跳到正文
研踪PaperTrail

YOUR RESEARCH, IN SIGHT

让下一篇重要论文，不再错过。

覆盖计算机全学科，以 AI 为快捷入口。关注方向、作者和关键词，把发现变成持续的研究积累。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+0新增论文
+0其中 AI 相关
+1已有论文修订
+100新增 AI 导读

arXiv 最近完成扫描：2026-11-09 03:03 · 全站最近入库：2026-11-08 06:02各信源进度 →

统计口径与数据新鲜度

统计 2026-11-08 03:23 — 2026-11-09 03:23（北京时间） · 点击数字查看对应简报；AI 相关包含在新增论文内。

最近完成检查：2026-11-09 03:03最近发现新增版本：2026-11-08 06:02库内论文最新更新时间：2026-11-06 02:59

每分钟刷新已保存的数据。空检查不代表采集异常；新收录可能包含历史论文。AI 导读按新增版本计数。

17,100 已收录论文14,661 AI 相关4,057 有修订112 HF 社区发现

浏览论文库

研究日报 / 周报 →按信源浏览 →

时间范围跟随排序：版本更新、本站收录或首次发表。最新收录可能包含历史论文。部分信源异常，查看状态 →

找到 3,292 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-09 03:01（北京时间）
版本更新 · v2arXiv:2609.08789

Silent Revision: Measuring Undisclosed Change in the Safety Frameworks of Frontier AI Developers

Louis Yiven Zhu

原文摘要

Frontier AI developers publish safety frameworks that commit them to evidencing whether their models are dangerous. The European Union and California now treat these documents as instruments of accountability and impose duties on their revision. Neither requires the developer's account to identify each change to a commitment. We introduce the silent revision rate, the share of material changes to a framework's commitments that the developer's published account does not identify, and we release the versioned, hash-pinned corpus needed to compute it. The corpus holds every public version of the safety frameworks of the twelve developers that have published one, with each provider's changelog, redline or announcement. We trace 710 commitment instances across twelve consecutive version pairs, code them against a frozen codebook, and check the coding with three independent re-runs, a blind re-coding of every disputed row and a second coding of a stratified sample. First, 64% of material changes (95% CI 59 to 69) are silent under a strict standard and 30% under a lenient one. Second, the form of the account matters less than its granularity. Narrative and itemised accounts differ by eight points that eight pairs cannot separate from chance, whereas the rate falls with the number of discrete items a provider publishes. Third, 75% of traced changes weaken or remove a commitment, and weakenings are silent more often than strengthenings, at 70% against 52%. A justification explains why a framework changed and an enumeration states what changed, and only the latter makes revision auditable. We argue that publication duties should carry an enumeration duty, which one provider already meets, voluntarily and incompletely.

cs.CY · cs.AI · cs.SE

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2508.13537

EAvatar: Expression-Aware Head Avatar Reconstruction with Generative Geometry Priors

Shikun Zhang, Cunjian Chen, Yiqun Wang, Yong Li

原文摘要

High-fidelity head avatar reconstruction plays a crucial role in AR/VR, gaming, and multimedia content creation. Recent advances in 3D Gaussian Splatting (3DGS) have demonstrated effectiveness in modeling complex geometry with real-time rendering capability and are now widely used in high-fidelity head avatar reconstruction tasks. However, existing 3DGS-based methods still face significant challenges in capturing fine-grained facial expressions and preserving local texture continuity, especially in highly deformable regions. To mitigate these limitations, we propose a novel 3DGS-based framework termed EAvatar for head reconstruction that is both expression-aware and deformation-aware. Our method introduces a sparse expression control mechanism, where a small number of key Gaussians are used to influence the deformation of their neighboring Gaussians, enabling accurate modeling of local deformations and fine-scale texture transitions. Furthermore, we leverage high-quality 3D priors from pretrained generative models to provide a more reliable facial geometry, offering structural guidance that improves convergence stability and shape accuracy during training. Experimental results demonstrate that our method produces more accurate and visually coherent head reconstructions with improved expression controllability and detail fidelity.

cs.CV · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02956

Specification-Aware Predictive Uncertainty Decomposition

Roman Vashurin, Mikhail Seleznyov, Assel Yermekova, Fedor Chernogorskii, Danila Zykov, Elena Tutubalina, Alexander Panchenko, Preslav Nakov 等 10 位作者

原文摘要

Classical decompositions of predictive uncertainty, such as the information-theoretic one, are defined for a fixed representation $x$ of the input object. They do not consider that $x$ may be incomplete, nor do they measure how much the missing information can affect the prediction. In Large Language Models, users can clarify an underspecified input without any retraining. The clarified input can describe the user's unobserved intent $z^*$ more accurately and change the predictive uncertainty about the outcome. In this paper, we formalize this setting probabilistically and derive a three-term decomposition of prediction risk into specification, residual aleatoric, and epistemic uncertainties. We show that at the original information level, when only $x$ is available, the aleatoric uncertainty, represented by a Bayes risk, has two components. The first, which we call specification uncertainty, is the expected reduction in this Bayes risk from revealing the intent. The second is the residual aleatoric uncertainty, i.e., the Bayes risk that remains once the intent is known, averaged over its possible values $z$ given $x$. The distinction is important, as specification uncertainty informs the decision to request clarification before making a prediction. We build approximations of these components and connect them to existing uncertainty measures. Our experiments across different domains and modalities show distinct roles for the three components. Specification uncertainty is the most informative for detecting underspecified inputs, while residual aleatoric uncertainty is the most informative for detecting in-distribution prediction errors.

stat.ML · cs.AI · cs.CL · cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2610.20431

SwarmRSI: Recursive Self-Improvement through Evolving Scientific Agent Swarm

Qianru Meng, Xi Zhang, Konstantinos Papakostas, Congfeng Cao, Wenxiang Hu, Tianyi Lu, Xiao Zhang, Lun Zhou 等 12 位作者

原文摘要

Recursive self-improvement (RSI) in scientific research aims to improve agents' research capabilities through their own experimental experience. Scientific agent swarms can combine evidence and implementations from different research directions to guide their agents' subsequent experiments. However, poor experimental results can leave unclear whether to revise a hypothesis or its implementation, while overall program scores can hide components useful to other agents. To address these challenges, we introduce SwarmRSI, a framework that evolves scientific agent swarms for recursive self-improvement. We devise Feedback-Guided Swarm Mutation to refine hypotheses and their implementations using experimental feedback from other agents. We also propose Complementary Swarm Recombination to reuse retained hypotheses and complementary program components across agents, allowing partial results to support new research. Recursive Swarm Evolution integrates both mechanisms into a feedback loop, using new experimental evidence to guide subsequent research. Experiments on OpenProblems, NatureBench and MLE-Bench Lite demonstrate SwarmRSI's state-of-the-art (SOTA) performance. Further evaluations show that accumulated research experience improves performance on held-out tasks. Biologists also reproduce its programs for follow-up biological analyses.

cs.AI · cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02952

Balancing Representations by Identifying and Generating Under-represented Data

Filipe Laitenberger, Hirokatsu Kataoka, Yuki Asano, Cees Snoek, Mohammadreza Salehi

原文摘要

Self-supervised learning inherits the coverage of its source data. When an unlabeled corpus is long-tailed, the representation becomes dense around frequent modes and thin where coverage is weak. Objectives for long-tailed SSL reweight the images already in the corpus and add no new ones. We introduce BRIDGE (Balancing Representations by Identifying and Generating Under-represented Data), a label-free feedback loop around the source set. The current encoder scores every source image by its mean kNN distance, and a budgeted acquisition rule selects images from a band of that score distribution where support is thin and the generator still produces faithful variants. A frozen image-to-image diffusion model then adds local variants of the selected anchors, and pretraining continues on the enlarged source set. We formulate acquisition as recoverability-weighted allocation of local support and derive when a finite budget should favor a sparse band that excludes the extreme tail and when reallocation after each repair is optimal under a fixed objective. Across five long-tailed, web-scraped and synthetic source regimes, BRIDGE has the highest seven-task linear-probe and kNN average of the compared methods, and paired experiments extend the gain to ViT-S under contrastive and self-distillation objectives, to low-label fine-tuning and to semantic segmentation. The linear-probe gain is smaller under masked reconstruction. Generated repair matches an oracle that restores the withheld real images, and under a frozen encoder the inserted images shrink the neighborhoods of the selected anchors about twice as much as those of matched unselected images. BRIDGE needs no labels and leaves the pretraining objective unchanged, so it can be added to pretraining on uncurated web data, where the encoder decides which images to generate.

cs.CV · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02950

Action Verification through Conjecture and Refutation for Language Model Agents

Minhyeon Oh, Chiyeong Heo, Geonwoo Noh, Hoyoung Kim, Jangwon Son, Jungseul Ok

原文摘要

Language model agents increasingly act on users' behalf by manipulating external environments through tools, often from high-level instructions that leave critical task- and environment-specific context unstated. To prevent resulting misaligned actions, recent work verifies proposed actions before execution, largely by establishing whether the conditions supporting them hold. We argue that this confirmation-oriented view leaves a fundamental blind spot: a wrong action often passes every check that asks whether it is right. This motivates a different question for verification: rather than asking what evidence supports an action, can the agent find evidence that refutes it? We pursue this direction with BlackSwan, a refutation-oriented, expect-suspect-test procedure that predicts the intended effect of each state-mutating action, formulates specific and empirically testable hypotheses for how it could be wrong, and tests them using existing observations or additional non-mutating actions, executing the action only if none survives testing. On challenging AppWorld tasks, BlackSwan improves success rate over the strongest baseline by 14.9 and 25.2%p with and without the backbone model's thinking mode, respectively. Even without thinking or in-context demonstrations, it outperforms stronger baselines equipped with them. The gains also transfer beyond AppWorld, with BlackSwan improving WebShop performance by 12.7%p. More broadly, our results suggest that language model agents have an underused capacity to recognize and self-correct misaligned actions, which can be systematically elicited through active conjecture and refutation before execution.

cs.AI · cs.CL · cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.41511

Specification-Guided Generation for Large Language Models

Alessandro Previti, Konstantinos Vandikas, Mario Alviano, Mats Carlsson, Albin Larsson Forsberg, Alexandros Nikou, Athanasios Karapantelakis, Aneta Vulgarakis Feljan

原文摘要

Large language models (LLMs) can generate structured outputs, but existing constrained-decoding methods mainly enforce syntax and cannot directly enforce satisfaction of a formal specification. We introduce a neuro-symbolic, specification-guided decoding framework that combines LLM generation with symbolic reasoning. At each step, the framework refines the decoding grammar by filtering out values that cannot be extended to an output satisfying the specification. We present a forward-only variant that guarantees specification satisfaction by construction, and a backtracking variant that revises earlier feasible choices when their downstream consequences leave only values that the LLM itself judges unsuitable. The framework is model-agnostic, requires no retraining, and builds on existing grammar-constrained decoding mechanisms. Experiments on configuration and constraint-solving benchmarks show that the forward-only variant reliably produces specification-compliant outputs while reducing repeated LLM inference compared with rejection sampling and incurring modest symbolic-reasoning overhead. The backtracking variant further reduces cases in which the model is forced to select values that it has itself judged unsuitable, thereby substantially improving adherence to the LLM's stated interpretation of the prompt.

cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2608.03461

Solver-Aware Decompositions for Programming-by-Example: When Dividing Requires Knowing How to Conquer

Janis Zenkner, Tobias Sesterhenn, Tim Grams, Christian Bartelt

原文摘要

Decomposition-based Programming-by-Example splits a task into subtasks: a decomposer predicts intermediate subgoals, and a learned synthesizer generates a program for each subgoal. Execution-decomposition approaches train the decomposer to imitate ground-truth (GT) subgoals and thereby treat decomposition quality as a property of the task alone. A learned synthesizer, however, has its own inductive biases, and GT subgoals reflect how the annotator factorized the task, not what this synthesizer can solve. A decomposer that imitates them can therefore propose subgoals that are logically valid but intractable for the synthesizer. Motivated by this observation, we propose Solver-Aware Decompositions (SAD), which keeps supervised training on GT subgoals as a structural scaffold and additionally trains the decomposer online with self-critical policy gradients against a frozen synthesizer. Each sampled subgoal is rewarded with the negative cross-entropy of the synthesizer on the target program, a continuous signal of how difficult the subtask is for this synthesizer. Our experiments show an accuracy paradox: with the synthesizer fixed, a GT-supervised decomposer agrees more often with the GT subgoals, yet its subgoals lead to lower synthesis accuracy, although the synthesizer was trained on the same GT data. On two list-processing domains, SAD improves synthesis and task accuracy over the GT-supervised decomposer, both in-distribution and under length generalization; the same holds under zero-shot transfer to an external benchmark and for a fine-tuned language model. Even an oracle that receives the GT subgoals at inference time, with the same synthesizer and search, fails on tasks that SAD solves, and fewer than 2% of SAD's solutions to these tasks match the GT program. Decomposition quality is therefore not a property of the task: it depends on the synthesizer that executes the subgoals.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02940

PlagiaCorpus: A Bilingual, Adversarial, and Cross-Lingual Dataset for Forensic Plagiarism Detection

Adel Ammar, Serry Sibaee, Wadii Boulila

原文摘要

PlagiaCorpus is a bilingual Arabic-English dataset for benchmarking plagiarism and text-reuse detection, including reuse produced by large language models (LLMs) and reuse across languages. It contains 198,643 text records: a source corpus of 34,885 passages drawn from English and Arabic Wikipedia and two openly licensed book collections, and 163,758 queries (147,381 for training and 16,377 for testing). The queries cover five reuse classes of increasing difficulty, namely verbatim copying, character-level obfuscation, same-language paraphrasing by two different LLMs, and English--Arabic translation with structural paraphrasing, together with a Safe class of non-reused passages. Every positive query is linked to its source passage, and the train/test split is grouped by source passage so that no passage contributes queries to both splits. The release was curated through language identification, removal of LLM generation artefacts, and exact and cross-lingual de-duplication. Technical validation reports per-class length statistics, automatic checks of the train/test split, and inter-annotator agreement on a manually rated sample. The dataset is distributed in JSON and Parquet formats under a CC BY-SA 4.0 license.

cs.CL · cs.AI · cs.CY · cs.DL · cs.IR

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.03754

The Web Doesn't Sit Still: Adversarial Self-Evolving Attacks on Search Agents

Geyuan Wang, Shunyu Liu, Siyuan Liang, Chenyang Lyu, Dacheng Tao

原文摘要

Large Language Models (LLMs) have shown promise for empowering search agents to tackle complex information-seeking tasks. However, since acquiring external information necessitates web search, these agents are highly susceptible to various adversarial attacks (e.g., malicious information injection or stealthy context manipulation). Existing attack strategies inevitably rely on static and human-crafted templates, failing to emulate the dynamic and complex nature of real-world threats. In this paper, we propose an adversarial self-evolution framework that integrates bi-level optimization between the attacker's strategy evolution and the defender's adaptive mitigation, to expose the vulnerabilities of search agents. Specifically, we introduce the Contrastive Rollout Evolutionary Optimization (CREO) method to drive the directional evolution of the attack strategy population via contrastive evolutionary signals. To provide a stable environment for this continuous evolution, we further construct a controlled sandbox and design fine-grained metrics to quantify internal vulnerabilities. Extensive experiments across various multi-hop QA benchmarks and frontier LLMs demonstrate that the proposed framework yields attack efficacy superior to static baselines, underscoring the urgent need to develop more robust defense frameworks.

cs.CR · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02931

Training-Time Domain Transfer for Generator-Free Diagnostics on Novel Modalities in Computational Pathology

Julia A. Mielcarz, Andrea Camilloni, Nita Mulliqi, Jessica A. Frank, Anders Blilie, Emilius A. M. Janssen, Martin Eklund, Kimmo Kartasalo

原文摘要

Novel imaging modalities promise improved diagnostics, but AI development is constrained by scarce training data. Inference-time domain transfer can address this by mapping novel modalities to established ones, enabling reuse of existing diagnostic models. However, this introduces hallucination risks, computational overhead, and regulatory complexity. Inspired by sim-to-real transfer, we propose training-time domain transfer, generating synthetic target-domain training data from archived routine samples to enable direct diagnosis on real inputs without inference-time generation. Using unstained prostate histopathology as a proof-of-concept, we compare this approach with inference-time virtual staining and conventional H&E-based cancer detection. Training-time transfer achieved higher concordance with the conventional benchmark than virtual staining for both raw (Cohen's $κ$=0.59 vs. 0.37; Gwet's AC1=0.73 vs. 0.29) and deparaffinized tissue ($κ$=0.66 vs. 0.61; AC1=0.73 vs. 0.64), supporting its potential for emerging diagnostic modalities.

cs.CV · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02916

Defining Autonomous Weapon Systems: A schematic framework for refined purpose-driven definitions

Björn Lundgren, Nathan G. Wood

原文摘要

The literature on autonomous weapon systems (AWS) includes many proposed definitions for these (types of) systems, with core disagreements between scholars, governments, and international regulatory bodies. In this article, we present and critique prevailing (classes of) definitions, arguing that what is needed is not a singular definition, but rather a definitional framework which allows one to classify different sorts of AWS within the same conceptual family. Recognizing we cannot foresee all future AWS, we show how variations on a broadly accepted definition can be used to produce a definitional schema which can be easily modified in order to classify various concepts belonging to the class of AWS.

cs.CY · cs.AI · cs.ET · cs.HC · cs.RO

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2502.11741

SQL-o1: Automatic Text-to-SQL Program Synthesis via Self-Reward Heuristic Search

Shuai Lyu, Haoran Luo, Ripeng Li, Zhonghong Ou, Yifan Zhu, Tiesunlong Shen

原文摘要

Efficient human-database interaction is a core problem in software engineering. Text-to-SQL (Text2SQL) tackles this by mapping natural language questions to executable SQL queries, enabling automatic SQL synthesis and reliable interaction. Although large language models (LLMs) have advanced significantly, current approaches struggle with poor transferability to open-source LLMs, limited robustness against logic and function errors in complex queries, and inefficiencies in structured search. We introduce SQL-o1, a self-reward-driven heuristic search framework built on an agent-based architecture to enhance model reasoning capabilities. SQL-o1 leverages Monte Carlo Tree Search (MCTS) for structured, multi-step exploration and incorporates a dynamic pruning strategy to accelerate inference without sacrificing accuracy. On the Spider and BIRD benchmarks, SQL-o1 achieves a 10.8-point execution accuracy improvement on the complex BIRD dataset, surpassing even GPT-4-based models. It exhibits strong few-shot generalization and robust cross-model transferability across open-source LLMs. Our code is available at https://github.com/ShuaiLyu0110/SQL-o1.

cs.DB · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02908

User Attributes and Political Ideology Shape LLM Receptiveness in Online Political Debates

Akram Elbouanani, Yousra Taleb, Gianluca Demartini, Adrian Popescu

原文摘要

Online political discussions increasingly involve AI-mediated participation, as users invoke LLMs to assess contested claims, arbitrate disputes, and respond to opposing viewpoints. As these systems become participants in online exchanges, they may alter which speakers and argumentative strategies receive validation within the discussion. We study this interactional dimension by reconstructing 5,376 Debate.org debates as LLM conversations and running nearly 150,000 controlled multi-turn conversations that vary user political orientation, expertise, rhetorical strategy, and defended position, including conversations with competing users. Across naturalistic and controlled debates, we find that LLMs are systematically more receptive to left-leaning than right-leaning users. The asymmetry persists when users with different political orientations defend the same position and when opposing users appear in the same conversation, indicating that it is not fully explained by stance agreement or simple accommodation to the current speaker. Repeated interaction increases receptiveness without eliminating this ordering; expertise also matters, while rhetorical differences are smaller. The pattern appears across multiple models, but is not fixed: fine-tuning on different political corpora can largely erase or reverse it. These results show that LLMs can respond systematically differently to participants in the same political debate, making such interactional asymmetries important for understanding their emerging role in online public discourse.

cs.AI · cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02905

What makes a theorem interesting? Operationalizing interestingness for open-ended exploration

Silvère Gangloff

原文摘要

An agent that explores, whether it generates conjectures or research ideas, must decide which of them are worth keeping. This judgment of interestingness is usually delegated to a hand-picked proxy or to a language model, neither of which is known to correspond to what humans actually value. I study it for mathematical theorems, by observing which results mathematicians single out in formal libraries and in papers. I model each way of singling out results as an evaluator that weights features of a result's statement, its proof and dependencies, and split its weights into a shared and a subjective part. In formalized mathematics, ex-ante features predict these selections as well as a large language-model judge, and the two complement each other. Interestingness is also largely shared. One set of weights, carried mostly by structural features such as the length of the statement and of the proof and the subfields the proof draws on, accounts for most of the signal and transfers in part across formal libraries. What remains subjective depends on practices and individual authors more than on subfields. The judge's advantage lies in its knowledge of the canon, and neither the judge nor the features rely on memory of the papers. The approach, which infers what a community finds interesting from what it singles out, can be applied to other sciences.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02903

Maswao: An automatically collected, prepared, and benchmarked dataset for the South African Sign language

Thapelo Sindane, Unarine Netshifhefhe, Vukosi Marivate, Abiodun Modupe, Herkulaas Combrink, Ketro Sithole

原文摘要

Recent advancements in sign language processing can be attributed to developments in both large-scale dataset collection and algorithmic complexity in areas such as pose estimation techniques, signal processing through Transformer models, and model optimization. However, only a handful of sign languages, such as the American, British, and German, from over 200 globally recognized sign languages, reap the benefits of such advancements, while the rest lag behind. In this study, we aim to address this technological disparity by proposing a framework for the automatic collection and preparation of sign language datasets, with very minimal human-labour intervention, using the South African Sign Language (SASL) as a testbed. Specifically, we automatically collect SASL video content using public resources such as news broadcasts. Additionally, we convert their accompanying audio streams into text using available Automatic Speech Recognition (ASR) tools. Furthermore, we crop the sign language area in the videos and use gradient descent to locate valleys of non-motion to determine end-of-sentence points within long videos to create our weakly sentence-aligned parallel corpora -- the Maswao (Signs) dataset. Finally, we benchmark the weakly aligned dataset on the task of Sign-to-text Translation using Recurrent Neural Networks and Transformer models to compare their individual intricacies. Our results show that Transformer models proposed in the literature outperform recurrent baselines, with the best-performing Transformer achieving BLEU-1 to BLEU-4 scores of 13.78, 5.21, 2.52, and 1.54, respectively, as well as ChrF = 18.09 and ROUGE-L = 10.51. These results indicate that the task remains promising and warrants further investigations.

cs.CL · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2609.05090

Constructing and Evaluating Clinical Reasoning Trajectories for Medical Agent

Yunqi Zhu, Wen Tang, Jinghao Niu, Wensheng Zhang, Xuebing Yang

原文摘要

Evaluation of medical artificial intelligence (AI) agents remains predominantly answer-centric, assessing only the correctness of final outputs while overlooking the quality of intermediate reasoning. We propose MedTraj, a framework that treats reasoning trajectories as primary objects for modeling the progressive refinement of clinical evidence toward diagnostic decisions. A reasoning trajectory captures the state transitions among medical concepts---from clinical observations through evidential reasoning to a final conclusion---making the implicit reasoning chain of a medical agent explicit and inspectable. Each trajectory is scored across five quality dimensions: coherence, evidence support, hallucination, completeness, and traceability. Controlled error injection introduces targeted faults into otherwise correct trajectories to establish causal links between specific reasoning failures and measurable quality degradation. Step-level filtering based on marginal contribution identifies which individual reasoning steps drive or undermine reasoning trajectory quality. Quality-weighted reasoning trajectory context feeds these evaluations back into the model at inference time. Experiments across CareQA, PubMedQA, and CECMed demonstrate that reasoning trajectory context consistently improves reasoning coherence. For the severity assessment on the CECMed, quality-weighted reasoning trajectory context nearly doubles the correctness over the zero-shot baseline while cutting the hallucination ratio by 87%. Experiments show that 15.3% of reasoning steps carry most of the quality signal, and that extending chains beyond four steps yields diminishing returns.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02896

Bridging Low-Level Image Fusion and High-Level Semantic Understanding

Guangyuan Dong, Lingdong Shen, Xudong Zhang, Haodong Jing, Haoyu Wang, Zihao Li, Zonghao Ying, Ziyu Song 等 11 位作者

原文摘要

Remote sensing image fusion and semantic interpretation are usually optimized separately, so reconstruction quality need not preserve downstream retrieval utility. We propose LANCE, which uses structured language as an explicit, indexable semantic state. A frozen vision-language model maps the input to a closed-vocabulary text record; a language-conditioned generator fuses the imagery; and a dense verifier with a differentiable continuous summarizer re-describes the output. Text-text consistency and anti-collapse contrastive text-mask losses with negative queries close the loop without dense annotations or VLM fine-tuning. A variational EM formulation yields a per-step contraction result under regularity assumptions, distinguishing semantic faithfulness from verifier overfitting. Across seven test sets on GF2, QuickBird, WorldView-III, and WorldView-II, LANCE improves reduced-scale reference metrics by up to 11.7% (median 2.1%) over the strongest learning baseline. Independent evaluation using a held-out VLM and a supervised aerial detector yields gains of +6.4 mAP and +0.072 open-vocabulary F1 over the closest competitor. Text perturbations, cross-verifier transfer, and pseudo-RGB ablations examine the mechanisms underlying these gains. The framework connects image fusion to verifiable semantic representations for multimodal retrieval and knowledge management.

cs.AI · cs.IR

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02893

Voice-Native LLMs Hear Frustration but Do Not Listen to It

Hanjun Luo, Yuhan Lu, Hanan Salam

原文摘要

Users convey frustration in what they say and in how they say it, and voice-native models, which hear speech instead of an ASR transcript, can register both. Existing evaluations read unchanged performance as robustness without checking perception. Following Brunswik's lens model, we measure on identical audio whether frustration is expressed, perceived, and utilized. We cross frustrated wording with frustrated voice in a within-item 2x2 design over 353 tasks and release the 1,412 clips as VoxAffect. Across 8 models, outputs look robust, with no detectable vocal effect on accuracy for 7. With neutral wording, all 8 label the frustrated voice frustrated in at most 3.1% of clips, although 7 of them detect acted anger in 73-100% of clips and human listeners classify the voice correctly in 65.3% of judgments. Yet linear probes decode the frustrated voice from the audio encoders of the 2 open-weight models, with paired accuracy 0.67 and 0.69, and in proprietary models too the voice raises perceived frustration once the words agree. The models thus hear the frustrated voice but do not listen to it, whereas their invariance to frustrated wording, perceived in 40-75% of clips, is genuine robustness. Our work is openly accessible at https://github.com/Astarojth/hear-not-listen.

cs.SD · cs.AI · eess.AS

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02892

JABER: Sequential Quality Control for Human-AI Data Annotation under Shared Bias

Fengxu Yang, Minghui Liwang, Yang Lu, Bo Han, Xianbin Wang, Seyyedali Hosseinalipour

原文摘要

High-quality annotations are essential for training and evaluating artificial intelligence (AI) systems. Quality control (QC) often acquires additional human or AI judgments before accepting a label. Nevertheless, shared biases can cause correlated judgments, leading conventional methods to mistake repeated agreement for independent evidence of correctness. To address this challenge, we propose JABER (Joint Alias-Preserving Belief for Evidence Routing), a bias-aware sequential QC framework that maintains a joint Bayesian posterior over label correctness and shared biases to guide evidence acquisition and termination decisions. Building on this posterior, we develop a finite-horizon algorithm that selects evidence-acquisition and termination actions to minimize expected QC loss under resource constraints. Our theoretical analysis shows that conditional-independence updating can become arbitrarily confident in an incorrect label under persistent shared bias. We further identify when repeated inspections cannot distinguish a correct label from an incorrect label supported by shared bias, when obtaining a new label can resolve this ambiguity, and why modeling the dependence between label correctness and shared bias can be necessary for QC decisions. Our experiments show that JABER reduces expected QC loss by an average of 4.65% compared with a method maintaining separate label correctness and bias probabilities when shared bias makes repeated inspections unreliable. Further, JABER's finite-horizon planning, which evaluates outcomes of successive QC actions, reduces expected QC loss by an average of 10.05% compared with a one-step lookahead method. In real-API experiments on FEVER, JABER achieves zero labeling errors while reducing QC costs by 2.67% and 16.99% compared with marginal-based belief modeling and one-step lookahead methods, respectively.

cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2508.06352

Leveraging generative AI for human understanding: meta-requirements and design principles for Explanatory AI as an emerging paradigm

Christian Meske, Justin Brenne, Erdi Uenal, Sabahat Oelcer, Ayseguel Doganguen

原文摘要

Artificial intelligence (AI) systems increasingly support decision-making across critical domains, yet explainable AI (XAI) approaches prioritise algorithmic transparency over human comprehension. While XAI reveals computational processes for model validation, end users require explanations integrating domain knowledge, contextual reasoning, and professional frameworks. This disconnect reveals a design challenge: existing approaches fail to address how practitioners need to understand and act upon AI recommendations. This paper introduces Explanatory AI as an emerging complementary paradigm where AI systems leverage generative and multimodal capabilities to serve as explanatory partners for human understanding. Unlike traditional XAI, which answers 'How did the algorithm decide?' for validation purposes, Explanatory AI addresses 'Why does this make sense?' for practitioners making informed decisions. Through theory-informed design, we synthesise multidisciplinary perspectives from cognitive science, communication research, and education with empirical evidence from healthcare and AI expert interviews. Our analysis identifies five dimensions distinguishing Explanatory AI from XAI: explanatory purpose, communication mode, epistemic stance, adaptivity, and cognitive design. We derive five meta-requirements and formulate ten design principles prescribing how to build such systems, illustrated through five UI mockups exemplifying implementation. Explanatory AI operates standalone or combined with XAI, serving distinct purposes in creating systems that genuinely support human reasoning and decision-making.

cs.AI · cs.HC

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02889

Tug of Watts: A study on the Energy - Accuracy Trade-off of Large Language Models on Medical Question Answering Tasks

Viju Sudhi, Giulio Billi, Andreea Scrob, Regina Stodden, Philipp Cimiano

原文摘要

Large Language Models (LLMs) are increasingly used in medical question answering (QA) systems. While recent works aim at incrementally improving the downstream task accuracy, an empirical estimation of the models' energy consumption and the associated carbon footprints is often overlooked. In this study, we demonstrate how practitioners can estimate energy consumption and arrive at reliable emission estimates for medical QA tasks available in the OpenMedical-LLM leaderboard by investigating 13 LLMs from 3 model families ranging from 270M to 14B parameters. We discuss how one can critically interpret energy versus accuracy trade-offs for a set of models for a given task. With this work, we underline that reporting disjoint energy and accuracy metrics alone does not suffice for careful trade-off assessments, but it also demands systematic construction and interpretation of joint trade-off metrics. We envisage such studies aiding practitioners in adopting better and more efficient models for their task and helping them become active participants in building sustainable LLM pipelines.

cs.CL · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2509.04192

Domain size asymptotics for Markov logic networks

Vera Koponen

原文摘要

A Markov logic network (MLN) $\mathbb{M}$ determines a probability distribution $\mathbb{P}_n^\mathbb{M}$ on the set $\mathbf{W}_n$ of structures, or ``possible worlds'', with domain $\{1, \ldots, n\}$. We study the properties of such distributions as $n$ tends to infinity. We show that with mild assumptions on an MLN $\mathbb{M}$ with one soft constraint with an arbitrary positive weight the distribution $\mathbb{P}_n^\mathbb{M}$ will behave quite differently from the uniform distribution $\mathbb{P}_n^{uni}$ on $\mathbf{W}_n$ for all sufficiently large $n$. For a language with only one relation symbol $R$ which has arity 1 we give an almost complete characterization of the possible asymptotic behaviours of $\mathbb{P}_n^\mathbb{M}$ as $n \to \infty$, where $\mathbb{M}$ may be any MLN for this language. The asymptotic behaviour depends on the soft constraints and weights of the MLN. This characterization is used to show that if the language under consideration contains at least one relation symbol of arity 1 then the following holds: (a) There is an MLN $\mathbb{M}$ such that for every lifted Bayesian network (LBN) $\mathbb{G}$ there are infinitely many $n$ such that $\mathbb{M}$ and $\mathbb{G}$ determine different distributions on $\mathbf{W}_n$. (b) There is an LBN $\mathbb{G}$ such that for every MLN $\mathbb{M}$ there are infinitely many $n$ such that $\mathbb{G}$ and $\mathbb{M}$ determine different distributions on $\mathbf{W}_n$. We also show that, in the limit, the weight dimension and the domain size dimension may behave completely differently.

cs.AI · cs.LO · math.LO

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02883

Integrating Domain Adaptation and Retrieval-Augmented Generation for Saline-Alkali Land Reclamation Planning

Zuguang He, Congcong Zheng, Wanyue Chen, Wenya Zhou, Zaijin Yang, Long Long, Yucheng Zhang, Shuqin Gao

原文摘要

Reclamation of saline-alkali land requires management plans that account for soil and water conditions, treatment evidence, and financial constraints. This study develops Wenyao, an agricultural decision-support system integrating domain adaptation with retrieval-augmented generation. A bilingual corpus of 8,664 research papers was processed using a 110-indicator schema to construct SA-Bench, comprising 4,860 paper-level planning instances. A QwQ-32B model was adapted through low-rank supervised fine-tuning and combined with case retrieval for budget-conditioned planning. Evidence extraction achieved an F1-score of 0.83 on 300 expert-annotated papers. Without retrieval, Wenyao achieved 95.83% accuracy on 936 source-disjoint multiple-choice questions, compared with 87.93% for the strongest general-purpose baseline; removing explanatory supervision reduced accuracy to 89.42%. Matched-evidence comparisons across 69 region-budget cases supported complementary contributions from adaptation and retrieval. In a blind assessment by 31 experts, Wenyao with retrieval obtained the highest mean scores for objectives and projected outcomes, whereas GPT-5 scored highest for cost estimates. These findings support literature-informed reclamation planning with professional review, while ongoing field monitoring will assess soil and crop responses.

cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2606.30266

Towards Continual Motion-Language Agents: LoRA Variants for Incremental Motion Understanding and Generation

Bertram Taetz, Hugo Albuquerque Cosme da Silva, Gabriele Bleser-Taetz

原文摘要

Motion-language agents must possess the bidirectional capability to both understand human movement (motion-to-text, M2T) and generate it from natural language (text-to-motion, T2M). While foundational models have achieved strong performance in static settings, autonomous agents operating in dynamic environments must continuously incorporate new motion concepts -- such as novel athletic styles or specialized gestures -- without catastrophic forgetting of previously acquired skills. We investigate the stability-plasticity trade-off in bidirectional motion-language learning under sequential task exposure. Building on a frozen large language model backbone, we introduce low-rank adaptation (LoRA) variants designed to mitigate inter-task interference. We specifically propose mixture-of-experts architectures that utilize an autoencoder-based router to select task-specific experts at inference time, so that no task-label is needed. To evaluate these methods, we establish a reproducible five-task benchmark derived from HumanML3D through semantic clustering of motion descriptions. Our experimental results demonstrate near-zero forgetting across both M2T and T2M directions while maintaining high generation and captioning quality. Furthermore, we show that hard expert selection via routing significantly outperforms soft expert blending in quality metrics, indicating that preserving expert isolation is critical for maintaining performance in our continual learning setting. Finally, we observe that a divergence between token-level accuracy and downstream generation quality may occur, highlighting the need for more comprehensive evaluation protocols in future research on lifelong motion-language agents.

cs.LG · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02856

Kolmogorov-Arnold Networks vs. Multi-Layer Perceptrons: A Rigorous Statistical Benchmark on Clinical Tabular Data

Rafał Wodziński, Paweł Szczerbiak

原文摘要

Kolmogorov-Arnold Networks (KANs) have recently been proposed as a promising alternative to Multi-Layer Perceptrons (MLPs). However, empirical evaluations on structured medical datasets remain scarce, and commonly reported metrics like AUROC can yield overly optimistic performance estimates on imbalanced clinical cohorts. In this study, we present a standardized benchmark comparing eleven architectures - nine KAN variants (FastKAN, WavKAN, ReLUKAN, ChebyKAN, TaylorKAN, GramKAN, JacobiKAN, LegendreKAN, and TabKAN), a tuned StandardMLP, and a Random Forest baseline - across seven heterogeneous medical tabular datasets spanning oncology, cardiology, nephrology, neurology, endocrinology, and obstetrics. Our experimental protocol employs a 3x5 Repeated Stratified K-Fold cross-validation scheme with nested Bayesian hyperparameter optimization and evaluates discrimination via Area Under the Precision-Recall Curve (AUPRC), calibration through the Brier Score, and computational efficiency. Formal equivalence testing is conducted using the Bayesian Correlated t-Test with a Region of Practical Equivalence (ROPE) at +/- 0.01. The results indicate that on datasets with severe class imbalance, dedicated KAN architectures (particularly TabKAN and ReLUKAN) can improve minority-class detection and achieve higher AUPRC than standard MLPs. Nevertheless, across pooled datasets, KAN variants demonstrate comparable overall predictive discrimination rather than a systematic advantage, while requiring 1.20x to 1.48x longer inference latency per sample and up to 12.5x more trainable parameters. Consequently, StandardMLP remains a strong baseline on the efficiency Pareto frontier, while KAN variants provide a viable alternative when precision on rare positive cases is the primary clinical objective. The code for reproducing the results is available at https://github.com/rafalwodzinski/kan_mlp_comparison.

cs.LG · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02852

Value-Conditioned Proposal Acquisition for Test-Time Reasoning

Zeyan Li, Jianfeng Xu

原文摘要

Sampling and selection spend test-time compute inside a candidate pool, but they cannot recover a correct answer that never enters it. We introduce ValueFormation, a test-time policy that decides after each primary-model draw whether to stop, continue sampling, or acquire one proposal from a heterogeneous reasoning specialist. The three actions are compared on one cost-regularized value scale learned from vote-state features; the calibration is frozen across tasks and primary models, and acquired proposals are accepted only through a frozen posterior comparison. On held-out MATH-500, one Llama-fitted calibration improves over the strongest of eight identically evaluated baselines by more than fifteen points for each of three 8B primaries and transfers unchanged to Qwen3-14B. Removing acquisition lowers Llama MATH accuracy by 20.43 points, while same-trace and same-access replays attribute the gain to state-conditioned acquisition rather than specialist access alone. Where primary coverage is high, as on GSM8K, acquisition is rare and gains mainly come from cheaper allocation; on MBPP, a fixed code route attains the top accuracy at the lowest interaction cost. Varying the token price monotonically reduces calls without retraining.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02849

TLAPS-Bench: Formally Proving Correctness of Real-World System Specifications with TLA+ Proof System

Qian Cheng, Ruize Tang, Saad Mohammad Rafid Pial, Zhizhen Cathy Cai, Yiming Su, Jackson Clark, Munim Thahmid, Mahdiya Rahman Sukanya 等 16 位作者

原文摘要

TLAPS-Bench is a benchmark and execution harness for constructing machine-checked proofs for real-world system specifications in TLA+ with agentic AI. Proving correctness of complex system specifications requires discovering inductive invariants, reasoning about intricate state transitions, and decomposing proofs into tractable obligations. TLAPS-Bench provides a proof harness that supports iterative proof development atop the TLA+ Proof System (TLAPS), where agents are equipped with proof tools and libraries, with safeguards against reward hacking. Using the harness, we have employed AI agents to prove desired properties for important system specifications that were not proved before, spanning cache coherence, Byzantine consensus, and write-ahead logs. TLAPS-Bench contains 265 proof targets (each encodes a property to prove) from 95 TLA+ specifications of real-world systems. We evaluated frontier AI agents, represented by Codex with GPT models, on a subset of 70 proof targets across 12 specifications. The highest-scoring model, GPT-6.1 Sol (max), proved 31 (44.3%) proof targets. The results show challenges and opportunities in end-to-end proof development.

cs.AI · cs.DC · cs.LO · cs.SE

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02846

SpecOPD: Branch-Parallel On-Policy Distillation for Speculative Decoding

Tianyu Liu, Jun Zhang, Wenjie Huang, Yuhao Shen, Tianqi Xu, Xinyue Hao, Hongyaoxing Gu, Xuyang Liu 等 15 位作者

原文摘要

Speculative decoding (SD) accelerates large language model inference by reducing the sequential cost of token-by-token generation. Advanced SD methods reduce proposal cost with parallel drafters that predict entire blocks in one forward pass, but overall speedup still depends on acceptance length. On-policy distillation (OPD) can push well-trained drafters beyond acceptance-length plateaus, yet conventional pipelines collect supervision through sequential draft-and-verify rollouts. Our key insight is that on-policy supervision need not inherit these dependencies: a draft block can be supervised using only its reference prefix and draft-sampled history. We propose SpecOPD, a branch-parallel OPD framework that parallelizes draft sampling and target supervision across independent branches drawn from the evolving drafter policy at reference anchors. We pack these branches into proposal trees to share target-prefix computation, using tree masks for full attention, branch-consistent indexer groups for sparse attention, and indexed state replay for linear attention to preserve each branch's history. We train the drafter with an E2E-TV surrogate that directly targets acceptance length. Our study spans 8B full-attention (GQA), 27B full-linear (GQA--GDN), and 125B sparse-linear (QSA--GDN) targets, three parallel draft architectures (DFlash, DSpark, and DFlash V2), and four task domains (dialogue, coding, mathematics, and agentic). SpecOPD achieves up to 41.45% higher mean acceptance length, $16.09\times$ supervised-token throughput over sequential-rollout OPD under native configurations, and up to $3.56\times$ generation speedup over autoregressive decoding.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.02843

Math Foundation Formalization Agent: A General-Purpose Approach to Formalizing Mathematical Foundations

Jiaxin Xie, Guowei Dou, Xiang Ji, Wensheng Yu

原文摘要

Mathematical formalization plays an important role in establishing rigorous and machine-verifiable foundations for mathematical reasoning. With the development of large language models (LLMs) and interactive theorem provers, intelligent agents offer a promising approach to automating the formalization of mathematical theories. This paper presents the Math Foundation Formalization Agent (MFF-Agent), a general-purpose intelligent agent designed to support formalization tasks across multiple mathematical domains. The agent integrates the reasoning capabilities of large language models with the formal verification capabilities of the Rocq proof assistant. It incorporates task classification, tool-assisted execution, verification feedback, and state-based task management to support the generation, refinement, and verification of formal proofs. Unlike an approach focused on a single mathematical theory, the agent is examined through case studies involving Morse--Kelley axiomatic set theory (MK), real numbers, propositional logic, and equivalences of the axiom of choice. These cases investigate its applicability across mathematical foundations and distinguish source-level completion, interactive verification, and reproducible compilation. The results demonstrate the feasibility of a unified operational framework for heterogeneous foundation-level tasks and clarify the gap between such a framework and a fully integrated mathematical foundation.

cs.LO · cs.AI

查看论文
arXiv ↗PDF ↗