跳到正文
研踪PaperTrail

论文库

检索计算机领域论文,按研究方向、作者或关键词筛选。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+197新增论文
+151其中 AI 相关
+16已有论文修订
+91新增 AI 导读

arXiv 最近完成扫描:2026-11-22 06:41 · 全站最近入库:2026-11-21 12:19各信源进度 →

统计口径与数据新鲜度

统计 2026-11-21 08:06 — 2026-11-22 08:06(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:2026-11-22 06:41最近发现新增版本:2026-11-21 12:19库内论文最新更新时间:2026-11-20 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。

29,158 已收录论文24,181 AI 相关8,992 有修订112 HF 社区发现

浏览论文库

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 21,490 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-22 06:39(北京时间)
新论文 · v1arXiv:2611.20196

DoF-Query Transformer for Adaptive Fusion of Egocentric Vision and Forearm HD-sEMG in Continuous Hand Kinematics

Jiahang Liu, Wenbo Zhang, Junsong Wang, Zhenzhi Ying, Siqi Hou, Naohiko Sugita, Liming Shu

原文摘要

Continuous and accurate estimation of human hand kinematics is important not only for intuitive human-machine interaction, but also for capturing fine-grained human manipulation demonstrations that can provide action supervision for embodied-intelligence datasets and dexterous robot learning. However, individual sensing modalities may be insufficient for reliable estimation under varying operating conditions. We present a finger-unencumbered multimodal framework for continuous regression of 20 hand joint angles from egocentric RGB images and 128-channel forearm high-density surface electromyography (HD-sEMG). The framework combines a Vision Transformer-based visual encoder, a Transformer-based EMG encoder, and a degree of freedom (DoF) query Transformer fusion decoder. To account for the varying informativeness of visual and EMG cues across joint DoFs and input conditions, the decoder assigns a separate learnable query to each DoF, enabling DoF-specific and context-adaptive fusion of the two modalities. The framework was evaluated on synchronized multimodal data from 10 participants. Averaged across participants, the proposed fusion model achieved a mean coefficient of determination ($R^2$) of $0.88$ and a mean absolute error (MAE) of $2.80^\circ$, compared with $0.82$ and $3.21^\circ$ for the vision-only model and $0.68$ and $5.13^\circ$ for the EMG-only model. It also outperformed the evaluated representative unimodal baselines and alternative fusion designs. DoF-wise analyses showed larger fusion gains for visually ambiguous joints and viewing conditions, while synthetic visual occlusion demonstrated greater robustness than vision-only estimation. These results highlight the complementary value of egocentric vision and forearm HD-sEMG for capturing fine-grained human hand kinematics, with potential applications in human demonstration acquisition and dexterous robot learning.

cs.CV · cs.HC · cs.LG

新论文 · v1arXiv:2611.20195

How Specification Explicitness Shapes Failure in LLM-Generated Code

Elmira Onagh, Maleknaz Nayebi

原文摘要

When LLMs turn requirements into code, every underspecified constraint becomes behavior that the model must decide during generation. We test when specifications move beyond structure and begin to control generated behavior. In a controlled experiment, 72 participants implemented and revised Python tasks with an LLM under three levels of specification explicitness. The results show a clear break between lightweight structure and explicit behavioral obligations. Lightweight structure reduced hidden-test failures by only 8.7%. Requiring participants to specify constraints, invariants, negative requirements, edge cases, tie-breaking rules, and exception handling cut failures by 47.2% and reduced the number of distinct failure modes by 20.5%. The largest drops were in failure modes tied to implicit obligations. Invariant violations fell by 66.1%, and unsupported assumptions fell by 59.5%. After a requirement change, specifications with explicit behavioral obligations produced regressions in more submissions, but each affected submission lost less previously correct behavior. Explicit specifications also changed how developers judged generated code. Without explicit obligations, participants overestimated generated-code correctness by about 23 percentage points. With explicit obligations, they produced better code but judged it more cautiously. Thus, the lesson is sharper than "better specifications help." Lightweight structure barely helped, while explicit behavioral obligations nearly halved failures, made regressions shallower, and improved calibration.

cs.SE · cs.AI

新论文 · v1arXiv:2611.20192

GlyphAtlas: Disentangling Stroke Detail from Layout for Scene Text Synthesis

Luozheng Qin, Changhai Zhou, Yiyi Zhou

原文摘要

Scene text synthesis requires stroke-accurate characters rendered in a specified layout, as well as visual coherence with the surrounding scene. To achieve this, existing methods commonly pre-render the target text at its intended locations to form a glyph image that jointly specifies layout and glyph structure. However, this representation ties the resolution of each glyph to its target region, limiting fine-grained stroke guidance for small, densely packed, or structurally complex text. To tackle this, we propose GlyphAtlas, a scene text synthesis framework that retains the glyph image as a spatial anchor and supplies fine-grained stroke references through a dedicated glyph atlas. The atlas renders and encodes each character or word in a separate cell, decoupling reference resolution from target-region size. We further introduce atlas-anchored attention, which lets each text region retrieve stroke detail only from its corresponding atlas cells to refine the coarse glyph structure provided by the glyph image. This coarse-to-fine conditioning simplifies stroke rendering while preserving explicit layout correspondence. We optimize GlyphAtlas with a mask-weighted flow matching objective and a timestep-weighted text perceptual loss for writing accuracy and background consistency. Random resolution augmentation of atlas cells supports adjustable atlas budgets at inference. Extensive experiments demonstrate that GlyphAtlas achieves stroke-accurate synthesis of small, dense, and structurally complex text, zero-shot transfer to languages unseen during task-specific training, and leading performance in scenes with many text regions. project page: https://fr0zencrane.github.io/project-glyphatlas/

cs.CV

版本更新 · v4arXiv:2604.24197

Seeing Is No Longer Believing: Frontier Image Generation Models, Synthetic Visual Evidence, and Real-World Risk

Shuai Wu, Xue Li, Zhijun Wang, Bolun Liu, Weilin Cai, Zihao Su, Ran Wang

原文摘要

Image generation systems can produce plausible photographs, readable documents, and consistent depictions of people and places. Presented as records of real events, these outputs can influence decisions in news, finance, identity verification, medicine, and law. This narrative review examines public model documentation, incident reports, empirical studies, governance materials, and illustrative English and Chinese practitioner commentary available through 1 October 2026. We separate vendor capability claims, documented incidents, experimental findings, and prospective harm pathways. Realism, text rendering, reference consistency, conversational editing, search grounding, and low generation cost shape the conditions under which synthetic images acquire evidentiary weight. Documented episodes show how fabricated visuals enter public and institutional workflows, although public reporting cannot measure misuse prevalence for current releases. Comparing provider safeguards, provenance credentials, watermarks, platform labels, and statutory duties shows that origin signals do not verify a depicted event or transaction. Our framework matches artifact types and decision stakes to specific controls. Consequential decisions require authenticated source records, independent channel corroboration, and proportionate review before action.

cs.CL · cs.AI

版本更新 · v5arXiv:2610.18826

WorldOracle: Benchmarking Action-Conditioned World Models against Real-World Executions

Haoyi Duan, Liliang Chen, Nan Wang, Yang Zhang, Yuhang Lu, Xiaotong Chen, Wei-Cheng Tseng, Shenyuan Gao 等 16 位作者

原文摘要

We introduce WorldOracle, a benchmark specifically designed to evaluate action-conditioned world models (ACWMs) against paired real-world executions. Given the same initial observation and low-level action sequence, WorldOracle asks whether a predicted rollout reproduces what happens in the real world, including physical consequences and task-state transitions. We evaluate models along four complementary dimensions: physical state correctness, interaction spatial alignment, key interaction area fidelity, and multi-view action coherence. To support this evaluation, we curate real-world training and evaluation suites that cover different interaction dynamics and embodiments. The evaluation suite contains long, multi-segment episodes with both successful and failed executions, enabling evaluation across varied action outcomes and behavior distributions. Experiments on seven representative ACWMs reveal distinct model strengths and failure modes. WorldOracle provides a systematic foundation for measuring and advancing action-conditioned world modeling for manipulation. We release the datasets, evaluation code, and leaderboard: https://worldoracle-benchmark.github.io/.

cs.RO · cs.CV

新论文 · v1arXiv:2611.20190

A Line is Not Just a Line: Visualizing Harmonious Complex Line Sets

Francesca Samsel, Daniel F. Keefe, Gregory Abram

原文摘要

Line is a foundational building block in visual design for traditional media and computer graphics. Controlled variation in line quality can be used to define form, create abstractions, distinguish categories, control affect, and more. Yet, when it comes to computer graphics visualizations of science, particularly those built upon 3D computer graphics, too often a line is just a line. This portfolio presents a series of visual design studies and lessons learned that explore the diverse character of line that might be useful for 3D visualizations. The imagery begins to show how more sophisticated use of this powerful visual building block could increase the clarity of data visualizations while also bringing a human, handcrafted aesthetic that reinforces the people and impacts behind scientific climate data.

cs.HC · cs.GR

版本更新 · v2arXiv:2611.15917

Owned Edits to a Shared Decision Rule: Learning Private Exceptions as Typed, Gated, Withdrawable Weight Edits

John J. O'Hare

原文摘要

Organisations that apply the same published decision rule often hold private evidence that the rule is wrong for their own cases, and will not pool it. We describe a learning mechanism for that setting whose unit of change is an owned edit. The published rule is compiled exactly into a small ReLU network, a structured prior. Each party's learner searches a small typed space of declarations, a few named values such as subject, scope, form and size, and proposes one only when its own records pass a selection-aware evidence gate. The learning is that gated search around the prior. In two pre-registered analyses, on synthetic data whose 20 hidden truths were sampled from that typed space, the learner at 30 records per party was non-inferior, within 2 points, to all seven tuned baselines (a confined gradient learner, Ditto, FedRep, federated averaging, two pooled networks and pooled boosted trees), scoring 96.6% against 91.7% for the unchanged rule; a descriptive greedy ablation with no prior, over a separate space, scored 84.0%. In a third registered test, typed edits over a network trained on shared records beat that network alone by +5.4 points and the same edits over the published rule by +3.1; oracle-tuned fine-tuning and Ditto matched it at 30 records. An acceptance body recomputes each edit from its declaration and writes it into units reserved for that party, so by construction an edit changes no decision about any other subject, withdraws exactly, discloses at most log2 of the typed space's size in bits, and the artefact can hide nothing beyond its declared values; timing is not covered. We report where it loses: truths shared by all parties, relaxations, ten records, a pre-stated gate criterion that failed in both runs, and a language model given the records that decided about as accurately on one draw. On a rule-only base, a table of owned exceptions makes the same decisions.

cs.LG · cs.CY

新论文 · v1arXiv:2611.20187

Cyber-Secure Wearable, Ultra-Low-Power Networked Sensors for Soldier Health Monitoring: Insights from the CAPSARII project

Luciano Bozzi, Christian Celidonio, Umberto Nuzzi, Massimo Biagini, Stefano Cherubin, Asbjørn Djupdal, Tor Andre Haugdahl, Andrea Aliverti 等 22 位作者

原文摘要

The CAPSARII project advances a cyber-secure, ultra-low-power wearable platform and Internet of Battlefield Things (IoBT) framework to support real-time monitoring of military personnel and data-driven decision support. The architecture integrates smart-textile sensing, a body/squad area mesh, edge-deployed AI on secure RISC-V microcontrollers, and cloud/federated analytics for cross-mission learning and comparative studies. Two representative use cases--pilot fatigue and warfighters operating in harsh environments--guide requirements, validation, and interface design (including HUD and haptic feedback). As an initial step toward embedded models, we benchmark stress classification on a public wearable dataset: tree-ensemble baselines achieve state-of-the-art performance for accuracy and F1 (e.g., XGBoost: $\mathrm{Accuracy}=0.901\!\pm\!0.023$, $\mathrm{F1}=0.920\!\pm\!0.019$), while a feature-based Transformer attains the highest AUC ($0.963\!\pm\!0.025$), informing feature-engineering and compression choices for edge deployment. Building on these results, CAPSARII employs HW/SW co-design, model compression (pruning, quantization, distillation), and precision tuning to meet stringent energy and memory budgets; it further investigates Processing-in-Memory accelerators for future efficiency gains. A structured gap analysis highlights the remaining challenges: establishing validation standards for physiological signal quality and textile durability; achieving low-latency, scalable BAN/SAN communications under mobility and interference; and deploying hybrid lightweight/post-quantum cryptography with robust device identity and side-channel resilience on constrained endpoints. By consolidating these elements into a coherent, secure IoBT stack, CAPSARII delineates a practical roadmap from laboratory baselines to mission-ready soldier monitoring in both training and operational scenarios.

cs.ET · cs.AI · cs.AR · cs.CR · cs.HC · cs.LG

新论文 · v1arXiv:2611.20185

RoboKit: Accelerating Parallel Differentiable Optimization for Robot Learning

Jianglong Ye, Xuanbin Peng, Conor Mc Gartoll, Changwei Jing, Lai Wei, Jiali Chen, Xueyan Zou, Xiaolong Wang

原文摘要

We present RoboKit, a high-performance, general-purpose toolkit for parallel differentiable optimization in robotics, built on three principles: (i) Performance and flexibility: GPU-parallel multi-seed solvers pair sampling with versatile optimizers, running 2-60$\times$ faster than task-specific baselines across seven tasks; (ii) Geometry, rendering and physics: hybrid and coarse-to-fine representations accelerate collision, differentiable rendering and simulation, optimizing 4096 grasps in 2.97 s; (iii) Integration with learning: costs are differentiable and simulation is parallel, so RoboKit both distills optimization into policies and trains RL; distilled policies run 9-240$\times$ faster at comparable accuracy, and RL reaches a GPU baseline's scores up to 2.1$\times$ sooner on the same simulator.

cs.RO · cs.AI · cs.LG

新论文 · v1arXiv:2611.20183

GeoReVis: Learning Selective Revision and Evidence Consolidation for Remote Sensing Segmentation

Ke An, Liang Chen, Yupei Wang

原文摘要

Recent methods try to inspect and revise masks to mitigate referential ambiguity and localization errors caused by complex spatial relations in language-guided remote sensing segmentation. However, these solutions often lead to Revision Degradation: revising missing regions may introduce errors into previously correct ones. Even a globally better mask can omit correct regions found in inferior masks. We term this Evidence Fragmentation: correct target evidence is scattered across imperfect masks. All masks generated by the model at this time are suboptimal. We propose GeoReVis, which unifies revision guided by execution feedback, Candidate Selection, and Cross-Mask Evidence Consolidation. It learns fine-grained revision from real actions that improve segmentation when executed and learns to select a better mask through quality comparisons between actual candidates, reducing the risk of degradation during revision. The selected whole mask can still discard useful regions. GeoReVis therefore retains all candidates across rounds, partitions their spatial disagreements into composable regions, and jointly reconstructs regions conditioned on the full instruction and context. This makes complementary and fragmentary evidence reusable regardless of its source candidate's overall quality, yielding improved segmentation. We further build a data pipeline to generate fine-grained multi-round segmentation trajectories and derive supervision for revision actions, candidate selection, and region composition from the model's own outputs, extending training beyond whole-mask IoU supervision. We also propose ReVis-Eval, a benchmark protocol evaluating multi-round revision and cross-round evidence consolidation accuracy. Experiments on ReVis-Eval and 5 benchmarks for remote sensing reasoning and referring segmentation demonstrate the effectiveness of GeoReVis.

cs.CV

版本更新 · v3arXiv:2607.11250

Do Multi-Agent LLMs Explore Each Other?

Hyeong Kyu Choi, Jiatong Li, Wendi Li, Xin Eric Wang, Sharon Li

原文摘要

Exploration is essential for reliable autonomy in multi-agent systems, yet it remains unclear whether large language model (LLM) agents can explore effectively when interacting with one another. We show that modern LLM agents often fail to do so, exhibiting myopic and polarized interaction patterns that lead to suboptimal coordination and increased regret. We formalize this challenge as the Multi-Agent Exploration problem, modeling it as a partially observable stochastic game (POSG), in which agents must probe peers to infer their capabilities and identify effective interaction strategies. To address this, we introduce Multi-Agent Contextual Exploration (MACE), a lightweight framework that explicitly promotes exploration through structured peer selection. Across both contextual and parametric diversity settings, exploration behavior and downstream task performance substantially improve when MACE is applied. We further provide theoretical insight that MACE achieves a sublinear expected-regret bound. Overall, our results highlight a fundamental limitation of current LLM agents and underscore the importance of explicitly guided exploration for reliable multi-agent autonomy.

cs.MA · cs.AI

版本更新 · v2arXiv:2609.14424

Shot noise can decide capacity comparisons of quantum models

Pavel Sulimov, Claude Lehmann

原文摘要

Shot noise can decide close comparisons of what quantum models appear able to learn. On an IBM processor, the standard estimate of a capacity measure, the effective dimension, placed a 48-qubit model above about 90% of classical networks with as many parameters at 256 shots per circuit. The split-shot estimate from the same data, which removes the bias of the Fisher matrix and needs no noise model, placed it above 12%. A registered run at six parameter points of a 16-qubit model repeated this: the standard estimate placed the three random points above every classical network, the split-shot estimate above 11 to 22%. In simulation the standard estimate reversed the comparison with the classical median at all 100 random points tried. The inflation is the size that shot noise predicts, and a forecast from a pilot run bounds it from above. For squared loss and a linear readout, the error of the standard parameter-shift estimate of a model's Fisher matrix is an exact polynomial in single-shot statistics. Its leading term grows with the effective rank of the matrix, so the shot budget for a stated precision follows from a pilot run; on hardware the observed errors were 0.85 to 1.09 times the predicted ones. The effective dimension can be bracketed from the data but not pinned down at a practical budget. Capacity reports should state their shot count and use the split-shot estimate.

quant-ph · cs.LG

新论文 · v1arXiv:2611.20180

UrbanROI: Causal Modeling for Regional ROI Forecasting under POI Entry

Zhaoqi Zhang, Miao Xie, Linyou Cai, Qianlong Xie, Xingxing Wang, Siqiang Luo, Gao Cong

原文摘要

As urban commercial environments continue to evolve rapidly, accurately modeling the economic impact of point of interest (POI) entry is essential for supporting data-driven advertising allocation, merchant onboarding, and regional business decision-making. Existing return on investment (ROI) prediction methods mainly focus on individual ads or merchants, overlooking the interaction effects among POIs and failing to capture the causal influence of structural changes in urban commercial ecosystems. In this paper, we introduce a novel research problem -- regional ROI forecasting under POI entry, which aims to predict the future ROI trajectory of a commercial region after the introduction of a new POI by modeling temporal ROI evolution and competition-driven interactions among surrounding POIs. To address these challenges, we propose UrbanROI, an intervention-aware causal forecasting framework for region-level economic outcome prediction. UrbanROI leverages an Interaction-aware Regional Graph Module to explicitly model competitive relationships among POIs based on user request contexts, spatial proximity, and request-level commercial signals, and constructs structurally aligned treatment and control graphs to simulate factual and counterfactual regional states. Extensive experiments on large-scale real-world POI and advertising datasets demonstrate that UrbanROI consistently outperforms strong baselines in post-entry regional ROI forecasting. The results demonstrate its effectiveness in temporal economic modeling, inter-POI interaction representation, and intervention-aware forecasting, providing support for regional economic analysis and business decision-making in dynamic urban commercial systems.

cs.LG

版本更新 · v3arXiv:2605.26321

Anchor: Preventing Artifact Drift for Reliable Agent Verification

Abhijay Rana, Maksim Ivanov

原文摘要

Reliable evaluation of long-horizon agents requires verifiers that match what the task actually asks. Verifiable rewards have driven progress in domains like math and coding, yet building similarly reliable environments for open-ended enterprise workflows repeatedly struggles to balance realism, verifiability, and scale. This creates a validity problem we call artifact drift: when instructions, environments, oracles, and verifiers are created separately, they frequently fail to agree on what a task requires, resulting in environments that are unsolvable, reward-hackable, or inconsistent. We introduce Anchor, a benchmark-construction pipeline that formalizes domain expert specifications of business workflows into constrained optimization programs. From a single parametric specification, the pipeline jointly produces a natural-language instruction, environment configuration, solver-certified ground-truth solution, and state-based verifier. With this pipeline, altering parameters yields new tasks with controlled difficulty and known optimal solutions, producing harness-agnostic environments whose rewards depend solely on end-state business correctness. We apply Anchor to produce ERP-Bench: a benchmark of 300 long-horizon tasks spanning procurement and manufacturing workflows in a production-grade ERP system. Our pipeline organizes tasks into diverse families with calibrated difficulty profiles, and allows generated environments to evaluate coding, browser, and computer-use agents using a shared verifier. Across these harnesses, we find that frontier models follow all rules and instructions in 25.9% of trials and reach fully optimal decisions in only 17.4% of trials. Overall, Anchor and ERP-Bench offer a concrete recipe for building auditable evaluation environments whose verifiers remain aligned with the economically valuable tasks they are intended to score.

cs.AI

新论文 · v1arXiv:2611.20174

Weights & Caches: LLMs Can Program Tabular Foundation Models

Shi Bin Hoo, Magnus Buehler, Omar Swelam, Frank Hutter, Lennart Purucker

原文摘要

Neural networks are typically programmed through optimization, rather than by explicitly specifying their computations. Recent work on weight-space programming suggests that large language models (LLMs) can instead reason about and manipulate neural-network weights directly. In this work, we show that LLMs can directly program statistical computations in tabular foundation models (TFMs), whose parameters enable inference across different datasets. We investigate three complementary settings using GPT-6 Astra, without gradient-based optimization. First, Astra directly programs nanoTabPFN from zero-initialized weights, achieving a mean ROC AUC of 0.849 across 20 TabArena datasets, compared to 0.876 for a pretrained baseline. Second, Astra diagnoses and repairs a signed-product regression failure in pretrained TabPFN-3.5, restoring near-perfect predictions on held-out synthetic datasets while preserving performance on TabArena-Lite, with only a 0.08% average relative increase in error. Third, it writes KV-cache-editing programs that steer TabPFN-3.5's predictions on unseen synthetic datasets, demonstrating transformations such as reversing linear trends and converting sine waves into approximate square waves, without modifying model weights. Together, our findings suggest that LLM-driven programming could complement gradient-based optimization across TFMs' development and usage, while offering a new way to investigate and intervene in their internal computations.

cs.LG

版本更新 · v3arXiv:2611.03100

LUCA: A Synthetic World for Experimental Reasoning

Anna Gogleva, Thomas Charman, Martin Balla, Karl Gemayel

原文摘要

[Note, November 2026: the experimental results in this version are affected by implementation issues in the simulation environment and should not be relied upon pending correction. See the author note on page 1.] An experimental agent can succeed for the wrong reasons. Evaluations, however, often focus on the outcome, not the experiments, evidence, and inferences that produced it. We introduce LUCA, a configurable synthetic world for evaluating this process under fully known ground truth. Agents learn unknown regulatory systems through budgeted experimentation, then predict unseen interventions, diagnose hidden faults, and attempt rescues. This setup allows us to evaluate experiment selection, prediction, diagnosis, and intervention separately. Across seven frontier LLMs and three baselines, these capabilities dissociate sharply. Agents frequently rescue malfunctioning systems while diagnosing the wrong fault; diagnosis varies 7x across models while prediction varies only 1.4x. Models also adopt surprisingly different experimental strategies, and greater exploration does not reliably yield better performance. In one failure mode, agents often perform the experiment needed to distinguish competing causes, but still fail to use the resulting evidence. LUCA turns these hidden failures into measurable behaviours, providing a controlled environment to study where experimental agents succeed and fail. Code: https://github.com/RelationRx/luca.

cs.AI

版本更新 · v2arXiv:2609.12099

Representation Before Weighting: Semantic Retention in Frozen 3D Foundation-Model Fusion

Silas Kwabla Gah, Ebenezer Owusu

原文摘要

Frozen foundation models are increasingly composed at inference time, yet a common implementation shortcut collapses each source to a Top-1 label before heterogeneous evidence can interact. We investigate this representation bottleneck in frozen two-source 3D semantic fusion, combining dense RegionPLC predictions with sparse cross-view promptable segmentation evidence. Source outputs, geometry, scene lists, vocabulary, and evaluation protocol are held fixed while the number of retained semantic alternatives is varied. Established information-theoretic results characterize hard routing as a deterministic information bottleneck and show that the total-variation error of normalized Top-$k$ retention equals the discarded probability mass. On 156 held-out ScanNet200 scenes, under a fixed uniform linear pool, Top-1 retention reaches 28.47 harmonic-mean (HM) IoU, Top-20 reaches 35.49, and Full retention reaches 34.87: keeping every alternative is worse than keeping twenty. Paired scene-level bootstrap comparisons show that Top-20 exceeds Top-10 by $+0.289$ HM (95\% CI $[+0.093,+0.475]$) and Full by $+0.617$ HM ($[+0.204,+1.086]$). The ordering also depends on the operator: under geometric pooling it reverses and Full retention is best, at 36.38 HM with a fit-selected exponent. Top-20 and Full differ on 9.57\% of valid points, exclusively at locations where the two sources disagree at Top-1. These changes yield 746,318 net point corrections and are associated with greater discarded dense-source tail mass. On 50 ScanNet++ validation scenes, distribution-preserving retention also outperforms Top-1, reaching up to 26.92 HM versus 23.02. Within ScanNet200 novel classes, Full favors the rarest frequency tertile, whereas Top-20 favors the most frequent tertile.

cs.CV

版本更新 · v3arXiv:2602.04585

ImmuVis: Hyperconvolutional Foundation Model for Imaging Mass Cytometry

Dawid Uchal, Marcin Możejko, Krzysztof Gogolewski, Piotr Kupidura, Fabian Ozga, Szymon Łukasik, Jakub Giezgała, Tomasz Nocoń 等 17 位作者

原文摘要

We present ImmuVis, a family of efficient foundation models for imaging mass cytometry (IMC), a high-throughput multiplex imaging technology that handles molecular marker measurements as image channels and enables large-scale spatial tissue profiling. Unlike natural images, multiplex imaging lacks a fixed channel space, as real-world marker sets vary across studies, violating a core assumption of standard vision backbones. To address this, ImmuVis introduces marker-adaptive hyperconvolutions that generate convolutional kernels from learned marker embeddings, enabling a single model to operate on arbitrary measured marker subsets without retraining. We pretrain ImmuVis on the largest dataset to date, IMC17M (28 cohorts, 24,405 images, 265 markers, over 17M patches), using self-supervised masked reconstruction. ImmuVis outperforms state-of-the-art baselines and ablations in virtual staining and downstream classification tasks at substantially lower compute cost than transformer-based alternatives, and is the sole model that provides calibrated uncertainty via a heteroscedastic likelihood objective. These results position ImmuVis as a practical framework for real-world IMC modeling.

cs.CV

版本更新 · v2arXiv:2608.08073

eIRWR: Enhanced Iterative Random Walk with Restart for Scalable Root Cause Analysis in Microservices

Saiful Khan, Afrah Farea

原文摘要

Root cause analysis (RCA) in microservice systems must find the faulty service behind symptoms that spread across hundreds or thousands of dependent services. Graph-based random walk methods propagate anomaly evidence over the service dependency graph. However, existing anomaly-restart walks restart from the raw anomaly vector, which is often dominated by strongly anomalous downstream victims rather than the less anomalous source. Through a controlled ablation of a basic anomaly-restart walk, the Iterative Random Walk with Restart (IRWR), we show that its ``resilience damping'' is equivalent to a higher restart probability in Personalized PageRank (PPR). Basic IRWR is therefore no stronger than a restart-tuned PPR, which we adopt as a fairer baseline. We then present Enhanced IRWR (eIRWR), which goes beyond restart tuning. It (a)~concentrates restart mass on the most suspicious nodes through power-law sharpening and (b)~adds self-loops and backward edges so that probability accumulates at cascade sources. On three large topologies (12K--25K nodes) from the Alibaba Microservice Trace Dataset, eIRWR is the most accurate of the compared graph-based methods at every level of root-cause visibility. At visibility $v = 0.3$, it reaches a Mean Reciprocal Rank (MRR) of $0.78$, about four times a restart-tuned PPR and nine times the best other baseline. At high visibility, it reaches MRR\,$= 0.95$ and runs in under 30\,ms on graphs with 17K nodes, so it is suitable for online use.

cs.DC · cs.SE · eess.SY

新论文 · v1arXiv:2611.20158

Who Can You Hide Behind? Author-Conditioned Style Transfer and the Evasion of a Commercial AI-Text Detector

Mohd Ruhul Ameen, M. Saifuzzaman Rafat, Akif Islam

原文摘要

AI-text detectors can be sensitive to changes in writing style. We examined this sensitivity by rewriting 991 AI-generated answers with Gemini, using prompts that included a writing sample from one of 55 named authors. We scored each original and its rewrite with the QuillBot AI detector. The median score fell from 100% to 0%, and the mean fell from 67.9% to 11.2%. Of the rewrites, 66% received a score of exactly 0%. Among the 699 originals scored at least 50%, 71% of rewrites scored below 10% (paired Wilcoxon signed-rank test, p-value below 1e-100). Scores varied across the authors used in the prompts. All rewrites associated with 10 authors scored 0%, whereas the mean score remained above 50% for six authors (Kruskal-Wallis test, p-value below 1e-80). Greater reuse of five-word sequences from the author sample was associated with lower scores, although rewrites with less than 5% reuse still averaged 15.2%. These results describe one detector and one rewriting system, not detectors in general. The study lacks a generic-rewrite baseline, did not measure preservation of meaning, and produced rewrites that were substantially longer than the originals.

cs.CL

版本更新 · v2arXiv:2611.17215

Faster Deterministic All-Pairs Shortest Paths with Real Edge Weights

Isaac Barouch Essayag, Aryeh Lev Zabokritskiy

原文摘要

We give a deterministic algorithm for all-pairs shortest paths in directed $n$-vertex graphs with arbitrary finite real edge weights and no negative cycles. It computes all distances in $O(n^{2.9929729})$ worst-case time, improving both the recent deterministic bound $O(n^{2.9997})$ and the Las Vegas expected-time bound $O(n^{2.993805})$. We use a model with exact real addition, subtraction and comparison at unit cost; auxiliary integer arithmetic and indexed memory access are charged at bit cost.

cs.DS

新论文 · v1arXiv:2611.20157

Kernel-Guided Learning of Lifted Sparse Polynomial Models of Nonlinear Dynamical Systems

Mohammad Itani, Timm Faulwasser

原文摘要

Polynomial models of dynamical systems enable constructive analysis via sum-of-squares programming. Yet, many physical systems involve non-polynomial nonlinearities which for certain classes of systems may be inferred from the underlying physics even when the exact dynamics are unknown. We show that fitting lifted polynomial models directly to lifted data is problematic, since the exact lifting constraints render the lifted derivatives linearly dependent. Instead, we learn the dynamics in the extended state, i.e., the state together with the known nonlinearities, and obtain the dynamics of the nonlinearities via the exact lifting relations. To avoid the combinatorially large libraries that sparse regression requires when the polynomial degrees are unknown, we propose KINDy, which selects the degree of each component by kernel ridge regression before applying sparse regression to the correspondingly reduced library. We prove that, for sufficiently small noise and regularization, provided the data satisfy a rank condition, the selected degree never underestimates the true degree. On the cart-pole system, KINDy recovers the true degrees at low noise levels and achieves speedups of up to 8.5 over sparse regression on the full library, while in almost all cases yielding models that are at least as sparse and have lower or equal test errors.

eess.SY · cs.LG

版本更新 · v2arXiv:2611.16013

Runtime Analysis of Cartesian Genetic Programming on Max: A Proven Exponential Speedup

Duc-Cuong Dang, Roman Kalkreuth, Andre Opris

原文摘要

Genetic Programming (GP) is a search paradigm inspired from natural evolution for the automated discovery of expressions, functions, and computer programs. Cartesian GP (CGP) is a flavor of GP that uses a graph-based model to encode candidate programs, as contrasted to the conventional Tree-based GP (TGP). Since its inception two decades ago, CGP has predominantly been analyzed empirically, thus little is known regarding its theoretical performance guarantees. This paper analyzes CGP for the Max problem, which has been only rigorously studied for TGP with a proven expected runtime superlinear in the optimal program length $n$. We prove that the expected time for the ($1$+$1$) CGP employing common mutation operators to evolve an equivalent optimal of the same output is polylogarithmic in $n$, specifically at most $O(\log^9{n})$. This is remarkable, as it shows an exponential speedup by switching to CGP. Our results shed light on the benefit and the compactness of the graph-based representation for programs, and on how CGP can navigate its search and fitness spaces efficiently. This is a first proven performance guarantee for CGP on an established benchmark problem. Experiments complement our theoretical findings.

cs.LG · cs.AI · cs.NE

版本更新 · v2arXiv:2604.08541

Seeing but Not Thinking: Routing Distraction in Multimodal Mixture-of-Experts

Haolei Xu, Haiwen Hong, Hongxing Li, Rui Zhou, Yang Zhang, Longtao Huang, Hui Xue, Yongliang Shen 等 10 位作者

原文摘要

Multimodal Mixture-of-Experts (MoE) models have achieved remarkable performance on vision-language tasks. However, we identify a puzzling phenomenon termed Seeing but Not Thinking: models accurately perceive image content yet fail in subsequent reasoning, while correctly solving identical problems presented as pure text. Through systematic analysis, we first verify that cross-modal semantic sharing exists in MoE architectures, ruling out semantic alignment failure as the sole explanation. We then reveal that visual experts and domain experts exhibit layer-wise separation, with image inputs inducing significant routing divergence from text inputs in middle layers where domain experts concentrate. Based on these findings, we propose the Routing Distraction hypothesis: when processing visual inputs, the routing mechanism fails to adequately activate task-relevant reasoning experts. To validate this hypothesis, we design a routing-guided intervention method that enhances domain expert activation. Experiments on three multimodal MoE models across six benchmarks demonstrate consistent improvements, with gains of up to 3.17% on complex visual reasoning tasks. Our analysis further reveals that domain expert identification locates cognitive functions rather than sample-specific solutions, enabling effective transfer across tasks with different information structures.

cs.CV · cs.AI · cs.CL

版本更新 · v2arXiv:2607.26057

Pass the Baton: Trajectory-Relayed On-Policy Distillation

Haolei Xu, Xiaowen Xu, Haiwen Hong, Zixuan Ni, Hongxing Li, Yiwen Qiu, Weiming Lu, Yongliang Shen

原文摘要

On-policy distillation (OPD) grounds token-level supervision in the student's own trajectory, yet suffers from prefix failure: once the student commits to a wrong reasoning direction, all subsequent generation builds on this deviation, producing misdirected continuations that elicit unreliable supervision and waste compute. We identify a teacher-student continuation asymmetry on failed prefixes, where the teacher tends to redirect while the student continues along the original direction, and convert it into a label-free handoff trigger in Relay On-Policy Distillation (Relay-OPD). During training, Relay-OPD constructs relay trajectories by letting the teacher briefly take over at detected trigger points to produce a teacher leg, after which the student resumes and is optimized on the resulting trajectory. A limited relay budget concentrates intervention on critical early positions while limiting departure from the student policy. With a Qwen3-4B-Instruct-2507 teacher and Qwen3-0.6B/1.7B-Non-Thinking students on eight mathematical reasoning benchmarks, Relay-OPD achieves the best or second-best results on every benchmark, outperforming standard OPD by +5.73% and the strongest baseline FastOPD by +1.49% on average for 1.7B, with consistent gains at 0.6B. Training trajectory length is reduced by over 50%.

cs.CL · cs.AI

版本更新 · v3arXiv:2502.16336

Rectifying Conformity Scores for Better Conditional Coverage

Vincent Plassier, Alexander Fishkov, Victor Dheur, Mohsen Guizani, Souhaib Ben Taieb, Maxim Panov, Eric Moulines

原文摘要

We present a new method for generating confidence sets within the split conformal prediction framework. Our method performs a trainable transformation of any given conformity score to improve conditional coverage while ensuring exact marginal coverage. The transformation is based on an estimate of the conditional quantile of conformity scores. The resulting method is particularly beneficial for constructing adaptive confidence sets in multi-output problems where standard conformal quantile regression approaches have limited applicability. We develop a theoretical bound that captures the influence of the accuracy of the quantile estimate on the approximate conditional validity, unlike classical bounds for conformal prediction methods that only offer marginal coverage. We experimentally show that our method is highly adaptive to the local data structure and outperforms existing methods in terms of conditional coverage, improving the reliability of statistical inference in various applications.

stat.ML · cs.LG · math.ST

新论文 · v1arXiv:2611.20146

Evacuating from Eventually Emerging Targets on the Line

Konstantinos Georgiou, Caleb Jones, Liza Krupsky

原文摘要

We introduce and study Eventually Emerging Linear Search, an online evacuation problem on the line that, to the best of our knowledge, has not been studied before. In this problem, $n$ unit-speed agents with wireless communication start at the origin and search for a hidden target at location $x$, where $|x|\geq1$, that becomes detectable at an unknown time $T$. Visiting $x$ before time $T$ reveals nothing, and the search terminates only when all agents have reached the target. We study two competitive measures. In the emergence-aware model, the target location becomes known to the offline algorithm at time $T$, giving cost $T+|x|$. In the fully-informed model, the offline algorithm knows $x$ from the beginning, giving cost $\max\{|x|,T\}$. For one agent, the optimal competitive ratio is $9$ under both measures, as in classical cow-path search. For two agents, we give an algorithm with competitive ratio $3+\sqrt6$ under both measures, and prove it optimal within the class of origin-separated algorithms, in which each agent performs strictly expanding origin-return excursions on a fixed half-line. We also prove the unconditional lower bounds $4.35$ and $4.38$ for the emergence-aware and fully-informed measures, respectively. These bounds are computer-assisted; we derive finite linear programming relaxations for arbitrary algorithms and certify their infeasibility using exact rational Farkas certificates. For three agents, we prove analytically the unconditional lower bounds $3.1014$ and $3.2823$ for the emergence-aware and fully-informed measures, respectively, and introduce a geometric three-peak family. Different parameter choices give the corresponding upper bounds $4.010599$ and $4.340445$. These bounds lie below the corresponding two-agent lower bounds, so a third agent provably improves both measures.

cs.DM

新论文 · v1arXiv:2611.20145

Rank, Don't Regress: A 279M Text Ranker Beats 2B and 7B VLMs on Referential Urdu GUI Grounding

Hamza Raziq Khan, Aimal Khan

原文摘要

UI grounding asks a model to find the screen element that a user instruction refers to. Most current systems use a vision-language model (VLM) that reads a screenshot and outputs pixel coordinates. We study this task on Urdu mobile web pages, a setting with no published UI grounding results. Instead of regressing coordinates, we take the candidate elements from the page DOM and train a 279M-parameter cross-encoder (mDeBERTa-v3-base) to pick one. The click point is the centre of the picked element's box, so it is exact whenever the right element is picked. We built a frozen test set of 304 instructions from six Urdu news sites; 149 of them (49%) are referential and never quote the full target text. Test domains were banned from training data before collection started. The ranker scores 73.4% point-in-box accuracy, against 29.6% for UI-TARS-2B and 23.7% for Qwen2.5-VL-7B, both zero-shot. On the referential slice the ranker scores 65.8%, against 30.2% for a string-matching baseline that needs no training. In a controlled comparison at small scale, with training data drawn from the same pages, a coordinate-regression model of similar size scores 1.6%. Adding element geometry as input tokens gives the best variant, at 75.7%. The ranker's errors concentrate in two places: elements that share identical text, and ordinal instructions such as "the fourth link".

cs.CL

新论文 · v1arXiv:2611.20144

HEAMS: Holistic Evaluation of Agent Memory Systems

Chirag Goel, Jacob Lavoie, Amirhossein Naghshzan, Jordan Bannister, Marzia Nouri, Mehar Bhatia, Matt Harris, Siva Reddy 等 10 位作者

原文摘要

Memory systems allow AI agents to persist information across interactions, making them more personalized and more capable on long-horizon tasks. This persistence introduces new risks: stored information can be disclosed in inappropriate contexts, resist correction or deletion, be retained beyond what users desire, or be poisoned to influence later agent behaviour. Yet memory systems are typically evaluated on task performance and efficiency alone. We argue that a memory system should instead be understood as the custodian of its user's persistent information, with responsibilities that extend beyond helping an agent complete its tasks, and that it should be evaluated holistically against all such responsibilities. We present Holistic Evaluation of Agent Memory Systems (HEAMS), an open-source framework for evaluating agent memory systems in this way. First, we state the responsibilities of agent memory systems as explicit behavioural desiderata spanning performance, efficiency, security, privacy, and user sovereignty. Second, we assemble a benchmark suite that covers these desiderata and acknowledge the gaps that remain. Third, we develop MilaMem, a prototype memory system, and a governance layer that can wrap any memory system, enabling controlled study of memory system design choices. Fourth, we compare memory systems across the full HEAMS benchmark suite. We find that no memory system leads every performance benchmark, and strong performance scores may coexist with poor privacy, user-sovereignty, and security scores. Governance strongly influences results: recall policies can reduce inappropriate disclosure at a cost in helpfulness, while both recall and retention policies improve poisoning resistance. Performance and efficiency benchmarks alone therefore cannot reveal how well a memory system meets its responsibilities. Code: https://github.com/milatechtransfer/heams

cs.AI · cs.CL · cs.CR

版本更新 · v4arXiv:2504.15564

OpenClassGen: A Large-Scale Corpus of Real-World Python Classes for LLM Research

Musfiqur Rahman, SayedHassan Khatoonabadi, Emad Shihab

原文摘要

Existing class-level code generation datasets are either synthetic (ClassEval: 100 classes) or insufficient in scale for modern training needs (RealClassEval: 400 classes), hindering robust evaluation and empirical analysis. We present OpenClassGen, a large-scale corpus of 324,843 Python classes extracted from 2,970 engineered open-source projects. Each entry pairs a human-written class with its corresponding skeleton, which comprises class and method signatures with associated docstrings, and is enriched with 27 static code metrics covering complexity, coupling, cohesion, and inheritance properties. Unlike prior benchmarks that require repository-level context resolution, OpenClassGen provides self-contained class skeletons that serve as complete generation specifications. We demonstrate the corpus's utility by evaluating three LLMs (GPT-o4-mini, Claude-4-Sonnet, Qwen-3-Coder) on a curated, executable subset of 300 classes, enriched with test suites achieving 58% branch coverage. Results show strong semantic similarity (CodeBERTScore-F3: 0.89) but moderate functional correctness (pass rate: 0.33), with substantial variance across models. This variance, along with diverse class characteristics, confirms that OpenClassGen enables meaningful differentiation of LLM capabilities. The dataset supports diverse use cases, including fine-tuning, retrieval-augmented generation, difficulty modelling, and failure mode analysis. The complete dataset and curation scripts are publicly available at https://zenodo.org/records/18409150.

cs.SE · cs.AI · cs.LG