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
+0已有论文修订
+100新增 AI 导读

arXiv 最近完成扫描：2026-11-09 06:10 · 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 07:08 — 2026-11-09 07:08（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 06:10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 06:08（北京时间）
新论文 · v1arXiv:2610.39454

Kairos: Semantics-Aware Orchestration of Planning Engines across Classical, Numeric, Temporal and Hybrid PDDL

Francisco Martín Rico, Esther Aguado González, Francisco Miguel Moreno Olivo, Rodrigo Pérez Rodríguez, José Miguel Guerrero Hernández, Francisco J. Rodríguez Lera

原文摘要

Planning at run time, as in robots that re-plan while they act, needs a planner that reads the PDDL its users write (durative actions, numeric resources, time windows, continuous change), answers quickly, and delivers only validated plans. Specialised planners are strong on individual fragments of PDDL, but they differ in coverage, latency and in what their failures mean. We study whether selecting heterogeneous engines from the semantics of each task can give broad coverage of PDDL, from STRIPS to PDDL+, while remaining competitive with specialised planners and making explicit what each answer guarantees. Kairos profiles each task to decide which of five engines apply (ground heuristic search, compressed and snap-action temporal search, time steps for hybrid dynamics, lifted search) and schedules them under one budget of time and memory. Every candidate plan is checked by a validator independent of the planner's transformations, and "unsolvable" is reported only when a search whose exhaustion is a proof exhausts its space. On 1,756 tasks from nine comparative suites, Kairos solves more tasks at the full limits than the strongest evaluated planner of each suite, with the largest differences on numeric, temporal, PDDL3 and PDDL+ planning, and it also solves the most tasks at 100 ms and 1 s. Resampling over domains, the gain is above zero in six of the nine suites, while on classical planning against LAMA it is within the noise. A further 450 feature-oriented tasks and 26 one-feature probes show that Kairos handles all seventeen features probed, where no other planner evaluated handles more than eight. Ablations identify compressed temporal search and the novelty fallback as major contributors, and a comparison with VAL traces every disagreement to a PDDL+ convention or to a documented VAL defect. Planner and benchmark are open source.

cs.AI · cs.RO

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39439

Timbre: Reliability-aware harmonic consensus for label-free multimodal test-time rexinforcement learning

Haotian Lyu, Haoyu Wang, Ru Wang, Feng Ji, Ziwen Li, Jiaxian Guo, Mingming Gong

原文摘要

Label-free test-time reinforcement learning adapts vision--language models using rewards derived from their own predictions. However, misinterpreted visual evidence and sensitivity to textual formulation can produce a consistent but incorrect consensus that majority voting reinforces. Cross-view aggregation can expose such errors, but image and text transformations may themselves degrade task-relevant evidence to different extents. We introduce \method, a reliability-aware harmonic consensus framework for multimodal test-time reinforcement learning. \method constructs a $2\times2$ grid combining original and perturbed images with direct-solving and reframing modes, selects pseudo-labels through weighted harmonic aggregation, and adapts modality-level weights using batch-averaged, temporally smoothed shifts in pseudo-label confidence as a heuristic reliability signal. We formulate consensus as a weighted $χ^2$-barycenter problem whose unique minimizer is the normalized weighted harmonic aggregate, without requiring correct-answer invariance. An exact support--dispersion decomposition characterizes when harmonic aggregation preserves or corrects arithmetic rankings, while a sensitivity analysis establishes how modality reweighting changes pairwise ranking margins, informing interpretation of the heuristic updates. Across six benchmarks and three Qwen3-VL model scales, \method outperforms TTRV in 16 of 18 benchmark--model settings on adaptation instances and 17 of 18 settings on held-out instances under matched estimation and optimization rollout counts. These results demonstrate improved target-task adaptation and transfer to unseen instances without further updates.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39438

ATLAS-Bench: An Agentic Tool-Use Benchmark for Local Area Search

Keyun Zhang, Dapeng Yin, Fanfu Wang, Shuang Lang, Zhenyu Zhu, Yulin Huang, Chong Yang, Pengjie Wang

原文摘要

Reliable local search is essential for LLM agents to support everyday decisions and longer plans by finding places that meet users' needs at given times and locations. Yet existing benchmarks provide limited coverage of whether search results satisfy mandatory requirements, account for preferences, and ground factual and experiential judgments in place attributes and reviews. We introduce ATLAS-Bench, a benchmark of 1,200 queries across three regions, expanded from real search events using place attributes, reviews, and existing user profiles from the platform. Each query distinguishes mandatory requirements from preferences and factual conditions from subjective judgments, with supporting evidence and criteria for checking each requirement. Queries are validated through iterative assertion checks and cross-model review. We also introduce ATLASBox, a standardized evaluation environment with search tools and a fixed point-of-interest (POI) knowledge base. Given a query, timestamp, and location, an agent searches place attributes and reviews to recommend evidence-supported options or report that no place meets all mandatory requirements. We evaluate answer quality, trajectory quality, and execution cost. Across 14 configurations of 11 models, GPT-6-Luna achieves the highest constraint satisfaction rate among returned recommendations (71.5%) and factual accuracy (95.6%). However, it returns at least one acceptable recommendation for fewer satisfiable queries than GLM-5.3 (67.3% versus 85.3%). These results highlight the challenge of satisfying individual requirements while providing usable recommendations across queries.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39435

Verifiable LLM-Guided Focal SMT Solving for Quantified Arrays

Kunhang Lv, Rui Han, Fuqi Jia, Yuhang Dong, Feifei Ma, Jian Zhang

原文摘要

Large language models can often identify useful semantic structure in formal artifacts, but their outputs cannot be trusted as proofs or solver decisions. We study whether LLMs can instead guide formal reasoning systems through a narrow, solver-verifiable interface. We instantiate this idea in quantified array SMT solving, where modern instantiation-based techniques can fail when key ground terms and equalities are hidden by auxiliary variables, axiom guards, and nested array terms. We observe that many instances are largely driven by a small set of focal ground terms or constraints, similar to the backdoor sets in SAT solving. Building on this, we present LinguaArray, a framework where an LLM proposes a small Semantic Focus Set (ground terms/constraints), while a conventional SMT solver remains the sole proof engine and certifies all results. LinguaArray uses these LLM-suggested, solver-validated finite sets to significantly enhance solver capability, solving additional satisfiable and unsatisfiable instances that the backend alone fails to solve. It interacts within a finite time and then falls back to the backend solver to mitigate potential regressions. Evaluation on 1015 SMT-LIB instances across five logics under 1200s timeout (including LLM latency) shows that LinguaArray solves up to 69.1% of instances and increases solved-instance counts by 105%--397% over the corresponding backends.

cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2605.29488

AnyMo: Scaling Any-Modality Conditional Motion Generation with Masked Modeling

Yiheng Li, Zhuo Li, Ruibing Hou, Yingjie Chen, Hong Chang, Hao Liu, Shiguang Shan

原文摘要

Conditional human motion generation remains a fundamental challenge in computer vision and robotics. Despite significant progress, current methods are often constrained by fixed modality configurations and task-specific architectures, leaving cross-modal interactions and the scaling laws of multimodal-conditioned synthesis largely underexplored. A key bottleneck is the scarcity of large-scale modality-aligned motion data, limiting generalization across diverse control signals. In this work, we introduce OmniHuMo, a large-scale, high-quality dataset comprising over 5,000 hours of motion and 3.2 million sequences with precisely aligned multimodal annotations (e.g., text, speech, music, and trajectory). Leveraging OmniHuMo, we propose AnyMo, a unified multimodal framework combining a Residual FSQ-based motion tokenizer with a scalable masked modeling transformer, enabling high-quality motion synthesis under arbitrary modality combinations. Extensive experiments show that AnyMo achieves high-fidelity synthesis while offering flexible control over both spatial and stylistic attributes.

cs.CV · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39423HF 社区推荐

FDSML: Formula-Driven Supervised Metric Learning with Parameter-Aware Triplet Loss

Hina Otake, Shinei Arakawa, Keitaro Tanaka, Yoshihiro Fukuhara, Hirokatsu Kataoka, Shigeo Morishima

原文摘要

Formula-Driven Supervised Learning (FDSL) enables vision pre-training without real images, but existing methods cast pre-training as classification on discretized labels, discarding continuity in the generative process. We propose Formula-Driven Supervised Metric Learning (FDSML), which derives relational supervision from image-generation parameters beyond discrete class labels. FDSML combines FineFractalDB, our refined fractal dataset with higher intra-class variation and more stable structures, with a Parameter-Aware Triplet (PA-Triplet) Loss that assigns positives and negatives according to distances in the underlying Iterated Function System parameter space. This objective trains embeddings to respect parameter-derived relative similarity orderings. Experiments show that FDSML improves transferability to image retrieval while remaining competitive on downstream classification and improving robustness under several corruption settings. On ResNet-50, Recall@1 improves from 65.6% to 74.0% on Stanford Online Products and from 56.2% to 78.4% on Stanford Cars over FDSL pre-training. These results suggest that parameter-derived relational supervision provides a stronger synthetic pre-training signal than label-based classification for transfer beyond recognition.

cs.CV · cs.AI · cs.LG · cs.NE

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39417

Well-Specified and Well-Tested? Assessing SWE-bench-Style Datasets with a Self-Evolving Agent

Zhonghao Jiang, Xuyang Zheng, Yanfu Yan, Zhongxin Liu, Lingfeng Bao, Shuiguang Deng

原文摘要

Large language models (LLMs) have shown strong performance on software engineering (SE) tasks, driving the need for high-quality datasets for reliable training and evaluation. SWE-bench-style datasets (SWE datasets) are widely used to evaluate LLM-based coding agents on real-world software issues, where task clarity and test coverage directly affect whether instances are solvable and verifiable. However, quality issues persist in existing SWE datasets, and manual annotation is costly and labor-intensive, motivating LLM-based annotation. Existing approaches typically rely on static, overly general rubrics, limiting their ability to handle diverse and nuanced scenarios and often yielding high false positive rates. In contrast, human annotators draw on accumulated experience to make context-aware judgments. Inspired by this, we propose ExLabel, an LLM-based, self-evolving, experience-guided assistant for task clarity and test coverage assessment in SWE datasets. ExLabel maintains an evolving experience bank that distills cross-instance knowledge into stable yet adaptive judgment criteria and reuses relevant experience to guide the assessment of new instances. For systematic evaluation, we construct SWE-TQ, a new benchmark spanning eight programming languages, by cleaning human annotations from SWE-bench and Multi-SWE-bench. Experiments with three backbone LLMs show that ExLabel achieves the best F1 across all settings, with relative improvements of 44.2%/49.9% and 24.0%/18.4% over the best baseline for task clarity and test coverage assessment on the Python/Multilingual subsets, respectively. Moreover, a coding model fine-tuned on SWE-rebench data filtered by ExLabel resolves 66.2% of issues on SWE-bench Verified, outperforming the model trained on data filtered by the best baseline by 14.5%.

cs.SE · cs.AI · cs.CL

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2608.16213

Process-Constituted Intelligence: A Shared Criterion for Humans and Machines

Michael J. Richardson, Manuel Varlet, Ayeh Alhasan, Cassandra Crone, M. Paula Diaz Monfort, Patrick Nalepka, Mark Dras, Rachel W. Kallen 等 9 位作者

原文摘要

Intelligence is constituted by \textit{process} (iterative activity through which output emerges), not in the output itself. Generative AI (GenAI) is trained on \textit{traces} (textual and visual residues of human cognitive processes), reproducing samples from a distribution of those traces. Its outputs resemble reasoning, problem-solving, and creativity, yet the activity that produces such outputs in humans remains largely absent. Current GenAI is, therefore, weakly equivalent to the cognition it imitates, matching outputs while process stays absent or opaque. The cognitive sciences have long distinguished between weak and strong equivalence. Here, we define \textit{strong} equivalence across seven process features, assessable against human and machine cognition. Our process-based account addresses a symmetric risk: GenAI tools that outsource a person's generative processes may leave critical capacities unbuilt. We specify design principles for GenAI that instantiate more process and preserve rather than erode human judgment and creativity, and outline process audits that make strong equivalence testable.

cs.AI · cs.ET

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2603.15220

InterPol: De-anonymizing LM Arena via Interpolated Preference Learning

Minsung Cho, Jaehyung Kim

原文摘要

Strict anonymity of model responses is a key for the reliability of voting-based leaderboards, such as LM Arena. While prior studies have attempted to compromise this assumption using simple statistical features like TF-IDF or bag-ofwords, these methods often lack the discriminative power to distinguish between stylistically similar or within-family models. To overcome these limitations and expose the severity of vulnerability, we introduce INTERPOL, a model-driven identification framework that learns to distinguish target models from others using interpolated preference data. Specifically, INTERPOL captures deep stylistic patterns that superficial statistical features miss by synthesizing hard negative samples through model interpolation and employing an adaptive curriculum learning strategy. Extensive experiments demonstrate that INTERPOL significantly outperforms existing baselines in identification accuracy. Furthermore, we quantify the real-world threat of our findings through ranking manipulation simulations on Arena battle data.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39410

Retaining Repair Hypotheses for Cost-Aware Execution

Jae Hun Hwang, Hong-In Won

原文摘要

Several corrections of an executable model can fit the same observations while predicting different execution outcomes. We study a repair-to-execution interface that retains compatible finite guard repairs, conditions their weights on diagnostic and ordinary-execution observations, and uses their predictions to choose routes, priced tests, or a backup. On 24 synthetic two-stage devices, an equal mixture of a language-model-generated repair and a uniform belief over compatible repairs (the AI-weighted belief) lowers no-test cost by 2.14 percent relative to committing to the generated repair. Access charges can erase the benefit of diagnosis on isolated tasks, while repeated work can recover it under the AI-weighted belief. With normal-execution learning in every repeated-work arm, workload-valued testing costs 3.33 percent less than same-budget random selection under the AI-weighted belief. Weighting changes the comparison: uniform normal learning costs 540.75 less than AI-weighted normal learning on this cohort, with a device-paired interval excluding zero. Testing recovers part of that gap, but uniform normal learning still has a lower aggregate cost than AI-weighted workload-valued selection. Under uniform beliefs, information-gain selection has the lowest total among the compared selectors, while workload-valued testing costs more than normal learning. These results support retaining alternatives to a generated repair and evaluating diagnostic selection together with repair weighting and learning from execution.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39409

Safe Multi-Objective Reinforcement Learning: Zero Constraint Violation and Primal-Averaging Convergence

Chenxi Li, Yao Shu, Siyang Gao, Shuang Qiu, Guangxin Jiang

原文摘要

Many applications of multi-objective reinforcement learning (MORL) require policies that balance conflicting objectives while satisfying safety constraints. We study safe MORL with zero constraint violation throughout learning, covering both cumulative and stepwise constraints. A key challenge is to explore sufficiently under unknown dynamics and constraint costs without violating the constraints. For a given preference over objectives, we develop an algorithm to learn a safe Pareto optimal policy by combining optimistic reward estimates with pessimistic cost bounds. With high probability, every policy executed over $T$ episodes incurs zero constraint violation. The algorithm returns a safe Markov policy induced by the averaged occupancy measure and achieves an $\widetilde{O}(1/\sqrt{T})$ suboptimality gap under smooth Tchebycheff scalarization. We further develop a preference-free algorithm that first explores safely without a prespecified preference and then computes a safe policy for any given preference without further interaction. Over $T$ exploration episodes and $K$ planning iterations, it retains the same safety guarantee and achieves an $\widetilde{O}(1/\sqrt{T}+1/\sqrt{K})$ suboptimality gap. Both algorithms require knowledge of a policy that strictly satisfies the constraints and its costs. We then relax these requirements, incurring at most constant cumulative violation. To our knowledge, we provide the first sample-efficient MORL algorithms that guarantee safety for every executed policy.

cs.LG · cs.AI · math.OC · stat.ML

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39403

SymKL: A Unified Representation-Space Measure of Signal Degradation

Dhananjaya Jayasundara, Sudarshan Rajagopalan, Vishal M. Patel

原文摘要

Quantifying degradation between a pristine signal and its degraded observation is fundamental to signal processing, yet existing full-reference measures are inherently modality-specific. We introduce SymKL, a unified full-reference degradation measure based on implicit neural representations (INRs). An INR is fitted once to the pristine reference and characterized by its diagonal Fisher information, which captures the local information geometry of the learned representation. Evaluating the Fisher information with the pristine reference and a degraded observation yields two parameter-sensitivity profiles whose symmetric KL divergence defines the degradation score. The proposed formulation requires only the reference and degraded observation, without knowledge of the corruption process or degradation severity, making it applicable across images, audio, and 3D data. We further derive an exact closed-form expression for the diagonal Fisher of SIREN, enabling efficient computation. Experiments on image, speech, and volumetric 3D benchmarks demonstrate that SymKL accurately tracks degradation severity, remains monotonic where conventional metrics saturate or fail, and provides a unified modality-agnostic measure of signal degradation. Code will be made publicly available after the review process.

cs.CV · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2609.13406

Generalized Agent Iteration: One Formal Framework for Iterative Policy Improvement and Recursive Self-Improvement

Hongyao Tang, Yi Ma, Pengyi Li, Yifu Yuan

原文摘要

When we speak of recursive self-improvement (RSI), are we speaking of a phenomenon, a mechanism, or a prospect? Towards autonomous and evolving intelligence, RSI is being claimed at many scales, while no single framework that formally describes these emerging instances exists. Its counterpart in the classical realm, iterative policy improvement, is characterized by generalized policy iteration (GPI), a framework of broad applicability with well-understood theoretical properties, but only where the update principle and the evaluation base lie outside the agent. In this paper, we propose Generalized Agent Iteration (GAI), a formal framework that describes iterative policy improvement and RSI as two cases of a single learning paradigm. GAI defines the agent as a configuration of modifiable components within a system and models the learning process as a cycle of agent evaluation and agent improvement. Two pivotal dials then distinguish the instances, in particular the systems that self-identify as RSI in recent work: whether the improving mechanism is part of the agent and whether the standard it is measured against is grounded outside it. The former dial delineates the boundary between GPI and RSI, and the latter determines a system's polarity as anchored, goal drift, or fully self-referential. Moreover, we use these coordinates to place existing systems on the same two axes and make the defects of recursive self-improvement statable one condition at a time. We see this paper as a first step toward exploring a formal characterization of RSI that rests on the classical account, makes existing systems comparable, and provides a principled basis for analyzing and designing new ones.

cs.AI · cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2606.06840

Characterize Then Distill: Mechanistic Reasoning in Large Output Spaces

Debjyoti Saha Roy

原文摘要

Reasoning-trained language models can select, zero-shot, a small set of relevant labels from thousands to hundreds of thousands of candidates. We ask how they do it mechanistically, and whether the mechanism can be distilled. We treat each decision as a token-level event scored by the model's own decision margin: the token that picks a coarse region of the label space, the tokens that pick a label within it, and the token where the output departs from a near-miss named earlier in the reasoning. Attribution, exact mean-ablation, knock-in into another example's context and a null calibration give individual attention heads causal standing. On clinical coding of MIMIC-IV discharge summaries, with all 5,651 candidate codes in context, a small, global, phase-structured set of heads is necessary and sufficient on essentially every summary; distinct head families read the candidate region and the earlier near-miss, and blocking what they read removes their effect; for mentions elicitable in the reasoning, a disjoint mid-layer set makes the region elicitable several tokens before the code. The structure recurs in a second reasoner. We introduce MISTILL, which, unlike chain-of-thought distillation, also supervises the student's pooled attention at these decision events. Read on heads found after training, it nearly doubles the causal recovery of the contrastive decision in a cross-family student and adds a small, seed-stable gain in one that already carries most of it, with no detected task difference at matched precision; under matched selection it exceeds a frozen random-head pool on both recovery and task score. A pre-registered evidence-flip test detected no improvement in evidence-dependent decisions through the tested routes.

cs.CL · cs.AI · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39395

OPERA: OPamp Design by an Expert-Taught Reasoning Agent, from Specifications to Layout

Vishal Saxena, Soumyajit Mandal

原文摘要

Analog flows from specifications to a signed-off layout rarely run without manual iteration. Recent large language model (LLM) agents size circuits with optimizer loops over the device dimensions and stop at the schematic. We present OPERA (OPamp design by an Expert-taught Reasoning Agent), in which a general-purpose LLM agent designs CMOS opamps from specifications to post-layout sign-off with open-source tools. First, the designer teaches the agent the gm/ID method: map the specifications to transconductances and minimum currents, choose gm/ID per device role from characterized lookup tables, and size from equations, leaving a few knobs to search and an explainable design. Second, the agent learns the gLayout framework and taught layout practice, and writes a reusable layout generator that closes the layout and extraction loop at designer-approved gates. In SKY130, the agent designed a shared constant-gm bias and six opamps (5T, folded-cascode and telescopic OTAs; Miller, folded-cascode and telescopic two-stage) for fun >= 50 MHz, PM >= 60 degrees, SR 25 V/μs into 1 pF and a minimum dc gain of 30 to 60 dB. All six layouts are DRC-clean and LVS-matched, and their extracted netlists meet every target at all 15 PVT corners ( fun >= 53.7 MHz, PM >= 62.1 degrees, SR >= 27.9 V/μs). The benchmark netlists are released at https://github.com/AMPICLab/AMPICBench.

cs.AR · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39385

UMI-Live for In-The-Wild Bimanual Robotics Data Collection

Govind Velamoor, Adrian Lin, Sam Jung, Spencer Mateega, Brandon Y. Yang, Amish Sethi

原文摘要

Universal Manipulation Interface (UMI) devices allow human experts to demonstrate tasks for robot learning without teleoperating physical robots. Using these demonstrations as training data still requires a consistent hardware setup, calibration, and data processing. We present UMI-Live, a full-stack open-source bimanual UMI system for in-the-wild data collection. We design our custom UMI-Live hardware to integrate camera and LiDAR--inertial sensing, onboard compute, and power, without external tracking equipment or calibration within the collection environment. We implement a simple user interface for recording data while an asynchronous pipeline performs parallel SLAM reconstruction and publishes training-ready datasets. We also release software that supports dual-YAM teleoperation and VLA policy training and inference. We validate UMI-Live data quality by fine-tuning $π_{0.5}$ and MolmoAct2, achieving success rates of 10/10 and 8/10, respectively on a bimanual pick-and-place task. Furthermore, we release 40 hours of UMI-Live data across three tasks, and complete hardware designs, software, and reproduction guides at our $\href{https://umi-live.github.io/}{\text{project website}}$.

cs.RO · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39380

Learning to Formulate: Agentic Hardware-Software Co-Design of AI Infrastructure through Progressive Optimization and Verification

Yi Ding, Li Shang

原文摘要

AI infrastructure, the chips and software that train and serve AI models, is designed in layers. Layers make the system manageable, but they also fix which choices an optimizer can consider. Gains that need the chip and its software to change together are excluded before optimization begins: the formulation leaves them out. Language-model agents make it possible to revisit these boundaries, but an agent's proposal cannot show by itself that a design is correct or optimal. We present Formulaic, a flow that divides the work. The agent writes the formulation, a mathematical solver optimizes within it, and verification and measurement check its assumptions and results. The agent refines the problem step by step, from the computation graph toward the chip. Each step is a template: a family of programs with the conditions under which they are correct. Wall relaxation searches the chip by pricing the hardware limits that bind the software. This paper takes a first step, through separate studies rather than one end-to-end run. On NVIDIA GB10, templates the agent wrote run complete Qwen3-8B layers $1.075\times$ faster than SGLang. In a simulated Ventus GPU study, joint design cuts the worst-scenario cycles of Qwen2.5-0.5B to 0.428 of the reference, against 0.934 for software alone. On a five-step GEMM refinement chain, reusing proofs makes checking $2.51\times$ faster.

cs.AR · cs.AI · cs.DC

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2609.02959

The Direction of Ignorance: LLMs Encode and Temper a Unigram Prior

Toni J. B. Liu, Jiajun Bao, Gurbir Arora, Mitchell Ostrow, Yizhou Liu, Raphaël Sarfati, Nicolas Boullé, Christopher J. Earls

原文摘要

What does a language model predict when it has few clues?The answer lies in its unembedding geometry: a single direction of the unembedding matrix encodes the unigram distribution of the training corpus, which serves as the Bayesian prior the model falls back on when uncertain. This structure, termed the direction of ignorance, appears across all model families examined (Llama, Qwen, Gemma, and Pythia), ranging from 0.4B to 405B parameters. Projecting the final prediction state onto this direction yields a per-token prior loading factor $λ$, which declines steadily as the context becomes more informative. Formally, the same projection decomposes the prediction state into two orthogonal vectors interpretable as the two components of a tempered Bayesian update: a unigram prior raised to the exponent $λ$ and a context-driven likelihood. This geometric probabilistic connection makes $λ$ meaningfully comparable across model sizes and families. Larger models generally exhibit lower prior reliance, and reasoning models exhibit lower reliance than their base counterparts. Finally, we show that $λ$ serves as a causal, inference-time control: reducing it can promote context-driven continuations and improve Llama-3.2-3B-Instruct on GSM8K, whereas amplifying it or applying matched perturbations along random directions degrades performance.

cs.LG · cs.AI · cs.CL · stat.ML

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39377

AgentDecept: Benchmarking Deceptive Behavior in Tool-Using LLM Agents under Pressure

Yehan Zhang, Qianqi Yan, Hang Yin, Ching-Chen Kuo, Shan Jiang, Ming Ouyang, Yijie Liu, Xin Eric Wang

原文摘要

LLM agents increasingly carry out multi-turn tasks across everyday applications, yet users often see only the agent's report rather than the observations, API calls, and state changes behind it. Existing deception benchmarks have studied long-horizon interactions and tool-use settings, but whether agents faithfully report their actions across realistic, stateful application workflows remains underexplored. To address this gap, we introduce AgentDecept, an executable benchmark for evaluating behavioral deception under pressure. AgentDecept contains 477 pressure variants derived from 257 source tasks across six pressure sources, each paired with a pressure-free counterpart. Interventions enter through event messages, task context, application state, tool responses, and service quotas, spanning explicit events and implicit task or environment changes. Our evidence-grounded protocol evaluates full trajectories by relating user-facing claims to time-ordered observations and executed outcomes, distinguishing falsification, concealment, and equivocation from honest failure and disclosed limitations. The resulting behavioral labels also provide a setting for testing existing activation-based probes. AgentDecept enables controlled analysis of when multi-turn agents' reports diverge from execution in realistic application interactions.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39373

What Makes Medical Examination Items Difficult? Using Large Language Models to Explore Patterns Associated With Empirical Item Difficulty

Xiaomeng Xiong, Marcus Walker, Yanlin Jiang

原文摘要

Research on medical item difficulty has largely focused on prediction rather than on identifying interpretable patterns associated with empirical difficulty. This study used empirical difficulty labels to guide large language models (LLMs) in inducing difficulty rubrics and examined how rubric guidance affected LLM difficulty judgments. Using training items and their empirical difficulty estimates from the United States Medical Licensing Examination (USMLE) and the Physician Assistant National Certifying Examination (PANCE), Claude Opus 5 induced separate rubrics for absolute and pairwise difficulty judgment. None of the induced rubrics treated surface features, such as stem length, option count, or images, as major sources of difficulty. Several patterns emerged across both examinations, including cue convergence, misleading cues, and option competition, although their specific forms differed. Retrieval of knowledge that could not be derived from the stem was particularly prominent in the PANCE rubrics, whereas the USMLE rubrics placed greater emphasis on integrating multiple clinical cues. We compared commercial and open-weight LLMs on held-out items under unguided and rubric-guided conditions. Rubric guidance changed how models used the difficulty scale but did not consistently improve performance. In absolute judgments, guidance encouraged broader use of the scale and generally improved coarse-grained agreement, without consistently improving rank ordering. In pairwise comparisons, its effects varied across models and examinations, and judgments became more sensitive to presentation order. These findings suggest that the value of label-guided rubric induction may lie more in generating interpretable, testable hypotheses from empirical difficulty data than in reliably improving difficulty judgments.

cs.CL · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39372

Cord: Confidence-Anchored Order Residual Decoding for Masked Diffusion Language Models

Chengwei Zhou, Zhaoyan Jia, Massoud Pedram, Gourav Datta

原文摘要

The decoding order of masked diffusion language models (dLLMs) can substantially affect generation quality, yet common decoders typically apply the same reveal rule throughout generation. Confidence provides a strong default, but it measures how decisively a position can be predicted now, not how useful revealing that position will be for subsequent decoding. We propose Cord, a lightweight prompt-conditioned policy that retains confidence decoding as a stable reference and learns when and how to depart from it through geometric residuals over generation direction, proximity to revealed context, and two-sided infilling. The residuals adapt to both the prompt and the evolving partial response. Rather than sampling a new policy action at every decoding step, Cord samples a single low-dimensional coefficient matrix per rollout, which induces an entire state-dependent reveal trajectory. This yields a contextual-bandit formulation trained directly from sequence-level rewards, without step-level credit assignment or backbone updates. Across three dLLMs and five reasoning, code, and constraint-satisfaction benchmarks, Cord improves average task success from 32.0% under confidence decoding to 48.2%, and further improves over the best per-domain fixed residual. Behavioral audits show that the learned residuals induce substantial, task-dependent changes in reveal locality and two-sided infilling. Cord adds only $0.53$M trainable parameters, requires no additional backbone forward passes, and incurs only 5.39% inference latency overhead.

cs.CL · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39358

Before the Next Rollout: Diagnosing Skill Utility Shifts from Policy Updates

Yifan Wang, Zhenyi Gai, Yiwen Zhao, Mingxuan Jiang, Keyang Chen, Zeping Li, Hongfeng Chai, Yixin Cao

原文摘要

Skill-evolution methods typically analyze execution trajectories to guide agent skill refinement. However, policy updates can change the behavioral effects and marginal utility of an unchanged skill. Attribution based on earlier trajectories may therefore no longer reflect the skill's contribution under the updated policy. Moreover, trajectory failures can arise from multiple factors, making individual skill contributions difficult to isolate. We evaluate whether policy-update signals contain predictive information about changes in skill marginal utility. At fixed contexts from pre-update training rollouts, we measure how policy updates change the contrast in token preferences between skill and control conditions. Combining these interaction shifts with advantage estimates yields reward-directed readouts for diagnosing utility changes. The readouts reuse existing training rollouts and require only forward passes for single-step action scoring, without additional post-update rollout evaluation. Our experiments show that policy updates can alter skill marginal utility and that policy-update readouts carry diagnostic information for identifying skills with declining utility. Integrating readout-guided prioritization into skill evolution improves downstream performance while reducing policy and editor token consumption.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39356

Semantic Mismapping in Generalized Category Discovery: Evaluation and Mitigation

Zitai Wang, Qianqian Xu, Cong Hua, Xi Chen, Zhiyong Yang, Xiaochun Cao, Qingming Huang

原文摘要

Generalized Category Discovery (GCD) aims to perform old-category recognition and novel-category discovery in the same unlabeled set. Early methods address both objectives through clustering and evaluate the exchangeable predicted indices with global Hungarian (GH) matching. Recent methods adopt trainable parametric heads, using supervision for old-category recognition and discovery-oriented objectives for new categories. In this way, the output space becomes partially exchangeable: old indices are semantically anchored, while new indices remain exchangeable. However, these methods still apply global matching, which can reassign anchored old indices to other categories and hide some prediction errors. Empirically, we are surprised to observe that this reassignment can induce an overestimation of new-category accuracy by up to 25.63 points. We term this phenomenon semantic mismapping and further derive a permutation-cycle condition that characterizes when semantic mismapping can increase the global matching score. To fix this issue, we propose Semantics-Preserving Evaluation (SPE), which preserves the supervised semantics of old indices and restricts Hungarian matching to the new region. Furthermore, we propose Mismapping-Mitigating Optimization (MMO) to alleviate the prediction errors that can give rise to semantic mismapping, combining a region-balance loss (RBL) for old/new routing with a novel-diversity loss (NDL) that encourages orthogonality among new prototypes. Theoretically, we show that minimizing the two losses can tighten an upper bound on spurious accuracy gains from semantic mismapping, with controllable residual terms. Empirical results under both GH and SPE demonstrate the effectiveness of our proposed method.

cs.CV · cs.AI · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39353

CoToan: Executable Rules and a Representation Contract for Reproducible Game AI Evaluation

Ha-Thanh Nguyen

原文摘要

Game AI results are comparable only when different implementations expose the same decision problem. Yet seemingly minor representation choices can silently change that problem by altering action probabilities or collapsing states with different outcomes. We study this issue through CoToan, an executable specification and representation contract for Vietnamese Mathematical Chess connecting the Game Description Language (GDL) with OpenSpiel. The game provides a useful case because a single capture may have multiple arithmetic justifications, while outcomes can depend on scoring rules and a finite playing horizon. CoToan therefore represents actions by their state transitions rather than by arithmetic witnesses, and treats time and victory configuration as part of task identity. Cross-representation tests show agreement in legal actions, successor states, and terminal outcomes across hundreds of sampled and constructed positions, complemented by exhaustive checks of all 32 operator masks and 512 opposing-material subsets. Controlled diagnostics further show that uniform witness-level sampling alters endpoint-action probabilities, while omitting remaining time or victory settings can merge states with different terminal behavior. These results show that action semantics, time, and victory configuration are not merely implementation details. They are part of the experimental task itself and must be represented explicitly for reproducible game AI.

cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2608.04565

Breadcrumbing Search Agents: Per-Turn Scheming Over Long-Horizon Trajectories

Xuebin Li, Hanqing Zhao, Siyuan Liang, Kejiang Chen, Weiming Zhang, Dacheng Tao, Nenghai Yu

原文摘要

LLM-based search agents are widely used for information-seeking tasks, but their reliance on external tool returns introduces a critical security risk: web content retrieved during execution is untrusted, exposing agents to prompt injection and goal hijacking. Prior work on search-agent safety primarily focuses on static web-content injection, but modern agents issue follow-up queries and cross-check competing sources, so a single injected page is often diluted or rejected. We show that the channel delivering search and page observations is a fragile security boundary: beyond exposing the agent to a single poisoned page, a mediated search interface can repeatedly steer how the agent gathers evidence and forms its final answer. Under a constrained tool-intermediary threat model, appending only one controlled result per query can substantially increase attack success when the evidence is coordinated across the agent's trajectory. We study this setting with a strategy-driven long-horizon attack system and introduce Authority-Chain Hijack (ACH), an expert-refined strategy that turns isolated search-result and page-content manipulations into a coherent evidence chain across seemingly corroborating sources. ACH achieves the highest Overall ASR among all baselines, reaching 55.9% / 83.3% ASR / MaxN ASR on the full SafeSearch test split. We further introduce Trace-Guided Strategy Evolution (TGSE), which automatically improves attacker strategies from execution traces, replacing manual redesign with trace-driven refinement; its strongest single setting reaches 71.4% / 95.0% in held-out evaluation. Code is available at https://github.com/MBOSSLESS/Breadcrumbing-Search-Agents.

cs.CR · cs.AI · cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39341

RetroChessBench: A Benchmark for Reconstructing and Proving Chess Histories

Tidor-Vlad Pricope

原文摘要

Frontier language models such as GPT 6 Astra have contributed to research mathematics, demonstrating substantial reasoning capabilities. Yet serious reasoning failures persist even in tasks governed by explicit rules. We demonstrate this in retrograde chess which is a playground for pure reasoning, where solvers reconstruct legal histories and distinguish possible events from forced ones. We introduce RetroChessBench, a 12-task benchmark covering exact-length constructions, unique endings, and exhaustive answer sets, with references, rubrics, and a terminal-reward environment. In two trials per task and condition, GPT 6 Astra High fails to produce the required final answer in 22/24 attempts without tools and 18/24 with code execution; mean terminal RL rewards are 14.75% and 27.19%, respectively. The examples identify faulty deductions: failed responses reject legal histories and treat temporary obstructions as permanent. Online retrograde archives often provide puzzles and final move sequences without detailed derivations. We hypothesize that limited exposure to these missing explanations in web-derived training data contributes to the observed reasoning failures. This questions how far frontier mathematical reasoning really generalizes beyond familiar solution patterns. Claims of general intelligence require evidence that models can transfer learnt principles to areas with few worked examples.

cs.LG · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2610.21565

BiShield-TEE: On the (In-)Security of Unilateral Weight Obfuscation in On-Device TEE-Shielded LLM Partition

Zhiyuan Wang, Yuezhi Zhou, Ziqi Zhang, Zhu Zhang, Zhixing Tan, Yongheng Deng

原文摘要

On-device deployment of large language models (LLMs) is increasingly adopted due to latency and privacy requirements, but exposes proprietary models to reverse engineering and intellectual property theft. Trusted Execution Environments (TEEs) provide hardware-enforced isolation for on-device inference, yet most TEE-based parameter obfuscation relies on unilateral matrix transformations that obfuscate only the output side. We identify a vulnerability in this design, termed the Solution Determinacy Property (SDP): the obfuscation secret is uniquely determined, exposing the structure of the true key and enabling attackers to break the protection. Under our attacks, representative methods GroupCover and ArrowCloak are reduced to negligible protection, with attack accuracy reaching over 2.4 times the full-shielding black-box baseline and recovering near-complete model functionality. To address this vulnerability, we propose BiShield-TEE, a bilateral framework that induces Ambiguous Solutions through joint obfuscation of input and output dimensions, hiding the true key among many equally valid candidates. Experiments across diverse models and datasets show that BiShield-TEE confines attack accuracy to the full-shielding black-box baseline, while achieving higher inference efficiency than prior methods on real hardware. Our code is available at https://github.com/laduiw/BiShield-TEE.

cs.CR · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2610.39331

Telling Is Not Seeing: Textual Personas Rarely Make Vision-Language Models Answer Like the People They Describe

Jiawei Chen

原文摘要

Vision-language models (VLMs) are increasingly prompted with personas, such as "You have low vision", to stand in for people in user studies. We ask whether a persona makes a model see like the person it describes. We compare persona answers with measured people with low vision, color blindness or hearing loss, and with simple rules that use only the facts the persona states. The persona never beats these rules, and it erases the real differences between groups and individuals. Describing blur leaves recognition unchanged in all 14 models we test, while blurring the photo lowers it as it does for people. Inside open models, the description changes neither the image information nor the answer probabilities. Yet the models know the effects they fail to show: they can estimate how blur and illusions affect people. Telling is not seeing: a persona makes a model say what it knows about people, not see what they see.

cs.CV · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.33775

The Hitchhiker's Guide to System 1 Models: Fast Decision-Making Beyond LLMS

Haggai Roitman

原文摘要

Large Language Models excel at reasoning and generation but their autoregressive, token-by-token inference makes them unnecessarily slow, expensive, and poorly calibrated for the many routine decisions (routing, gating, scoring, classification) that modern AI systems require at sub-second latency. A growing class of models addresses this gap: System 1 decision models, architectures that map an input directly to a typed, structured decision in a single forward pass. This guide provides a comprehensive, practitioner-oriented treatment of the field. We propose a five-family taxonomy covering encoder-based classifiers, reward models, non-autoregressive decision engines, contrastive decision models, and extraction models, and we provide in-depth coverage of each family's architecture, training methodology, and deployment trade-offs. A central theme is calibration: we develop the theory of proper scoring rules, analyze Reinforcement Learning for Calibrated Decisions (RLCD), and examine recently discovered phenomena including IIA violations, ordinal-scale bias, and few-shot recalibration under distribution shift. On the integration side, we describe how System 1 models serve as decision infrastructure within agentic workflows and System 1 + System 2 cascades that route between fast and slow models based on calibrated confidence. The guide covers 18 application domains, maps the rapidly growing open-source ecosystem (over 70 model reproductions in weeks), provides runnable code throughout, and includes a practitioner's checklist, an open-source model zoo, and 22 self-assessment questions with detailed answers. Spanning 190 references, it is intended as a unified entry point for researchers and engineers working at the intersection of fast decision-making, calibration, and production AI systems.

cs.AI · cs.CL

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2609.11234

NovGauge: A Fine-Grained Benchmark for Diagnosing LLMs' Capability in Paper Novelty Assessment

Guoqiang Zhang, Kexin Tan, Ming Zhang, Li Ju, Wenqing Jing, Zhonghan Yue, Jiayi Chen, Shiqiang Wu 等 15 位作者

原文摘要

Large language models (LLMs) are increasingly used in peer review at major AI conferences, yet novelty remains a persistent weak point. Existing benchmarks assess novelty as a single holistic score, making it difficult to diagnose which dimension a model misjudges or whether its evidence is faithful. We present NovGauge, a human-anchored benchmark for fine-grained novelty assessment diagnosis. The benchmark contains 619 paper pairs and 50 multi-paper sets, drawn from two expert sources: ICLR reviewer overlap claims and survey co-citations. Instances are independently labeled along three dimensions: task, problem, and method, capturing application goals, technical challenges, and solution approaches. We propose a cascading diagnostic pipeline that verifies per-dimension correctness, evidence grounding, and logical support. Evaluation of 18 LLMs shows hallucination rates ranging from 0% to 39% across dimensions, and among non-hallucinated correct-positive judgments, over 70% cite evidence fails to logically support the stated reason. The best-performing model, GPT-5.5, achieves 43-72% Verified F1 across dimensions, while most models retain less than half of their raw F1 after faithfulness verification. These results suggest that current LLMs remain far from reliable scientific novelty assessment, particularly when correctness is conditioned on faithful evidence grounding.

cs.AI

查看论文
arXiv ↗PDF ↗