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

+0新增论文
+0其中 AI 相关
+0已有论文修订
+91新增 AI 导读

arXiv 最近完成扫描:2026-11-22 08:44 · 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 14:29 — 2026-11-22 14:29(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 08:44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 1,456 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-22 08:42(北京时间)
新论文 · v1arXiv:2611.26601

SoK:评估LLM在软件漏洞研究中能力的演进

SoK: Assessing the Evolution of LLM Capability in Software Vulnerability Research

Amine Lbath, Manan Suri, Aurelien Delaitre, Vadim Okun, Massih-Reza Amini, Dinesh Manocha

AI 速读 基于摘要

该论文提出一个基于自动驾驶分级思想的自主性评级框架,系统化梳理LLM在漏洞发现、验证与修补方面的证据,并分析其能力演进对公开漏洞记录的影响。

cs.CR · cs.AI · cs.SE

查看原文摘要

LLM-based agents are becoming increasingly capable at discovering, validating and patching vulnerabilities in real software, and vendors and researchers report that this is already changing how vulnerabilities are found and fixed. These claims, however, are scattered across papers, vendor reports and disclosures and are rarely comparable. We introduce a framework that rates these capabilities on an autonomy scale modeled on the levels of driving automation, and qualifies each claim by its correctness, its availability and the strength of its evidence. We use this framework to systematize the published evidence on LLM capability in vulnerability research. We find that agents discovered, validated, and patched vulnerabilities in real-world software without human intervention before the restricted frontier models of 2026, and that open-weight models subsequently attained comparable autonomy for discovery and validation. We then measure the consequences of evolving LLM security capabilities in a decade of public vulnerability records, advisories and exploitation data. In 2026 the number of disclosed vulnerabilities more than doubled, with an increasing share rated critical or high, and severe bugs that had survived a decade or more in widely used code were reported several times more often than before. Yet the time between the stages of a vulnerability's life, from report to fix and from disclosure to attack, has not shifted noticeably so far. We grade the mitigations proposed in the literature with an explicit rubric and draw guidance for each group of stakeholders from the best-supported ones.

新论文 · v1arXiv:2611.26582

端到端量子算法求解微分方程:在生物分子系统模拟中的应用

End-to-End Quantum Algorithm for Differential Equations: Application to Simulating a Biomolecular System

Mohsen Bagherimehrab, Daniyaal Farooqi, Gavin K. Brennen, Barry C. Sanders, Alán Aspuru-Guzik

AI 速读 基于摘要

作者提出并实现了一个针对线性化泊松-玻尔兹曼方程的端到端量子算法,从经典输入到经典输出,展示了在网格点数上相对于经典方法的指数级扩展优势。

quant-ph · cs.DS

查看原文摘要

Quantum algorithms for differential equations have the potential for substantial algorithmic speedups over classical methods. However, they rely on certain assumptions and abstract oracle specifications that leave their practical feasibility and computational advantages for realistic applications unclear. Here we formalize the concept of an end-to-end quantum algorithm for differential equations as one that accepts classical inputs and returns classical outputs, and provide implementations for algorithmic primitives in constructing such algorithms. We present a tailored end-to-end quantum algorithm for the linearized Poisson-Boltzmann (LPB) equation, a partial differential equation (PDE) with variable coefficients widely used in simulating a biomolecular system. We explicitly implement all aspects of the algorithm, from preparing an initial state to generating a quantum solution and estimating properties of the system. In particular, we develop a new discrete representation for variable-coefficient PDEs on a grid that enables efficient implementation of the operator oracle, and build on a recent technique based on optimal wavelet preconditioning to generate a quantum solution with a cost independent of the condition number of the discretized operator. We show the algorithm can be implemented by quantum circuits of size polylogarithmic in both the number of grid points $N_g$ and the inverse of the desired accuracy $\varepsilon$, with an overall gate cost polylogarithmic in $N_g$ and quasilinear in $1/\varepsilon$. We also show the classical preprocessing to construct the circuits has polylogarithmic cost in $N_g$ and $1/\varepsilon$. Our work provides a concrete end-to-end quantum algorithm for a PDE of practical value, demonstrates an exponential improvement in scaling with $N_g$ over classical methods, and provides a foundation for assessing its potential quantum advantage in practice.

新论文 · v1arXiv:2611.26557

MM-SystemOneBias:评估多模态Jev风格系统一决策模型中的偏差

MM-SystemOneBias: Evaluating Bias in Multimodal Jev-Style System One Decision Models

Jinzhe Li, Qi Wang, Dandan Guo, Li Liu, Yuan Wu, Yi Chang

AI 速读 基于摘要

作者提出MM-SystemOneBias基准,通过六种配对干预在九个公开数据集上比较13个多模态决策模型与其骨干模型的行为差异。

cs.AI

查看原文摘要

When multimodal decision models determine actions, moderation outcomes, or quality judgments, their reliability depends on preserving semantic decisions across presentation changes, ignoring irrelevant additions, and distinguishing information consistent with task evidence from information that conflicts with it. Existing evaluations study decision-interface biases and multimodal model behavior in separate settings, leaving no common protocol to compare multimodal decision models with their backbones on practical tasks. We introduce \textbf{MM-SystemOneBias}, which evaluates 13 models on 5,323 original instances from nine public datasets and 11 task scenarios using six paired interventions: option order, option label, external judgment, image--text description, irrelevant information, and rating scale. Across seven matched decision-model--backbone pairs, decision models are more stable under option-label, irrelevant-information, and rating-scale changes in all seven pairs; six pairs also show lower option-order flip rates. They incur smaller accuracy losses under incorrect external judgments in all seven pairs and conflicting descriptions in six. In six pairs, they gain less accuracy from correct external judgments and show less favorable accuracy changes under consistent descriptions. Among pairs with unchanged semantic decisions, mean absolute changes in answer probabilities range from 4.56 to 11.55 percentage points, revealing input effects that accuracy alone misses.

新论文 · v1arXiv:2611.26553

智能体安全事件的复现:审计与调查(ROGUE-AI)

Replication of Agentic Security Events: Audit and Investigation (ROGUE-AI)

Mengxiao Wang, Mahbub Alam, Kalyan Nakka, Nitesh Saxena

AI 速读 基于摘要

提出多智能体框架 ROGUE-AI,将媒体报道的 LLM 智能体安全事故离线复现,并设置仅缺少被归因原因的控制组,以检验事故成因与防御效果。

cs.CR · cs.AI · cs.MA

查看原文摘要

Agents based on large language models (LLMs) can now delete files and send payments on behalf of their users. News media reports routinely describe failures of these agents (e.g., Replit's coding agent deleting a production database), but a report often only says what went wrong and rarely (in)validates the incident or finds its root cause. Incident catalogs collect these reports, and benchmarks record whether agents fail, but neither replicates an incident nor tests the cause its report might identify. In this paper, we present ROGUE-AI (Replication of Agentic Security Events: Audit and Investigation), a multi-agent framework that closes the above gap and rebuilds each reported incident offline together with a matched control that lacks only the blamed cause, and runs agents on both. Our results suggest that ROGUE-AI is able to reproduce the reported harm in 89% of its 108 attempted replications. Agents built on five current GPT, Claude, Gemini and DeepSeek models made the reported mistake in 18% to 33% of runs, against at most 1% without the blamed cause. Of the 72 incidents with real-world harm, 62 had no underlying attacker, and in these, agents did what their task asked for but missed an ordinary hazard, such as a database shared with the live site. Replications without an attacker caused more harm than replications of prompt injection for all five models. Defenses against prompt injection look for injected instructions, but these replications contain none. Further, a warning after each tool output that names the hazard and its consequence lowered harm for all five models, whereas a note of the same length that does not mention the hazard left most of the harm in place. Built from real incidents, the ROGUE-AI replications form a reusable audit framework that vendors can rerun before each new release.

版本更新 · v2arXiv:2602.06363

面向不确定模态的鲁棒行人检测

Robust Pedestrian Detection with Uncertain Modality

Qian Bie, Xiao Wang, Bin Yang, Zhixi Yu, Jun Chen, Xin Xu

AI 速读 基于摘要

作者构建RGB-NIR-TIR三元组数据集并提出AUNet,以在模态组合不确定时实现鲁棒行人检测。

cs.CV

查看原文摘要

Existing cross-modal pedestrian detection (CMPD) employs complementary information from RGB and thermal-infrared (TIR) modalities to detect pedestrians in 24h-surveillance systems.RGB captures rich pedestrian details under daylight, while TIR excels at night. However, TIR focuses primarily on the person's silhouette, neglecting critical texture details essential for detection. While the near-infrared (NIR) captures texture under low-light conditions, which effectively alleviates performance issues of RGB and detail loss in TIR, thereby reducing missed detections. To this end, we construct a new Triplet RGB-NIR-TIR (TRNT) dataset, comprising 8,281 pixel-aligned image triplets, establishing a comprehensive foundation for algorithmic research. However, due to the variable nature of real-world scenarios, imaging devices may not always capture all three modalities simultaneously. This results in input data with unpredictable combinations of modal types, which challenge existing CMPD methods that fail to extract robust pedestrian information under arbitrary input combinations, leading to significant performance degradation. To address these challenges, we propose the Adaptive Uncertainty-aware Network (AUNet) for accurately discriminating modal availability and fully utilizing the available information under uncertain inputs. Specifically, we introduce Unified Modality Validation Refinement (UMVR), which includes an uncertainty-aware router to validate modal availability and a semantic refinement to ensure the reliability of information within the modality. Furthermore, we design a Modality-Aware Interaction (MAI) module to adaptively activate or deactivate its internal interaction mechanisms per UMVR output, enabling effective complementary information fusion from available modalities.

新论文 · v1arXiv:2611.26444

Mind2Act World:面向机器人操作中具身心智、物理动作与心智-动作组合的诊断基准

Mind2Act World: A Diagnostic Benchmark for Embodied Mind, Physical Action, and Mind-Action Composition in Robotic Manipulation

Yuzhi Huang, Jie Wu, Fanding Huang, Junchao Lin, Weichen Zhang, Ziyi Xiong, Yang Li, Zhengwei Xie 等 43 位作者

AI 速读 基于摘要

作者提出一个包含30个仿真操作任务、180种配置的诊断基准,用Mind、Act、Mind2Act三个视图区分系统是缺少某种能力还是无法组合已有能力。

cs.RO · cs.AI · cs.CV

查看原文摘要

Embodied intelligence aims to let robots perform complex physical tasks with human-like adaptability and reliability, which requires three complementary capabilities: history-based cognitive reasoning, precise and timely physical execution, and the composition of reasoning and execution through environmental feedback. Yet existing benchmarks either evaluate reasoning and execution within the same tasks or assess them separately without their composition, so they reveal little about whether a failure comes from reasoning, execution, or their composition. We introduce Mind2Act World, a diagnostic benchmark of 30 simulated manipulation tasks from five real-world application domains. Each task has three difficulty levels and two scene settings (Clean and Random), yielding 180 configurations. The tasks form three views: Mind tests decisions based on past observations, Act tests precise and timely execution, and Mind2Act combines both in closed-loop tasks where action outcomes guide later decisions. Comparing the three views shows whether a system lacks a capability or fails to compose capabilities it already has. A diagnosis harness further attributes the failed episodes of three representative systems to Mind, Act, or their composition and summarizes each system in a profile card. Our evaluation of multimodal agents, vision-language-action models (VLAs), and world-action models (WAMs) shows that competence in a single view does not carry over to Mind2Act tasks. Even the strongest agent succeeds in 52.7% of Mind and 39.0% of Act episodes but in only 12.0% of Mind2Act episodes. The strongest agent retains task state but struggles with sustained contact, whereas end-to-end policies execute practiced motions but retain little task state. These findings point to composing reasoning and execution through environmental feedback as the main open problem.

新论文 · v1arXiv:2611.26293

LocoMuJoCo: A Comprehensive Imitation Learning Benchmark for Whole-Body Control

Firas Al-Hafez, Davide Tateo, Oleg Kaidanov, Aditya Bhatt, Artur Zelik, Zhengyi Luo, Jan Peters

原文摘要

LocoMuJoCo is an open-source benchmark for imitation learning (IL) and reinforcement learning (RL) on whole-body control. The library couples MuJoCo (CPU) with MJX and the MjWarp backend (GPU) under a single API, exposes 9 humanoid, 3 (musculo-)skeletal, and 4 quadruped embodiments, and ships a curated native motion-capture dataset for every humanoid, to which it can also retarget AMASS and LAFAN1. A modular plug-in architecture makes observations, rewards, control modes, terrains, and domain randomization fully interchangeable, while compact JAX implementations of many RL and IL algorithms provide JIT-compiled end-to-end training, and the companion loco-mujoco-real repository provides sim-to-real deployment. We describe the design, the SMPL-to-robot and robot-to-robot retargeting pipeline, and a usage example. LocoMuJoCo is released under MIT at https://github.com/robfiras/loco-mujoco.

cs.RO

版本更新 · v2arXiv:2611.22476

CheckCo: Checkpoint-Coordinated Load Balancing for Skewed Event Streams at Scale

Muhammad Anis Uddin Nasir, Reza Sadraei

原文摘要

Event streams are rarely uniform: a few dominant keys can generate orders of magnitude more events than the rest, overloading the workers that receive them and capping throughput. We present CheckCo, a checkpoint-coordinated load-balancing algorithm built on a dynamic oracle that estimates per-key frequencies from the live stream, with no prior statistics. Workers count events per key locally; at each checkpoint barrier, the counts are merged into a partitioning function that divides each key into a number of splits roughly proportional to its estimated frequency and packs the splits onto workers, and this function routes the next checkpoint window. We derive approximation bounds on the resulting load imbalance from classical scheduling results and bound its total key fan-out. On a production pipeline processing over 100 billion events per day, the most loaded worker receives only 9.8% more events than the average over a day (imbalance ratio 1.10), against 8.03 for key-based routing in simulation; in simulation, CheckCo also needs less than half the fan-out of earlier partitioners that reach comparable balance. Shifts in the key distribution are reflected in routing from the next checkpoint on, and the load stays balanced through job restarts.

cs.DC

版本更新 · v2arXiv:2611.21472

Event-Triggered Visual Perception for Real-Time Harvest Outcome Estimation in Mechanical Blueberry Harvesting

Nicolas Miroslav Husár, Jelle Driessen, Petra Bosilj, Başak Sakçak

原文摘要

Mechanical blueberry harvesting requires a balance between removing ripe fruit and avoiding premature removal of unripe berries that could remain on the bush for a later harvest. However, operators generally lack real-time information about the maturity composition of the fruit being collected. This study develops an event-triggered machine-vision system that measures the maturity of harvested blueberries directly on the harvester. When a filled crate is moved by the machine, an RGB image is automatically captured and analysed using a three-class detector that identifies ripe, semi-ripe, and unripe berries. The system operates locally on the existing CPU-only industrial computer and therefore requires neither continuous image processing nor dedicated GPU hardware. During field testing, 79 crate images containing 48,634 visible blueberries in manually verified reference annotations were evaluated. The vision system estimated an overall ripe-fruit fraction of 90.22%, compared with 90.97% determined manually. Per-image ripe-fruit estimates showed a mean absolute error of 0.935 percentage points and a Pearson correlation of r=0.967. Maturity estimates remained similar between the two sides of the harvester for which the light conditions varied despite poorer absolute counting performance under direct sunlight. All field events were processed within the available 14 s per-side crate interval. The proposed approach provides a practical means of quantifying one component of harvest selectivity while the machine is operating. Such information could support operator decision-making and, when combined with measurements of fruit recovery and losses, contribute to future feedback-based control of mechanical harvesting.

cs.CV

新论文 · v1arXiv:2611.26248

TwinLab-Bench: Evaluating AI Scientists in Agentic Twins of Real Labs

Yang Wan, Linchao Zhu

原文摘要

AI agents are increasingly expected to act as AI scientists, which requires three increasingly open abilities: carrying out a given computational workflow correctly, choosing the next measurements from a given design space, and designing their own experiments. Existing benchmarks mainly evaluate the first two abilities, in a terminal or in a closed loop that selects from predefined candidates. Evaluating the third ability requires reliably executing an experiment that an agent designs, with readings grounded in a real laboratory. We introduce TwinLab-Bench, which evaluates AI scientists on self-designed experiments in 52 tasks from laboratories in the physical sciences, the life sciences, and engineering. The benchmark supports more than 100 experimental action types, and the agent's report is scored by comparison with a hidden answer taken from real laboratory records. Each task has its own TwinLab agent, an agent that serves as the environment and executes the experiments that the tested agent designs, reading the raw materials of the laboratory and calling calculators that are built from its records. Compared with a language model that sees only the task description, the TwinLab agent almost eliminates readings that deviate from the real records and inconsistent answers to a repeated request, and closed-book probes control for answers recalled from memory. We find that these tasks remain hard for frontier models: the strongest, Claude Fable 5.2, averages a score of only 45.0, and we conduct an empirical study of where it fails.

cs.AI · cs.CL · cs.LG

版本更新 · v3arXiv:2608.29519

FuncRoom-Agent: Sequential Feed-Forward 3D Functional Indoor Scene Generation

Hao Feng, Zhi Zuo, MingJian Liang, Jingyu Hu, Xiaowei Hu, Liupengfei Wu, Dian Zhang, Guoxin Fang 等 9 位作者

原文摘要

We introduce Function-Room Generation, a new indoor 3D scene generation setting that creates rooms supporting explicit functional goals rather than merely visually plausible layouts. Existing agentic and executable methods improve controllability, but often depend on costly test-time generate--evaluate--revise loops, making functional room generation slow and computationally expensive. We address this challenge with three technical contributions. First, we design a recursive domain-specific language to effectively organize the hierarchical object compositions required by functional rooms, from room structure and major furniture to dense support-surface and nested small objects. It represents rooms as staged executable programs with explicit geometric and functional relations. Second, we propose a sequential feed-forward scene construction framework that distills recursive construction traces into a scene construction expert. At inference time, the expert writes executable DSL code stage by stage, and a deterministic executor directly instantiates each stage without teacher agents, online critics, or iterative repair. Third, we introduce ScenePRM, an execution-grounded process reward framework that improves the expert through reinforcement learning with functional, geometric, relational, and future-constructability feedback. We further establish a function-oriented benchmark and show state-of-the-art performance on both general indoor scene generation and function-room generation, achieving stronger functional completeness, relation correctness, geometric executability, and generation efficiency.

cs.CV

版本更新 · v2arXiv:2601.04423

Learning Multinomial Logits in $O(n \log n)$ time

Flavio Chierichetti, Mirko Giacchini, Ravi Kumar, Silvio Lattanzi, Alessandro Panconesi, Erasmo Tani, Andrew Tomkins

原文摘要

A Multinomial Logit (MNL) model is composed of a finite universe of items $[n]=\{1,\dots,n\}$, each assigned a positive weight. A query specifies an admissible subset---called a slate---and the model chooses one item from that slate with probability proportional to its weight. This query model is also known as the Plackett--Luce model or conditional sampling oracle in the literature. Although MNLs have been studied extensively, a basic computational question remains open: given query access to slates, how efficiently can we learn weights so that, for every slate, the induced choice distribution is within total variation distance $\varepsilon$ of the ground truth? This question is central to MNL learning and has direct implications for modern recommender system interfaces. We provide both adaptive and non-adaptive algorithms for this task. Our first adaptive algorithm makes $O(\frac{n\log n}{\varepsilon^2})$ queries and learns every choice distribution in total variation distance $\varepsilon$. We also provide a second adaptive algorithm that with $O(\frac{n \log n}{\varepsilon^3})$ queries provides a stronger guarantee: each choice distribution is approximated to a multiplicative $1\pm\varepsilon$ error (except for subsets of mass smaller than $\varepsilon$). We complement these algorithms with lower bounds of $Ω(\frac{n\log n}{\varepsilon^2})$ for total variation distance and $Ω(\frac{n \log n}{\varepsilon^3})$ for the multiplicative error guarantee, showing the optimality of our adaptive algorithms. Our non-adaptive algorithm makes $O\left(\frac{n^{2}}{\varepsilon^{3}}\log n \log\frac{n}{\varepsilon}\right)$ queries to obtain guarantees under the multiplicative error notion. We complement this result with an $Ω(\frac{n^2 \log n}{\varepsilon^2})$ lower bound for any non-adaptive algorithm, showing that our algorithm is almost tight in $n$ (within a $\log n$ factor).

cs.DS · cs.LG · math.ST · stat.ML

版本更新 · v3arXiv:2611.06470

How many sorted projections determine any n points in k dimensions?

Christopher Gorham Lester, Raymond Hau Sørensen

原文摘要

Let $d_1,\dots,d_m$ be directions in $\mathbb{R}^k$, where $k\ge 2$. A multiset $X$ of $n$ points of $\mathbb{R}^k$ projects to the $m$ multisets of numbers $\{\{x\cdot d_j : x\in X\}\}$, or equivalently to the $m$ sorted lists of these numbers. How many directions are needed for these projections to determine every multiset of $n$ points? Two simple bounds are known in every dimension: more than $(k-1)\log_2 n$ directions are always needed, and $(k-1)n+1$ directions in general position always suffice. In the plane much more is known: Matousek, Privetivy and Skovron showed that almost every choice of $O(\log n \log\log n)$ directions suffices. Whether anything similar holds in higher dimensions was left open by Dym, Wellershoff, Tsoukanis, Levy and Balan. We observe that two lemmas of Hair and Sahai, proved for list recovery of random linear codes, answer that question, and by tightening their argument we show that in every dimension almost every choice of $(4k-2)\ln(2n)\ln(k\log_2 2n)+k+1$ directions suffices. For $k\le 10$ this beats the linear bound once $n$ is larger than about 100: for example, 144 directions suffice for multisets of 1000 points in the plane, and an exact computation lowers this to 133. Letting some of the random blocks in their argument fail removes a factor $\log k$: almost every choice of $O(k\log n\log\log n)$ directions suffices, uniformly in $k$ and $n$. We also show that $2k-1$ directions never determine three points, so that the logarithmic lower bound is not tight; that every $m$ directions in the plane fail on some pair of multisets of size $O(1.65^m)$, improving the bound $O(1.80^m)$ of Matousek, Privetivy and Skovron; and that for each $k$, $m$ and $n$, either almost every choice of $m$ directions works or almost none does.

math.CO · cs.DM · cs.IT · cs.LG

版本更新 · v2arXiv:2611.23445

Why Sub-4-Bit Vision CNNs Collapse and Where Mixed-Precision Recovery Stops

Michael Schroeder

原文摘要

Deep vision models operating in resource-constrained environments depend heavily on low-bit quantization, yet compressing weights below 4 bits frequently induces catastrophic accuracy collapse. While mixed-precision recovery strategies aim to rescue accuracy by selectively retaining high-precision parameters, the fundamental mechanics of where and why these models collapse remain poorly understood. In this study, we systematically analyze the anatomy of quantization collapse across standard Vision CNNs (including ResNet and EfficientNet backbones) evaluated on the full 50,000-image ImageNet-1k validation set. First, we map the exact progression of degradation, demonstrating that collapse is highly non-linear and concentrated in late-stage pointwise projections and depthwise layers. Second, we introduce a structured, multi-step recovery waterfall that evaluates selective integer precision (INT4/INT8) and selective floating-point scales. We show that keeping less than 5% of critical layer activations in 8-bit precision recovers up to 92% of the accuracy lost during aggressive sub-4-bit compression. Finally, we bridge the hardware reality gap by evaluating deployment performance on standard ARM and x86 CPU architectures. We show that while low-bit quantization reduces memory storage requirements, standard runtime software bit-unpacking introduces a 1.48x to 21.5x execution slowdown depending on batch size, highlighting a critical trade-off between model footprint and actual edge throughput. Our findings provide a practical blueprint for navigating the limits of sub-4-bit deployment without sacrificing hardware performance.

cs.AI · cs.CV · cs.LG

版本更新 · v2arXiv:2611.15073

When Does Causality Help in Multivariate Time-Series Anomaly Detection?

Long Zhang, Wei Huang, Aishui Rao, Jian Hu, Binxiang Wu, Jie Li, Chen Gong, Linxi Feng

原文摘要

Causality-aware scoring is common in multivariate time-series anomaly detection, yet the literature rarely states when it helps. Holding one causality-guided autoencoder (DeCauAE) fixed, we vary graph quality, anomaly type and protocol against ground truth. Graph quality fixes the payoff: on a twenty-variable DAG the gain over a graph-free autoencoder decays as the graph is corrupted. Anomaly type fixes the target: the structural factor lifts window F1 from 0.800 to 0.976 where anomalies break conditional structure, and its advantage vanishes below a break severity s 0.3. Protocol fixes the ranking: point-adjusted F1 lets a random score match the full model, while range-aware VUS separates the autoencoder family from the classical baselines. On real benchmarks the causal module pays only where density makes a system eligible and cross-subsample reproducibility selects entities inside it; enabling it everywhere costs 0.012 VUS-ROC. The structural factor also transfers to classical detectors .

cs.LG

新论文 · v1arXiv:2611.26055

When Do Rankings Beat Signs? Gradient Direction Recovery under Even Nuisance

Zhuangzhuang Pan, Yan Xia, Chee Seng Chan

原文摘要

Can ranking a batch of antipodal query pairs recover a gradient direction with fewer evaluations than using within-pair signs alone? We study fixed-center recovery in $d$ dimensions for a linear signal plus an unknown differentiable even nuisance, counting function evaluations and feedback rounds separately. Shared pair offsets create an information hierarchy: Mirror retains within-pair signs, Reverse couples reversed cross-pair orders to cancel offsets, and Full enforces all ranking inequalities. Relative shell variation is the nuisance range over the query sphere normalized by the linear-signal scale. When it is at most $c\varepsilon/\sqrt{d}$, Reverse localizes every consistent direction to angular accuracy $\varepsilon$ using $\widetilde{O}(d+\sqrt{d/\varepsilon})$ evaluations in one round at fixed confidence, without a variation bound supplied to the decoder. This matches a lower bound up to logarithmic factors, versus the $Θ(d/\varepsilon)$ one-round sign-only minimax cost on the same class. Over well-conditioned positive-definite quadratics with no common relative-curvature bound, the $R$-round minimax cost is instead $Θ(dR(\varepsilon^{-1/R}-1))$, even with cumulative rankings. We extend validity beyond exact evenness under supplied error bounds. Matched-endpoint experiments certify information beyond signs and short cycles. On two ridge tasks under joint evaluation and round caps, short-cycle recovery attains lower terminal mean gaps than the tested batch-capable controls. An additional optimization benefit from fitting the complete ranking remains unestablished.

cs.LG

版本更新 · v3arXiv:2603.17927

RoboForge: Cross-space Motion Refinement for Physically Executable Humanoid Whole-body Tracking

Xichen Yuan, Zhe Li, Zhenzhe Zhang, Yangyang Wei, Xinying Guo, Shanghang Zhang, Jianfei Yang

原文摘要

Whole-body motion tracking has rapidly expanded the skill repertoire of humanoid robots, yet nearly all of this progress has come from stronger tracking policies, while the motion references they imitate are treated as fixed supervision. We argue that reference quality is a first-order factor in tracking performance. Human motion data carry capture and estimation artifacts, and retargeting adds physical inconsistencies due to the human-robot embodiment gap; discarding such motions sacrifices valuable skill content, whereas retaining them uncorrected undermines policy learning. We present RoboForge, a data-centric framework that resolves this dilemma by refining references instead of filtering them, repairing each error in the space where it arises. In human motion space, a quality-guided diffusion model localizes corrupted segments and inpaints only those segments, leaving the rest of the motion intact. In robot motion space, kinematic repair corrects foot-contact and self-collision defects, and annealed sampling-based optimization under physics rollouts turns kinematically plausible references into dynamically executable ones. RoboForge requires no change to the tracker: it improves three retargeting methods and four tracking policies, raising the success rate of BeyondMimic from 89.51% to 96.20% while reducing the mean per-joint position error (MPJPE-L) by 31.6%. Refined references also accelerate policy learning, restore tracking performance from heavily corrupted inputs, and enable more than 30 high-dynamic motions, including flips and the Thomas flare, on a real Unitree G1.

cs.RO

新论文 · v1arXiv:2611.26036

Presenting Your Figures: Data-informed Animated Transitions Across Scientific Figure Panels

Lu Ying, Can Liu, Yuxiao Yang, Jianing Yin, Xiaolin Wen, Yong Wang

原文摘要

Multi-panel figures are commonly used in research papers to show scientific data, and researchers often reuse them when presenting their work. Presenting these figures often involves explaining how data across chart panels are related. Animated transitions can progressively reveal these relationships, but two static panels specify only the endpoints, leaving what intermediate keyframes should show and how the transition should progress unclear. To support transition design across diverse scientific figures, we developed Figure Space and Transition Space through iterative analysis of 48 published multi-panel figures across eight scientific subject areas. Figure Space represents panel composition, data transformations, and visual encodings while retaining the links needed to trace panel results back to records. Transition Space uses this representation and source data to construct and sequence keyframes that make corresponding data visible and align them while preserving the meaning of each panel's position encodings. Building on these spaces, we developed FigFlux, a human-in-the-loop authoring system that guides AI agents in generating animated transitions from a multi-panel figure and its associated materials. Researchers can revise keyframes in natural language before animation generation. We evaluate FigFlux through cross-domain case studies, a study with 16 scientific authors, and expert assessment. Authors reported lower overall workload than with their familiar tools, while experts gave FigFlux transitions higher ratings than transitions created with participants' familiar tools or generated by AI alone. The example gallery have been published online: https://figflux.github.io.

cs.HC

新论文 · v1arXiv:2611.25988

Sequential Contract Design for Budgeted Crowdsourcing

Michal Feldman, Vasilis Gkatzelis, Shaul Rosner, Inbal Talgam-Cohen

原文摘要

A principal wishes to crowdsource a collection of tasks under a fixed compensation budget. The quality of a task's outcome is stochastic and depends on the worker's effort level. Since this effort is unobservable, the principal offers contracts to workers specifying their compensation for each possible outcome. The principal offers these contracts sequentially, using previously observed outcomes and payments to decide whom to approach next and what contract to offer. The objective is to maximize the expected value of the work produced while ensuring that total payments never exceed the budget, regardless of the realized outcomes. Contract design must therefore account both for a worker's incentives and for the budget available to incentivize others. For valuations that are additive across tasks, we characterize how the approximability of the optimal adaptive policy depends on the numbers of workers, effort levels, and outcomes. Polynomial-time constant-factor approximation is possible whenever the number of workers, effort levels per worker, or outcomes per task is constant, but impossible in general when all three are unrestricted, unless $P=NP$. We obtain tight guarantees for every fixed number of workers, establish computational hardness even in restricted settings, and obtain exact algorithms and approximation schemes in more structured settings. We also investigate simple contract structures, demonstrating both the benefits of adapting contracts to previous outcomes and the limitations of local contract rules, while identifying conditions under which a local policy achieves a constant-factor guarantee. Finally, we extend our analysis beyond additive valuations, obtaining constant-factor approximations for monotone submodular valuations with binary outcomes, including guarantees against a stronger full-knowledge benchmark under additional assumptions.

cs.GT

版本更新 · v2arXiv:2611.03168

Rollout then Recycle: 3,718 Labelled Real-Robot Episodes from Benchmarking 7 Policies on 12 Tasks

Praveen Selvaraj, Lorenzo Uttini, Ville Kuosmanen

原文摘要

We release all 3,718 episodes from the ArmnetBench v0.1 evaluation (arXiv:2607.24481), which contains 600 teleoperated demonstrations and 3,118 rollouts of seven imitation-learning and vision-language-action policies on 12 tabletop tasks, recorded on low-cost SO-101 arms; 73% of the rollouts are failures. Every policy for a task was trained or fine-tuned on the same 50 demonstrations and rolled out on the same cell, so the corpus contains many policies' failures on shared tasks. An operator labelled every rollout successful, suboptimal or failure. Each episode has three camera views, joint states and actions and is available on the Hugging Face Hub in LeRobot v3.0 and a reward-model format. This note documents how the episodes were collected and the limitations that could affect their use.

cs.RO

版本更新 · v5arXiv:2212.06123

Security of Deep Reinforcement Learning for Autonomous Driving: A Survey

Ambra Demontis, Srishti Gupta, Maura Pintor, Luca Demetrio, Kathrin Grosse, Hsiao-Ying Lin, Chengfang Fang, Battista Biggio 等 9 位作者

原文摘要

Reinforcement learning (RL) enables agents to learn optimal behaviors through interaction with their environment and has been increasingly deployed in safety-critical applications, including autonomous driving. Despite its promise, RL is susceptible to attacks designed either to compromise policy learning or to induce erroneous decisions by trained agents. Although the literature on RL security has grown rapidly and several surveys exist, existing categorizations often fall short in guiding the selection of appropriate defenses for specific systems. In this work, we present a comprehensive survey of 86 recent studies on RL security, addressing these limitations by systematically categorizing attacks and defenses according to defined threat models and single- versus multi-agent settings. Furthermore, we examine the relevance and applicability of state-of-the-art attacks and defense mechanisms within the context of autonomous driving, providing insights to inform the design of robust RL systems.

cs.LG · cs.RO

版本更新 · v2arXiv:2609.07369

DF26: We Cannot Tell Fake From Real Anymore

Severyn Shykula, Andrii Yermakov, Ivan Samarskyi, Dmytro Mishkin, Jan Cech, Anastasiia Mishchuk

原文摘要

We introduce DF26, a novel benchmark for detecting AI-generated videos containing fully synthetic clips produced by recent text-to-video and image-to-video models. The videos capture single-person public-speaking scenarios, spanning direct-to-camera recordings, official statements, and studio interviews - 271 real and 2,420 synthetic videos generated by seven modern video models. The study on DF26 shows that human accuracy in detecting AI-generated videos is 52.6% while recognizing 76.0% of real ones, and that state-of-the-art deepfake detectors, using the current evaluation setup, perform close to random chance. Our results highlight the limitations of current evaluation protocols and motivate the need for benchmarks that explicitly measure robustness to modern generative model distribution shifts. DF26 can be accessed at https://huggingface.co/datasets/DF26/DF26

cs.CV

版本更新 · v2arXiv:2607.04172

Lower Bound of Networked Control with Multiple Sensors and One Controller And The Application to the Side Information Case

Sijie Li, Takashi Tanaka, Hyeji Kim

原文摘要

This paper studies the tradeoff between the communication rate and the control performance for the Linear Quadratic Gaussian(LQG) plant with multiple sensors/encoders and one decoder/controller. Assuming the use of prefix codes for each encoder, we show a causally conditioned directed information lower bound and the sufficiency of a class of linear policies for optimizing this weighted sum lower bound with the full observation condition, reducing it to a finite-dimensional optimization problem. We then apply our analysis to the point-to-point case with side information(SI) at the decoder only, which can be viewed as an extreme case for the multi-encoder setting. We recover the structure result for the sequential rate-distortion function(SRDF) for the point-to-point case with SI. We further present semidefinite programming (SDP) formulations for a broader class of system parameters that recover the past forms as special cases, and show an equivalent single-letter SDP form for the infinite-horizon time-invariant system under certain conditions.

cs.IT · eess.SY

版本更新 · v2arXiv:2611.11502

Does Merge Form a Hopf Algebra? Coproducts, Traces, and the Prospects of Algebraic Minimalism

Daisuke Bekki

原文摘要

Marcolli, Chomsky, and Berwick (2025; henceforth MCB) formulate the Merge operation of Minimalism on a Hopf algebra of Connes-Kreimer type, and claim that language and physics obey the same algebraic structure. This paper examines the premise of that claim: whether Merge on workspaces can be described as a Hopf algebra. We first survey five coproducts for coassociativity and a connected grading: the Connes-Kreimer coproduct $Δ_ρ$, MCB's two coproducts $Δ_d$ and $Δ_c$, the contraction coproduct of Marcolli and Skigin (2025), and the coproduct of Xiao (2026). $Δ_d$ is not coassociative. For $Δ_c$, MCB gives no rule for the labels of trace leaves in trees that already contain trace leaves; under the literal extension it is not coassociative, but once a rule and a grading are supplied it yields a graded connected Hopf algebra. The coproduct of Xiao is coassociative, and so is that of Marcolli and Skigin once its definition is extended to trees with trace leaves. These properties do not tell the coassociative coproducts apart; what distinguishes them is the representation that movement leaves behind. We then review how generative grammar has treated moved elements, and confirm that semantic interpretation requires two things of the representation that results from movement. Retention requires that the structure the lower position had before movement be recoverable; Identifiability requires that, in interpreting the lower position, it be determinable which upper element, if any, corresponds to it. None of the five coproducts meets both conditions: Retention is met only by $Δ_c$, and there only in the form of a label, and Identifiability by none. In light of the results on coassociativity, we also reassess the Birkhoff factorization that MCB employs at the syntax-semantics interface. Finally, we indicate at least three directions for algebraic Minimalism.

cs.CL · math.QA

新论文 · v1arXiv:2611.25909

The Linguistic Structure of Dependency-Conditioned Embeddings

Bach Phan-Tat, Tung Kieu, Kris Heylen, Dirk Geeraerts, Stefano De Pascale, Dirk Speelman

原文摘要

Dependency-based distributional models capture syntactically informed lexical similarities, but the geometric organisation of dependency-conditioned lexical representations remains underexplored. We investigate the global and local geometry of lemma-dependency relation embeddings using four representations trained on WikiText-103: two count-based PPMI configurations and two Skip-Gram with Negative Sampling (SGNS) configurations. At the global level, we analyse dependency-relation centroids, control for lexical identity through within-lemma centring, and examine their alignment with part-of-speech (POS) distributions. At the local level, we evaluate cross-relation geometric consistency, cross-method agreement, lexical similarity, and polysemous usages. Results show that global geometry is structured strongly by context POS across all representations. Locally, broader lexical similarity structures are partially preserved across dependency relations, while immediate neighbourhoods vary substantially, highlighting different usages. PPMI-based representations exhibit greater cross-relation consistency than SGNS-based representations, with cross-method agreement varying by relation. These findings reveal systematic linguistic organisation at multiple geometric scales and its sensitivity to representation choices.

cs.CL

版本更新 · v2arXiv:2610.32678

Probing the limits of materials design via a comprehensive crystal database of 14.6 million structures with 140 properties

Zhilong Song, Zijun Pang, Lixue Cheng

原文摘要

Forward screening and inverse design drive materials discovery, yet their limits remain unquantified because records of known crystal structures and properties are fragmented across databases. Consequently, forward screening leaves known crystals underexplored for new functions, whereas generative and autonomous discovery has repeatedly reported known phases or close variants as new materials. Here we match structures across 55 databases to assemble MateriaViva, a unified database of 14.6 million distinct crystals and over one billion property values. To provide the missing properties of every structure, a task we term property completion, we develop an equivariant Crystal Graph Transformer Network with state-of-the-art accuracy on nine benchmark tasks and apply 140 property-prediction models across the database. These predictions reveal the underestimated potential of forward screening, identifying tens of thousands of high-performance candidates among previously synthesized or computationally studied materials that lack recorded target-property values. Conversely, comparison with MateriaViva shows that the novelty of inverse design has been overestimated, as 55-100% of inverse-designed structures are not new. Under equal density functional theory verification budgets for bulk-modulus design, inverse design yields a higher fraction of qualifying candidates, whereas forward screening reaches the higher bulk modulus. MateriaViva thus provides a comprehensive reference for probing the limits of both paradigms, revealing unrecognized performance in known crystals and directing inverse design beyond reported structures.

cond-mat.mtrl-sci · cs.LG

新论文 · v1arXiv:2611.25756

Self-Supervised Operator Learning for Channel Estimation in Fluid Antenna Systems

Erqiang Tang, Zixin Wang, Wei Guo, Shenghui Song, Jun Zhang, Khaled B. Letaief

原文摘要

Fluid antenna systems (FAS) have emerged as a promising technology for enhancing wireless communications through adaptive antenna repositioning. Recently, Fourier neural operators (FNOs) have shown strong potential for FAS channel estimation owing to their ability to learn resolution-invariant mappings and generalize across different port densities. However, existing operator learning approaches require densely sampled channel realizations as training labels, whereas a practical receiver only collects noisy pilot observations from a small subset of ports. This mismatch between training requirements and practical channel acquisition poses a major obstacle to deployment. To address this challenge, we develop a self-supervised framework that trains the FNO-based estimator directly from noisy sub-sampled pilots without any labeled channel fields. The proposed objective combines a pilot splitting strategy, which creates supervision at ports hidden from the model input, with a correction based on Stein's unbiased risk estimate (SURE) that eliminates the measurement noise bias. We prove that this objective admits the same minimum mean square error (MMSE) solution as the ideal supervised criterion. Since the loss relies solely on received pilots, the estimator can be continually adapted online on the live pilot stream. Simulation results show that the self-supervised FNO achieves accuracy comparable to supervised training, generalizes zero-shot to unseen port resolutions, and rapidly recovers from distribution shift via online adaptation.

cs.IT · eess.SP

新论文 · v1arXiv:2611.25725

On the Hamel Coefficients and the Boltzmann-Hamel Equations for the Rigid Body

Andreas Müller

原文摘要

The Boltzmann-Hamel (BH) equations are central in the dynamics and control of nonholonomic systems described in terms of quasi-velocities. The rigid body is a classical example of such systems, and it is well-known that the BH-equations are the Newton-Euler (NE) equations when described in terms of rigid body twists as quasi-velocities. It is further known that the NE-equations are the Euler-Poincare, respectively, the reduced Euler-Lagrange equations on SE(3) when using body-fixed or spatial representation of rigid body twists. The connection between these equations are the Hamel coefficients, which are immediately identified as the structure constants of SE(3). However, an explicit coordinate-free derivation has not been presented in the literature. In this paper the Hamel coefficients for the rigid body are derived in a coordinate-free way without resorting to local coordinates describing the rigid body motion. The three most relevant choices of quasi-velocities (body-fixed, spatial, and hybrid representation of rigid body twists) are considered. The corresponding BH-equations are derived explicitly for the rotating and free floating body. Further, the Hamel equations for nonholonomically constrained rigid bodies are discussed, and demonstrated for the inhomogenous ball rolling on a plane.

cs.RO

新论文 · v1arXiv:2611.25691

Exploring Semantic Table Search in Data Lakes

Martin Pekár Christensen, Matteo Lissandrini, Katja Hose

原文摘要

Discovery of relevant datasets within data lakes remains a key challenge. Union and join search are effective techniques for discovering data when augmenting an existing table. In exploratory search, instead, keyword search is commonly adopted to find a relatively small set of highly relevant tables matching some search terms. In this context, semantic search is a new technique that exploits the presence of named entities in exemplar tuples used as queries and in target tables. Hence, keyword search and entity search are able to find a diverse set of tables. In this work, we propose \texttt{Thetis++}, which combines keyword and semantic table search to discover relevant tables that combine the semantic signal with the presence of exact keyword matches. Incorporating multiple paradigms and sources of information allow us to combine the best of multiple worlds. Therefore, we study many alternative techniques for combining rankings and evaluate these for the table search task in data lakes. We explore both unsupervised and supervised techniques, including TOPSIS. Finally, alternative to Thetis++ combining keyword and semantic relevance signals, we study a variant of learning-to-rank for the table search task in data lakes. Our evaluation highlights both the limitations of many standard approaches along with some promising directions offered by learned approaches. Furthermore, while indexes for keyword search are quite efficient, here we provide insights into how to improve the scalability of prefiltering techniques for semantic search, comparing LSH indexes with Hierarchical Navigable Small Worlds. Our experiments show that HNSW drastically improves the runtime of semantic table search by 4.7X on average whilst retaining the high-quality tables. Together, our findings plot the line for effective and efficient data discovery techniques for semantic data lakes.

cs.IR

版本更新 · v2arXiv:2609.17488

LimiX-2: A Contextual Mechanism Network Towards General Structured-Data Intelligence

Xingxuan Zhang, Gang Ren, Hao Yuan, Hao Zou, Hongze Tan, Hui Wang, Jianhao Song, Jiansheng Li 等 60 位作者

原文摘要

We introduce LimiX-2, a new model in the LimiX family, developed through model and data scaling guided by our previously established scaling laws. LimiX-2 adopts the Contextual Mechanism Networks (CMNs) paradigm and is pretrained with Context-Conditional Masked Modeling (CCMM). CMNs shifts the organizing principle of in-context learning from target-centric prediction to mechanism-oriented joint modeling. Rather than centering the network on the $p(y \mid x, D_{\mathrm{context}})$ objective of conventional tabular PFNs, it is designed around learning $p(x, y \mid D_{\mathrm{context}})$, a context-dependent representation of the joint structure underlying data generation. Pretraining uses synthetic datasets generated by structural causal models (SCMs) spanning diverse graph structures, functional mechanisms, and observation processes. Evaluations on TabArena, TALENT, and BCCO show that LimiX-2 outperforms current dataset-specific models and tabular foundation models. Beyond predictive performance, the CMN paradigm also promotes causal awareness in LimiX-2: its feature attention encodes direct causal relationships, enabling accurate causal skeleton recovery.

cs.AI