, Xingsu Li, Jingqiao Yu et al. (+8 more)
Repeated event forecasting requires extracting predictive signals from noisy evidence and evaluating their value across successive decisions. We introduce JockeyBench, a
benchmark and evaluation protocol that studies this problem through historical Japanese horse racing. The protocol combines time-gated evidence, priced tools, paired frozen-program controls, and two workflows: executing an LLM-written predictor or acquiring evidence for a fresh per-race LLM forecast. Evaluation separates forecast quality,
benchmark-specific net utility, and retrospective betting returns and risk under fixed decision rules. Across 28 model–scaffold configurations on 324 races, paired tests identify seven revision gains and two declines after multiple-testing correction. Yet every configuration has negative mean profit under capped-Kelly settlement. In a four-model comparison, the reusable-program workflow has higher mean betting profit and lower research fees, although both workflows lose on average. Further diagnostics reveal that mean probability bias reverses on selected bets and that profitable individual runs can depend on concentrated returns. These findings motivate evaluating predictive competence, adaptation, resource use, and decision value separately when assessing
agents for repeated event forecasting.
Read the full abstract ▾
All authors Zhongwei Xie, Xingsu Li, Jingqiao Yu, Zihao Wang, Hong ChiTat, Zhaowei Wang, Weizhi Du, Yuxuan Liu, Huihao Jing, Jiaxin Bai, Yangqiu Song
Artificial Intelligence
Who's Behind the Text? Robust Model Authorship Attribution
Minseok Ko, Hyunseok Lee, Woomin Song et al. (+3 more)
As large
language models (LLMs) are increasingly used to generate text, their misuse has become a growing societal problem, with implications for academic misconduct, model
distillation, and cyberattacks. For these reasons, Model Authorship Attribution (MAA), which identifies the source model of a given text, is important for promoting the responsible use of LLMs and improving AI safety. In practice, attribution systems must remain reliable beyond the distributions observed during training. They may encounter texts from unseen domains and writing settings, as well as generations from future versions of model families that were not available when the system was trained. However, existing MAA methods often exhibit substantial performance degradation under such distribution shifts. To address this challenge, we propose Authorship Tracing via Likelihood-based Attribution of Sources (ATLAS), which formulates MAA through source-conditioned text modeling rather than direct source prediction. ATLAS learns how text from each source family is generated and attributes a given text by comparing its likelihood under each learned source condition. We aggregate this likelihood evidence at the
token level with clipping so that no single
token dominates the decision. We further introduce MAABench, a broad MAA
benchmark for evaluating practical OOD settings such as domain and version. Our experiments show that ATLAS consistently outperforms existing baselines across MAABench and OpenTuringBench. In particular, when both the generation domain and the model version change on MAABench, ATLAS achieves 74.27% accuracy, 9.00pp higher than the strongest baseline. Beyond MAA, ATLAS can also help identify
distillation source models by measuring which source model a target LLM most closely resembles. Our extended experiments show that ATLAS recovers the known teacher of every distilled model we evaluate.
Read the full abstract ▾
All authors Minseok Ko, Hyunseok Lee, Woomin Song, Beomjun Kim, Jinwoo Shin, Kunwoong Kim
Artificial Intelligence · Cross-listed from cs.MA
XferAgent: Cross-Agent Experience Transfer for Self-Improving LLM Agents
Weihao Li, Jun Bai, Shijie Chen et al. (+3 more)
Large
language model (LLM)
agents can reuse experience from prior task executions to improve performance on subsequent tasks. Although existing methods enable cross-
agent experience sharing, differences in
agents' execution histories and task requirements make shared experiences difficult to reuse effectively. We propose XferAgent, a framework for cross-
agent experience transfer that organizes reusable experience in local and shared experience graphs. XferAgent reconciles compatible contributions from heterogeneous
agents while avoiding unreliable merges. For each new task, it selects shared experiences and integrates them with the target
agent's local experience to support task execution. Experiments across diverse task domains show improvements over representative
memory-based methods and effective experience transfer between heterogeneous
agent frameworks. Cumulative task success also increases over successive task passes as
agents update and reuse their experiences.
Read the full abstract ▾
All authors Weihao Li, Jun Bai, Shijie Chen, Qincheng Lu, Xinyu Wang, Ziyang Song
Artificial Intelligence · Cross-listed from cs.LG
Learning Across Temporal Graph Representation Silos
Minwoo Yu and Young-guk Ha
Temporal graph data are recorded in different representations, but whether these differences also determine the scope of learned knowledge reuse remains unclear. This study investigates whether predictive structure learned from relational histories can be reused between continuous-time dynamic graphs (CTDGs) and temporal knowledge graphs (TKGs). Bidirectional reuse emerges while retaining distinct input and prediction interfaces and independent vocabularies, and sources with the same representation are not always more useful. Controlled reorganizations of source histories show that the organization contributing to reuse varies across source-target conditions. We show that learned aspects of Relational Dynamics, the predictive structure present in relational histories, can be reused across conventional representation boundaries. Independently pretrained TKG and CTDG learners can also be fused at the score level to improve prediction on a third
dataset without entity or relation-semantic
alignment. Two proofs of concept in recommendation and security detection illustrate that external temporal sources without direct
alignment can also become useful pretraining resources.
Read the full abstract ▾
Artificial Intelligence
Tiantian Peng, Guannan Lv, Hongyang Li et al. (+5 more)
On-policy
distillation (OPD) transfers knowledge from a teacher model to a student along student-generated trajectories, typically by minimizing the reverse Kullback-Leibler (RKL) divergence. However, RKL can provide weak corrective signals for teacher-supported
tokens to which the student assigns very low probability. This raises two questions: which positions are most consequential for the behavioral gap between the student and teacher, and what additional supervision is needed at these positions? Through cross-sampling, we find that divergence-based routing more effectively pinpoints positions contributing to the teacher-student performance gap than
entropy-based routing. Further analysis reveals teacher-mode under-coverage at high Jensen-Shannon (JS) divergence positions, where standard OPD increases the large teacher-student probability gaps. Motivated by these findings, we propose Divergence-Routed On-Policy
Distillation (DrOPD). DrOPD retains RKL supervision at all positions and selectively adds forward Kullback-Leibler (FKL) divergence supervision at positions with high JS divergence. Experiments across six math
benchmarks show that DrOPD improves over standard OPD and
entropy-based routing. DrOPD also reduces severe under-coverage cases by 37.9 percent relative to OPD, supporting targeted FKL as an effective complement to OPD.
Read the full abstract ▾
All authors Tiantian Peng, Guannan Lv, Hongyang Li, Qiuhe Hong, Hang Cheng, Hongjian Dou, Xiangyu Wu, Yuyang Liu
Artificial Intelligence · Cross-listed from cs.CV
Frame Evidence:A Benchmark for Reference Frames and Visual Evidence in Multi-Image Spatial Reasoning
Chunxi Luo, Qishuai Ding, Xingchen Peng et al. (+6 more)
A spatial answer depends on the reference frame specified in
language and its supporting visual evidence. Item accuracy alone obscures whether correctness persists when either changes. We introduce Frame Evidence, a
benchmark of 320 matched triplets and 960 Chinese multiple-choice questions from captured and rendered scenes. One family compares three grounded reference frames over fixed images. The other pairs image deletion annotated as insufficient with answer-preserving addition. Joint metrics require correct answers across all three frames and paired evidence variants. Across 20 configurations on 288 held-out questions, GPT three-shot reaches 50.00% item accuracy, but the best three-frame group accuracy is 5/48 (10.42%), reached by GPT two-shot. Few-shot prompting and task-specific adaptation improve item accuracy without consistently improving joint correctness. Input controls show where images support matched success, while repeated executions reveal answer instability beneath similar scores.
Read the full abstract ▾
All authors Chunxi Luo, Qishuai Ding, Xingchen Peng, Hengwei Ye, Ke Zhang, Wenxin Lian, Tianyi Lu, Chunlong Yu, Zhenya Lin
Artificial Intelligence · Cross-listed from cs.CL
REWIND: Uncertainty-Guided Trajectory Repair for Test-Time Reasoning
Yi Ding, Mengyu Wang, Min Zhou et al. (+1 more)
An early mistake in the generated
reasoning can constrain all subsequent
reasoning. Repeated-generation methods, like MCMC-based approaches, attempt to improve output quality by sampling multiple generations, but their effectiveness can be limited by randomly selecting where to regenerate. On MATH500 with Qwen2.5-Math-7B, we find that incorrect responses have substantially higher mean
token entropy than correct responses, with an average within-question relative increase of approximately 112%, suggesting that
entropy can signal
reasoning errors. Motivated by this finding, we introduce REWIND, a training-free controller that uses
entropy dynamics to propose targeted suffix regenerations. Across three model families and five
benchmarks, REWIND outperforms all baselines, with gains of 0.51-10.00 percentage points over the strongest baseline. These results demonstrate the effectiveness of targeted, uncertainty-guided trajectory repair across models and tasks. Our code is available at https://anonymous.4open.science/r/rewind-D310.
Read the full abstract ▾
All authors Yi Ding, Mengyu Wang, Min Zhou, Menglin Yang
Artificial Intelligence
$\tau$-Emotion-Bench: Exposing the Interaction Fragility of LLM Agents through Real-World User Emotions
Rishitosh Kumar Singh, Amir Saeidi, Shamanthak Hegde et al. (+2 more)
In long-horizon tool-calling tasks, LLM
agents can fail not only because of tool-use errors, but also because of ineffective interaction with users. Yet existing
benchmarks largely assume cooperative and behaviorally stable users, leaving the effect of emotional variation on
agent reliability underexplored. We introduce $\tau$-Emotion-Bench, a ten-domain
benchmark for evaluating tool-calling
agents under controlled emotion-conditioned user interactions while preserving the underlying task objective. To scale both
benchmark construction and
agent training, we introduce Tracer, a scalable agentic data-generation pipeline that constructs grounded, policy-valid tasks and collects trajectories through an interactive
agent–user–environment harness. By allowing dissatisfied users to terminate inefficient interactions, the harness provides an interaction-aware signal that filters trajectories with excessive user dependence and favors effective tool use. Experiments with two user simulators show that frontier
agents remain unreliable, with substantial gaps between average task success and consistent task completion. Tracer-generated supervision consistently improves Qwen3 models across scales, generalizes to unseen domains, and transfers to $\tau$-Bench and $\tau$-Trait. Behavioral analyses further show that higher task success is associated with a better balance between tool execution and necessary user follow-up. Overall, $\tau$-Emotion-Bench and Tracer provide a scalable framework for evaluating
agent reliability under diverse user behaviors and improving tool-calling
agents through interaction-aware supervision.
Read the full abstract ▾
All authors Rishitosh Kumar Singh, Amir Saeidi, Shamanthak Hegde, Shrinidhi Kumbhar, Chitta Baral
Artificial Intelligence · Cross-listed from cs.LG
TEX-MEX: Similarity-Guided Temporal Contrastive Learning for Time Series
Aditya Radhakrishnan, Maxwell A. Xu, Wanting Mao et al. (+3 more)
Contrastive learning for time series depends on deciding which pairs of inputs should be treated as similar and dissimilar. Because few augmentations reliably preserve meaning in time series, prior methods have relied on two disjoint notions of similarity between windows: structural similarity, which relates windows with similar morphology, and temporal similarity, which relates windows that occur close together in time. We introduce TEX-MEX, to our knowledge the first approach to unify the two into a single relationship. Given a random batch of windows drawn across the
dataset, TEX-MEX uses the encoder to identify which windows are structurally similar or dissimilar to each other. It then pulls each window toward the temporally shifted versions of similar windows and pushes it away from those of dissimilar ones, learning what typically accompanies similar windows in time. Across PPG,
EEG, ECG, accelerometry, and continuous glucose monitoring, TEX-MEX broadly achieves state-of-the-art performance against contrastive, reconstructive, JEPA-based, and modality-specific approaches, establishing it as a strong general SSL objective. We further train a 560K-parameter PPG foundation model that achieves state-of-the-art performance across a broad range of downstream
benchmarks while being roughly 50x smaller than the next-best unimodally pretrained PPG foundation model in our evaluation. We will release this foundation model open-source and open-weight to the community to support future work and practical applications.
Read the full abstract ▾
All authors Aditya Radhakrishnan, Maxwell A. Xu, Wanting Mao, Santosh Kumar, Harish Haresamudram, James M. Rehg
Artificial Intelligence · Cross-listed from cs.CL
Evidence-Anchored Overdose Death Prediction from Death Investigation Reports
Raavi Gupta, Ziqing Yang, Karli R. Hochstatter et al. (+8 more)
Confirmation of overdose deaths in the United States is often delayed due to the complexities of certifying drug-involved deaths. Death investigation reports represent a rich but underutilized data source that could be used to predict whether a death is a suspected overdose (i.e., a death preliminarily identified as an overdose prior to the final cause and manner of death determination). We present an automated framework that classifies suspected overdoses from death investigation reports, while producing supporting explanations guided by expert knowledge. Through an expert-guided design process, we investigate 1) what type of explanations are best for model prediction performance, and 2) which architecture is better suited for this task: pipeline models that rationalize first and then predict, or self-rationalizing models that jointly predict and rationalize? Our experiments show that a pipeline model with explanations grounded in the input death investigation reports outperforms the other settings. Our best model also outperforms a rule-based system currently in deployment by 15%, while allowing expert interventions for final decisions.
Read the full abstract ▾
All authors Raavi Gupta, Ziqing Yang, Karli R. Hochstatter, Talia Nadel, Nicole D'Anna, Jason K. Graham, Hannah Johnson, Kelsey Opozda, Nabila El-Bassel, Anubhav Jangra, Smaranda Muresan
Artificial Intelligence
LabCompiler: A Framework for Verifiable Human-Machine Laboratory Execution from Imperfect Protocols
Jingyuan Li, Dongzhan Zhou, Minting Pan et al. (+10 more)
Closing the loop of AI-driven science requires reliably translating natural-
language protocols into executable laboratory workflows despite protocol defects and heterogeneous hardware. Existing approaches often couple scientific interpretation with device-specific code generation, making errors difficult to localize and limiting reuse across laboratory platforms. We introduce LabCompiler, a robust protocol-to-execution framework that transforms imperfect natural-
language protocols into verifiable human-machine workflows. Its protocol firewall diagnoses and repairs defects before compilation while preserving unaffected content, guided by troubleshooting knowledge accumulated through perturbation-guided self-
evolution. A two-level, device-agnostic graph representation separates experimental semantics from hardware-specific implementation and supports capability-based device assignment with coordinated human execution. We further introduce a protocol recovery
benchmark covering four controlled perturbation types and naturally occurring errors collected from protocols written by iGEM participants. Experiments show that LabCompiler outperforms frontier LLMs in protocol error correction while enabling coordinated human-machine execution across heterogeneous, partially automated laboratory settings.
Read the full abstract ▾
All authors Jingyuan Li, Dongzhan Zhou, Minting Pan, Yujie Liu, Yuan Gao, Yankai Jiang, Xinyu Gu, Zhe Liu, Yang Li, Bo Zhang, Lei Bai, Tong Si, Weinan Zhang
Artificial Intelligence · Cross-listed from cs.CL
Typed Decision Models on Tabular Data
Baowen Zhang, Hangting Ye and Hongyuan Zha
Typed decision models are
language models that answer a question with a fixed set of possible answers by returning a probability for each answer instead of generating text. We study them on tabular data and compare four decision models with three large
language models (LLMs) on row classification, fact verification and anomaly detection. Every request is sent in its original form and under 21 types of rewrite that keep its content, such as renaming the options or writing the table in another format. The decision models perform as well as the LLMs at a lower cost. Three of them are also far more robust to how the question is formulated. No model we test is robust to how the table is organized. For every model, at least a third of the row decisions change under some rewrite of the table or the question. The decisions to which a decision model gives low probability are the ones most likely to change. Decision models are a practical choice for tabular decisions, but their robustness remains an open problem.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CL
Retrieve What Helps, Only When It Helps: Agentic Reinforcement Learning for Selective Retrieval-Augmented Machine Translation
Shijun Li, Yang Liu, Najmeh Sadoughi et al. (+5 more)
Retrieving external knowledge can improve large
language model (LLM) translation by providing supplementary information such as idiom interpretations and terminology. However, existing
retrieval-augmented machine translation systems typically apply predefined
retrieval strategies uniformly across all inputs. We argue that translation operates in a partial-evidence regime: LLM's parametric knowledge suffices for many inputs, while external evidence benefits a subset and usually supports rather than directly provides the desired translation. This creates two challenges: deciding when to retrieve and what evidence to acquire. We introduce Selective
Retrieval-Augmented Translation (SRAT), an agentic reinforcement learning framework that learns when to translate directly or retrieve, what to retrieve, and how to integrate these decisions by optimizing final translation quality. To align evidence
retrieval with downstream translation quality rather than simple semantic similarity, we further develop joint Plackett–Luce optimization and translation-quality contrastive training for retriever-training. Experiments on idiom, knowledge-intensive, and document-level translation with offline or online
retrieval show that SRAT consistently outperforms never- and always-retrieve baselines, and quality-aligned retriever training provides further gains when similarity-based
retrieval is insufficient. The results demonstrate the benefit of treating
retrieval as a learnable translation action rather than fixed preprocessing.
Read the full abstract ▾
All authors Shijun Li, Yang Liu, Najmeh Sadoughi, Dongmei Wang, Xinjie Li, Zhu Liu, Rohith Mysore Vijaya Kumar, Yuzong Liu
Artificial Intelligence
Revisiting Triplet Contrastive Learning for 3D Shape Understanding
Chaoyang Zhu, Zhe Liao, Jiayi Zhang et al. (+3 more)
In multi-modal 3D pre-training, aligning 3D shapes, their 2D image counterparts, and
language descriptions has made tremendous progress. However, we find that supervision from rendered images is unreliable, the semantics encoded in synthetic views are distorted due to domain gap, self-occlusion, and rendering artifacts. Instead, we learn transferable 3D representations from
language only. We observe that, in real world, 3D shapes are organized in a tree-like structure, ranging from general concepts to fine-grained instances with specific attributes, while
language typically denotes generic concepts and subsumes these various 3D instantiations. Hence, we propose to embed 3D representations in
hyperbolic space, whose geometry is well-suited for modeling tree-like data without distortion. To capture the asymmetric hierarchy between
language and 3D shapes, we further impose a partial-order relation that
language entails 3D shapes. Our method,
Hyperbolic Contrastive Learning (HyperCL), jointly optimizes a
hyperbolic geodesic-based contrastive objective and an entailment loss to learn discriminative cross-modal representations while explicitly preserving this hierarchy. We also introduce a rigorously verified
benchmark for comprehensively evaluating cross-modal 3D
alignment. Across two widely used
benchmarks and ours, we improve the performance up to 4.6% absolute percentage on 3D shape classification and 5.5% on cross-modal text-to-3D
retrieval under the same parameter budget. The code and evaluation
benchmark will be made publicly available.
Read the full abstract ▾
All authors Chaoyang Zhu, Zhe Liao, Jiayi Zhang, Jie Li, Jiayi Ji, Liujuan Cao
Artificial Intelligence
VLRLA: Deployment-Time Adaptation of Vision-Language-Action Models through In-Context Reinforcement Learning
Diyuan Shi, Yiqi Tang, Hongyin Zhang et al. (+2 more)
Vision-
language-action (VLA) models connect visual perception and
language understanding with
robot control, providing a foundation for instruction-conditioned manipulation. However, conventional pretrained VLA policies lack a mechanism for gradual adaptation through accumulated interaction experience during deployment. We therefore propose VLRLA, a general framework that equips existing VLA models with in-context reinforcement learning capabilities for deployment-time adaptation. Our approach augments a VLA model with a reinforcement learning module that learns from historical observations, actions, rewards, and termination signals. During deployment, the policy gradually refines its action generation and progressively improves task success while keeping model parameters fixed. Our approach is also designed for general applicability and compatibility with diverse learning history construction strategies and model training pipelines. Finally, through extensive experiments, we systematically investigate the learning properties of VLRLA in simulation and demonstrate its ability to adapt to unseen scenarios during real-world deployment.
Read the full abstract ▾
All authors Diyuan Shi, Yiqi Tang, Hongyin Zhang, You Wu, Donglin Wang
Artificial Intelligence
Structure Before Intelligence: Gating What Agentic Scientific Workflows Release
Ashish Acharya, Aman Al Amin and Hong Qin
An agentic scientific workflow can run to completion and still release wrong results: a silent numerical error, a stale input, or a defect in a shared upstream step can contaminate several outputs while a sibling branch stays correct. We treat trust in such results as a release decision. A harness around the workflow's directed acyclic graph (DAG) decides which outputs may be used: it enforces mandatory checks, contains failures to the affected branch, recomputes from tracked versions, adds validators, and releases a result only if a risk model estimates its probability of being invalid as low.
Language-model
agents work inside the harness, choosing corrective actions and reviewing releases, but cannot override it. On 60 held-out paired instances of two workflows with injected faults, we switched components on and off. Each supplied a different property. Containment and recovery raised the number of correct results. Validation supplied safety: fixed rules without validation released 37 invalid results, while validating every candidate released none. Risk estimation supplied efficiency, matching that safety with 31% fewer validators at somewhat lower yield, although better targeting than random validation was not shown. Lineage evidence helped on one workflow. The
language model added no measurable benefit: whether invalid results escaped depended on the harness, not the model. This proof of concept points to a practical path: placing
agents inside explicit release-control structure can keep released scientific results trustworthy, and it gives more capable
agents a measurable baseline to improve on.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from cs.CL
Structure-Preserving Document-Level Data Augmentation via Discourse Parsing
Ruilin Shang, Shilong Hou, Zi Long et al. (+1 more)
Data augmentation (DA) plays a vital role in improving model performance in low-resource or limited-supervision scenarios within Natural
Language Processing (NLP). Existing DA methods, such as synonym replacement and back-translation, have demonstrated effectiveness on the word or sentence level; however, they frequently neglect discourse-level coherence and logical flow, which are essential for complex tasks dependent on inter-sentential relationships. In this paper, we propose a structure-preserving document-level data augmentation framework based on large
language models (LLMs) and document-level discourse structure parsing. Our approach identifies rhetorical relations between sentence pairs and extracts key phrases, which are then replaced with semantically coherent but topically shifted content while preserving the original discourse structure. Experimental results on text summarization and question answering show that the proposed augmented data provides complementary supervision and improves combined-data training across the evaluated tasks. Comparisons with a standard RST parser and controlled ablations further support the proposed discourse representation.
Read the full abstract ▾
All authors Ruilin Shang, Shilong Hou, Zi Long, Xianghua Fu
Artificial Intelligence · Cross-listed from cs.CV
Robust Real-Time Monocular Human Pose Estimation with Insole-Based Sensor Fusion
Liang Yu, Tao Zhu, Xiangqing Lu et al. (+2 more)
Real-time monocular human pose estimation is an important sensing component for human-centered intelligent systems. While recent camera-based methods have achieved impressive accuracy, fundamental challenges persist, particularly when visual observations are incomplete, such as during lower-body occlusion or short out-of-frame intervals. In these challenging but common scenarios,
vision-only systems often struggle to reconstruct physically plausible lower-limb motion and reliable foot-ground contact. Due to the highly unobtrusive and wearable nature of smart insoles, this paper investigates whether simple plantar pressure and acceleration sensors can effectively resolve these visually induced limitations. To systematically investigate this, we first collect a synchronized
multimodal dataset and then propose a part-decoupled temporal fusion framework. The proposed model fuses foot-sensing features into the lower-body branch to compensate for incomplete visual evidence, while a lightweight coordinator exchanges information between the upper- and lower-body temporal representations to maintain whole-body consistency. The framework is evaluated with three representative real-time visual backbones based on 2D keypoints, 3D landmarks, and SMPL-based human mesh recovery. We evaluate our framework on three representative real-time visual backbones. Quantitative experiments demonstrate that incorporating simple foot sensors consistently improves pose accuracy, motion smoothness, and gait-related measurements over
vision-only baselines, with the most significant gains observed under occlusion and short visual-dropout conditions. Ultimately, our results demonstrate that unobtrusive foot sensing provides vital and complementary physical observability when camera observations are temporarily unavailable.
Read the full abstract ▾
All authors Liang Yu, Tao Zhu, Xiangqing Lu, Wenjian Luo, Liming Chen
Artificial Intelligence · Cross-listed from cs.RO
IMU-Bench: Benchmarking Wrist IMU Representations Across Kinematic, Semantic and Clinical Tasks
Yiquan Li, Aidan Bradshaw, Chenfeng Gao et al. (+1 more)
Wrist-worn inertial measurement units (IMUs) provide a common sensing modality for tasks ranging from activity and gesture recognition to pose estimation, inertial odometry and clinical motor assessment. Existing wearable
benchmarks evaluate performance across multiple tasks, but leave open how supervision for one task affects transfer to others and whether these tasks benefit from a shared representation. We present IMU-Bench, a
benchmark of ten tasks from eight
datasets for studying IMU multi-task performance and cross-task transfer. We compare six self-supervised methods pretrained on unlabeled recorded and simulated wrist motion, evaluating classification, pose and velocity estimation, and IMU-
language retrieval. We find that our masked autoencoder, InrtlWrist, achieves the best or joint-best frozen-probe performance on nine of ten tasks, and fine-tuning improves on training from scratch on nine of ten tasks, including full-body pose, odometry and
retrieval. Across transfer and joint-training experiments, supervision that benefits one target can reduce performance on others, with supervised sources transferring within kinematic estimation while, under frozen probing, none approaches the pretrained encoder on
language retrieval, and joint-training gains that come at a cost to the paired task. These findings distinguish broad downstream usefulness from complementary task supervision. Code, task cards and a leaderboard are publicly available.
Read the full abstract ▾
All authors Yiquan Li, Aidan Bradshaw, Chenfeng Gao, Karan Ahuja
Artificial Intelligence
Beyond Retrieval: Evidence Access and Event Aggregation in Agent Memory
Zerui Yang, Quan Xue, Chaorui Zhang et al. (+8 more)
Questions about an
agent's history often require aggregating events, which
retrieval alone does not resolve. We combine source-preserving evidence access with executable aggregation. Trajectory
Memory (TM) retains the original trajectory and constructs a source-addressable reading view at question time. TallyMem augments it with a reusable event ledger, model-generated event specifications and deterministic execution. With a commit-specific adaptation on 776 questions from 99 held-out MINTEval repositories, TallyMem raises Qwen3-32B exact match from 35.2% to 49.0%, a 13.8-point gain over TM; gains span 12.8-13.9 points across three readers. A controlled probe with event sets parsed independently from raw logs shows an 11.0-point gain, from 82.8% to 93.8%, without routing. On full AMA-Bench, the routed system yields a smaller 1.3-point gain, from 69.8% to 71.1%. These results support executable aggregation on tasks that require combining events, with benefits dependent on the task and event specification, the largest residual system-error category in our diagnostic analysis.
Read the full abstract ▾
All authors Zerui Yang, Quan Xue, Chaorui Zhang, Xueyan Niu, Xuanyu Zheng, Xiyuan Wang, Haolin Ruan, Guangcheng Song, Weixi Zhang, Wei Han, Bo Bai
Artificial Intelligence
MemCode: Discrete Semantic Representations for Long-Term Agent Memory
Xin Li, Liang Hu, Duoqian Miao et al. (+2 more)
Long-term
memory remains a fundamental bottleneck for large
language model (LLM)
agents across sessions. Existing systems typically adopt a write-then-retrieve paradigm, storing context-dependent dialogue fragments and retrieving them via nearest-neighbor search in a continuous
embedding space. This design yields unstable access units and systematically misses logically related but lexically distant evidence. We propose MemCode, a long-term
agent memory framework that couples
retrieval-oriented writing with discrete associative
retrieval. MemCode rewrites dialogue into self-contained atomic
memory units (AMUs), maps AMUs into a learnable codebook with isomorphic semantic quantization (ISQ), and uses codebook co-occurrence
topology as a
retrieval path complementary to vector search. For long-term
memory stores, MemCode separates online atomic writes from asynchronous offline consolidation. On LongMemEval-S and LoCoMo with GPT-4o-mini and Qwen3-30B-A3B-Instruct-2507 backbones, MemCode achieves the best overall accuracy while keeping per-dialogue
memory construction within roughly 40~130 seconds, making it 3~21 times faster than LightMem and over an order of magnitude faster than A-MEM and MemoryOS.
Read the full abstract ▾
All authors Xin Li, Liang Hu, Duoqian Miao, Ming Peng, Qi Zhang
Artificial Intelligence
PFedMA: Who Should Join Next as Federated Multi-Agent Orchestration Evolves?
Shuai Li, Shengyuan Liu, Fan Qi et al. (+1 more)
Federated Multi-
Agent Systems (FedMAS) offer a promising paradigm for collaborative
reasoning among LLM-based
agents while keeping raw client data decentralized. However, reliable multi-round collaboration remains challenging.
Agent selection must take into consideration evolving task requirements and the
agent's historical reliability, while
privacy rules must adapt to changing collaboration contexts and meet the client's personalized needs. To address these challenges, we propose PFedMA, a
federated multi-
agent framework for feedback driven multi-round collaboration under heterogeneous
privacy requirements. Across rounds, PFedMA maintains historical capability and reliability states to guide
agent selection, evaluates task responses before aggregation, and refines subsequent queries according to accumulated collaboration feedback. PFedMA enables bidirectional local to global and global to local private rule
evolution across heterogeneous clients. Clients generalize locally updated
privacy rules and selectively incorporate globally maintained rules, thereby facilitating cross-client
privacy rule
evolution while preserving personalized constraints. Extensive experiments across
reasoning,
privacy sensitive, and
adversarial settings show that PFedMA improves task performance, reduces measured
privacy leakage, and improves system robustness.
Read the full abstract ▾
All authors Shuai Li, Shengyuan Liu, Fan Qi, Xiaoshan Yang
Artificial Intelligence · Cross-listed from cs.LG
MIPS: Mitigating Mode Collapse in Reinforcement Learning for Large Outcome Spaces
Armaan Sandhu, Abhijeet Sinha and Dianbo Liu
Expected-reward maximization in reinforcement learning concentrates probability mass on a few outcomes, even in tasks that require many distinct, high-quality solutions, a failure known as mode collapse. Inverse Probability Scaling (IPS) counters this in group-based policy gradients such as GRPO by dividing each reward by the outcome's probability, estimated from its frequency within the sampled group. When the outcome space is large relative to the group, outcomes rarely repeat and these estimates become uninformative. Since an outcome can be constructed along many trajectories, its probability is a sum over all of them, and computing it directly is intractable. We propose Marginal IPS (MIPS), which performs marginal evaluation by combining the sampled trajectory with a learned model of the trajectories that lead to each outcome. On Hypergrids, where the exact sampling distribution is computable, MIPS stays near the reward-distribution target as the outcome space grows relative to the rollout group, while frequency-based IPS degrades. In molecular synthesis and phylogenetic inference, MIPS sustains diverse, high-quality sampling where GRPO and frequency-based IPS collapse, showing that MIPS serves as a reliable solution to mode collapse in RL.
Read the full abstract ▾
Artificial Intelligence
AdCoEvo: A Co-Evolving Multi-Agent Advertising Delivery System
Tianzhu Hu, Hang Yu, Zeyuan Zhao et al. (+9 more)
Digital advertising is an important means for platforms to reach potential users and drive business growth. Advertising delivery is the core link between marketing objectives and actual returns. Existing methods have achieved partial automation in individual tasks such as bidding and budget allocation, but they still rely on human operators for overall coordination, making it difficult to support large-scale operations. They also often fail to achieve optimal performance, and individual experience remains difficult to turn into reusable operational capabilities. In addition, as advertising
markets change, advertising delivery systems need timely strategy updates to remain effective in dynamic
markets. To automate the entire advertising operations process and improve adaptability to
market changes, we propose AdCoEvo, a co-evolving multi-
agent advertising delivery system. Specifically, the system connects delivery decisions across different timescales and operational granularities through role specialization. Diagnostic results guide account planning and strategy adjustment in the next period. The multi-
agent system updates its model parameters based on actual delivery outcomes and continually improves delivery strategies through mutual feedback among planning, adjustment, and diagnosis, thereby achieving co-
evolution. Experiments show that, relative to the rule-based operational baseline (Rule), AdCoEvo increases advertiser value by 14.02% while reducing cost per acquisition by 7.07%. Compared with a multi-
agent system with fixed parameters, AdCoEvo increases advertiser value by 3.96% and reduces cost per acquisition by 1.24%, indicating that co-
evolution can yield greater delivery returns.
Read the full abstract ▾
All authors Tianzhu Hu, Hang Yu, Zeyuan Zhao, Zeyu Pan, Huan Luo, Chun Gan, Mingzhe Cao, Liubin Wang, Mingming Li, Jiao Dai, Jizhong Han, Songlin Hu
Artificial Intelligence
Are You Reading the Right Thing? Rethinking Internal Signals for Prompt-Injection Detection
Sarthak Gupta and Priyanka Tembey
Several recent studies propose detecting
prompt injections from a
language model's internal state. However, these methods do not measure a single security property. Residual-stream probes,
attention scores, role projections, exposure probes, and concept readouts may capture injection presence, recognition, information routing, or behavioral influence. It is therefore unclear what differences in
prompt-injection detection performance reveal about the internal signals these methods read. We study how five representation-reading methods vary under a common structured indirect
prompt-injection setup on Qwen3.6-35B-A3B and Gemma-4-26B-A4B-it. We evaluate each method by how well it distinguishes injected
prompts from benign controls. We then further analyze each detection method's ability to detect attacks in cases where they are successful and in cases where they are not. This helps us better assess what properties each detection method might rely on. Methods trained to detect injection presence or attack-related concepts generally separate unsuccessful injections more strongly from benign inputs, whereas
attention diversion, exposure, and role-confusion-based methods separate successful injections more strongly on both models.
Attention Tracker provides the strongest consistent successful injection separation, reaching 0.949 AUROC on Qwen and 0.887 on Gemma. These results suggest that signals closer to response formation may better encode information about an attack's success, and that detection performance can depend strongly on the
prompt template. We share the data associated with this study at https://anonymous.4open.science/r/paper-evaluation-data-7877
Read the full abstract ▾
Artificial Intelligence
xLightGCN: Explainable Graph Neural Network Based Recommendation Systems with Most Influential Peers
Calder Johnson and Robin Cohen
Recommendation systems, systems that predict a user's preference for items in order to curate that user's feed, are a ubiquitous and central feature in online social networks. While they are a central piece of our modern digital infrastructure, many modern approaches are opaque black boxes, making them difficult to audit, understand, and assess the trustworthiness of. In this paper we present an exact, graph-based approach for generating explanations for decisions made by recommendation systems. We argue that in the risk-sensitive sociotechnical context of user-facing AI such as recommendation systems, explainability is a key pillar for establishing the trustworthiness of models and validating their effectiveness. We propose a technique built on LightGCN like architectures, an influential and popular class of recommendation models. We describe our approach and argue that it provides an appropriate assurance mechanism for these AI systems leading to specific advantages, including interpretable explanations and a theoretically guaranteed measure of explanation fidelity. We conclude by discussing limitations and directions for future research.
Read the full abstract ▾
Artificial Intelligence
YingQiao Wang, Eric Bigelow, Tomer Ullman et al. (+2 more)
Multimodal language models often need to reason about the physical dynamics of the world in order to answer questions appropriately. Various methods have emerged for solving physical
reasoning tasks, and in particular, two leading approaches are Chain-of-Thought
reasoning and physics-grounded simulation. However, relatively little is known about which particular problems can be solved by each approach. Some tasks may benefit from computationally expensive generative
video simulations, whereas others may be solved equally well (or even better) with
language-based
reasoning. Our work addresses this by systematically comparing simulation and
language reasoning across a variety of
multimodal tasks involving physical
reasoning, such as answering questions about fluid dynamics and elasticity. We draw a parallel with this to resource-rational theories of human cognition, which propose that people optimally trade off simulation and
reasoning depending on task demands. We present a novel cognitively-inspired hybrid model which optimally trades off simulation and
reasoning to improve model performance and computational efficiency. Our model is able to decide the appropriate method to solve a physical
reasoning problem at test time, and finds an optimal balance between accuracy and cost. Further, our model can be trained on a smaller
vision-
language model that is only 4B parameters and deployed to improve performance on much bigger base models with $>$3T parameters like Gemini 3 Pro.
Read the full abstract ▾
All authors YingQiao Wang, Eric Bigelow, Tomer Ullman, Yujin Tang, Sebastian Risi
Artificial Intelligence
Qianhong Guo, Wei Xie, Xiaofang Cai et al. (+3 more)
Static
benchmarks for large
language model (LLM)
agents face saturation and data contamination risks, motivating continual updates to evaluation tasks. However, aggregate scores and success rates offer limited insight into specific problem-solving bottlenecks and provide limited guidance for subsequent task design. We propose EvoProbe, an evaluation-driven
benchmark evolution framework that integrates task generation, diagnostic evaluation, and feedback-driven regeneration into a closed loop. It combines verifiable subtask outcome scoring with evidence-supported multi-judge trajectory evaluation and distills phase-level bottlenecks, success patterns, and failure patterns into diagnostic feedback for generating new executable tasks. Across two rounds of cybersecurity experiments, evaluations of eight models on 66 tasks reveal phase-level performance differences and common failure patterns beyond aggregate scores. In a cross-round comparison with the Examiner model set and solver configuration held fixed, all eight models achieve lower Outcome scores on the updated tasks, with relative decreases of 11.8% in mean Outcome and 40.0% in mean Pass\textasciicircum{}3. These results support using fine-grained diagnostics to guide iterative task updates and probe
agents' current capabilities and limitations.
Read the full abstract ▾
All authors Qianhong Guo, Wei Xie, Xiaofang Cai, Shuoyoucheng Ma, Chengxin Hu, Baosheng Wang