跳到正文
研踪PaperTrail

版本更新

查看已发布论文的最新修订，在详情页对照实际观测到的版本。

部分采集异常

全站更新概况

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

arXiv 最近完成扫描：2026-11-15 21:08 · 全站最近入库：2026-11-14 15:09各信源进度 →

统计口径与数据新鲜度

统计 2026-11-14 22:07 — 2026-11-15 22:07（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：2026-11-15 21:08最近发现新增版本：2026-11-14 15:09库内论文最新更新时间：2026-11-13 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。

23,348 已收录论文19,580 AI 相关6,729 有修订112 HF 社区发现

论文修订列表

研究日报 / 周报 →按信源浏览 →

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

找到 6,729 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-15 21:07（北京时间）
版本更新 · v3arXiv:2602.02799

Joint Learning of Hierarchical Neural Options and Abstract World Model

Wasu Top Piriyakulkij, Wolfgang Lehrach, Kevin Ellis, Kevin Murphy

原文摘要

Building agents that can perform new skills by composing existing skills is a long-standing goal of AI agent research. Towards this end, we investigate how to efficiently acquire a sequence of skills, formalized as hierarchical neural options. However, existing model-free hierarchical reinforcement algorithms need a lot of data. We propose a novel method, which we call AgentOWL (Option and World model Learning Agent), that jointly learns -- in a sample efficient way -- an abstract world model (abstracting across both states and time) and a set of hierarchical neural options. We show, on a subset of Object-Centric Atari games, that our method can learn more skills using less data than baseline methods and possesses learning and generalization capabilities that the baselines do not have.

cs.LG · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2601.22510

Shattered Compositionality: Counterintuitive Learning Dynamics of Transformers for Arithmetic

Xingyu Zhao, Darsh Sharma, Rheeya Uppaal, Yiqiao Zhong

原文摘要

Large language models (LLMs) often achieve strong benchmark accuracy yet remain brittle under small distribution shifts. While recent mechanistic studies reveal the discrepancy between LLMs and humans in skill compositions, the learning dynamics of skill acquisition and the role of data distributions remain elusive. In this study, we train transformers on synthetic arithmetic tasks with black-box model-agnostic metrics for analyzing non-human skill compositions. We discover that transformers often acquire skills for arithmetic in reverse order or in parallel instead of human-like sequential rules--a phenomenon we refer to as shattered compositionality. To explain these behaviors, we provide evidence that correlational matching to the training data, rather than causal or procedural composition, shapes learning dynamics. As a consequence, this non-human acquisition creates competition between partially learned skills, producing characteristic mixing errors and weaker robustness under controlled distribution shifts. We further show that the same qualitative behavior persists in modern LLMs and is not mitigated by pure model scaling or scratchpad supervision. Our results highlight a mismatch between training-time skill acquisition and the human-like hierarchical compositions, with implications for reasoning reliability and out-of-distribution robustness.

cs.LG · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2606.18062

Security and Privacy Prompts in the Wild: What Users Ask LLMs and How LLMs Respond

Hobin Kim, Xiaoyuan Wu, Omer Akgul, Lujo Bauer, Nicolas Christin

原文摘要

Large language models (LLMs) are widely used to answer users' information needs: users ask LLMs about the weather, pose educational questions, and consult them for legal assistance. One particularly understudied area is digital security and privacy (S&P), where users may seek LLMs' help on how to secure their online accounts or protect their computers from cyber attacks. To the best of our knowledge, no prior study has collected or analyzed the S&P questions users ask LLMs; prior research on LLM response quality relied on expert-authored S&P misconceptions or FAQs rather than user queries. Drawing from WildChat, a dataset of 3.2M user-LLM conversations collected in the wild, our study identifies 14,727 S&P prompts and categorizes them into nine categories covering a wide range of S&P topics. We sampled 450 S&P prompts, and performed a thematic analysis to characterize the S&P questions users ask LLMs. Separate from the thematic analysis, we curated 270 advice-seeking S&P prompts, where users ask for recommendations, guidance, or specific S&P information. We measured LLM response quality and consistency when posing each prompt to the LLMs 10 times. We found that commercial LLMs outperform open-weight models (GPT 5.5 provided "good enough" responses on 98% of prompts; Llama 4 on 47%). However, among prompts that received high-quality responses on average, commercial models sometimes produce semantically inconsistent responses across runs (some are contradictory, and others are different in specificity), thereby risking confusing or misleading users.

cs.CL · cs.AI · cs.CR · cs.HC

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2608.04222

TIDE: A Physically Diverse 3D Turbulence Benchmark Dataset for Advancing Scientific Machine Learning

Yilong Dai, Yiming Sun, Yiheng Chen, Shengyu Chen, Peyman Givi, Xiaowei Jia, Runlong Yu

原文摘要

Turbulence is a central testbed for machine learning on physical dynamics because its governing laws are known exactly. However, most existing studies remain in 2D, while 3D turbulence has fundamentally different physics and is far more costly to simulate. Existing 3D resources also typically provide only one realization per configuration, making it difficult to distinguish learning the dynamics from fitting the statistics of a single flow. In this paper, we introduce TIDE (Turbulent Incompressible DNS Ensembles), a 256^3 DNS corpus and benchmark for 3D incompressible turbulence, with 15 configurations on eight controlled axes, independent ensembles, pressure fields, and equation-level verification. The benchmark includes five tasks, standardized learned baselines, controlled generalization splits, and physical-fidelity metrics alongside pointwise error. Across the main forecasting configurations, current learned models trained under a fixed reference recipe barely outperform persistence and still make about twice the error of a spectral solver given the true equations. Moreover, lower pointwise error can coincide with severely distorted small-scale dynamics, showing that accuracy alone does not ensure physical fidelity. Rollout stability depends on the configuration, the training seed, and how densely training samples each trajectory. Joint training across neighboring regimes does not close the transfer gaps we test, and the forced-to-decay contrast bounds a conditioning limit: an estimate of the current forcing state removes 20-41% of the error a spectral solver makes without forcing input, while history or forcing inputs help only some learned architectures. Closing these accuracy, fidelity, and conditioning gaps is the central open problem made measurable by TIDE. https://github.com/Dyloong1/TIDE-dataset-benchmark

physics.flu-dyn · cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2611.06717

Combinatorial Boolean Matrix Multiplication in $n^3 / 2^{Ω(\sqrt{\log n})}$ Time

Piaoyang Cui, Yiming Liu, David P. Woodruff

原文摘要

For over fifty years, combinatorial algorithms for Boolean matrix multiplication (BMM) and triangle detection saved only polylogarithmic factors over cubic time. At STOC 2024, Abboud, Fischer, Kelley, Lovett, and Meka broke this barrier with a deterministic $O(n^3 / 2^{Ω((\log n)^{1/7})})$-time algorithm, and asked whether $n^3 / 2^{Ω(\sqrt{\log n})}$ time is achievable. We answer this question in the affirmative, giving deterministic combinatorial algorithms for both BMM and triangle detection running in $O(n^3 / 2^{Ω(\sqrt{\log n})})$ time. Following Abboud et al., we decompose $(A,B)$ into pairs of submatrices $(A_k,B_k)$, called \emph{pieces}, each of which is either sparse or pseudorandom in a sense that forces $A_k\star B_k$ to be nearly all-ones on its vertex sets. Whereas their decomposition produces $2^{O(d^7)}$ pieces at regularity parameter $d$, we obtain $2^{O(d^2)}$ pieces, directly improving the runtime exponent from $(\log n)^{1/7}$ to $\sqrt{\log n}$. Prior work found pseudorandom pieces by repeatedly restricting to denser submatrices, paying worst-case vertex losses at each step that compounded across nested recursion levels, together with an expensive derandomization step. We eliminate the extra powers of $d$ by matching the vertex loss at each step to the size of its density jump so losses telescope across scales, derandomizing witness search via $L_d$-norm estimation so the sampler length stays $2^{O(d)}$, and bounding overlap across the recursion with a fractional density-weighted volume budget rather than raw submatrix volume.

cs.DS

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2606.03871

Visual Instruction Tuning Aligns Modalities through Abstraction

Luis Palacios, Lorenzo Basile, Diego Doimo, Alberto Cazzaniga

原文摘要

Visual instruction tuning effectively adapts a pre-trained Large Language Model (LLM) to process image information alongside text. Yet it remains unclear how visual features are embedded within the layer-wise hierarchy of abstractions in the LLM backbone. Across a diverse set of vision-language architectures, we show that instruction tuning primarily serves as a bridge, embedding visual features directly into the intermediate semantic layers of the LLM, bypassing the early layers devoted to unimodal processing. With probing analyses and causal interventions, we show that these intermediate layers constitute the semantic core of vision-language processing and play a critical role in performance across a broad set of multimodal benchmarks. In addition, by comparing the geometry of semantically equivalent visual and textual representations, we find that fine-tuning extends and strengthens the existing abstraction phase, aligning visual features with pre-existing textual ones. Finally, we confirm the functional role of this localized alignment by restricting fine-tuning to intermediate layers alone: this strategy preserves the performance of full fine-tuning on vision-centric benchmarks while reducing training time. Our results suggest that multimodal integration is a localized phenomenon driven by the repurposing of the internal abstraction engine of the LLM.

cs.CV · cs.CL · cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.03391

Native Action-Prior Learning from Videos for World Action Models

Zhaochong An, Fei Zhang, Menglin Jia, Duncan Frost, Zijian Zhou, Yikai Wang, Xudong Wang, Aditya Patel 等 13 位作者

原文摘要

World action models integrate future visual dynamics with robot action prediction, but their scalability remains limited by the need for action-annotated robot trajectories. Observation-only videos contain rich evidence about interaction dynamics, but existing approaches typically use them either to pretrain visual representations that must later be adapted for control, or to infer latent actions that are subsequently grounded to robot commands. We present NAVA-WAM, which introduces native action-prior learning by directly pretraining the action policy from observation-only videos, avoiding indirect representation-to-control transfer or a separate latent-action model. Our training consists of two stages. First, we pretrain on observation-only videos, where future-video flow-matching supervision over visual transitions is propagated through transition-structured joint attention to optimize the Action-DiT and learn action-relevant priors. Second, we use action-labeled demonstrations to post-train the Action-DiT for robot control through joint video--action flow matching, while asymmetric attention decouples the visual branch from iterative action denoising and enables efficient action-only inference. Extensive experiments show that NAVA-WAM consistently outperforms prior approaches under both in-distribution and out-of-distribution settings, while demonstrating strong action-label efficiency and effective real-robot generalization. These results establish native action-prior learning as an effective approach to directly pretrain action policies from observation-only videos, providing a scalable path beyond action-labeled robot data.

cs.CV · cs.RO

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2601.20154

Spectral Ghost in Representation Learning: A Unified View from Component Analysis to Self-Supervised Learning

Bo Dai, Na Li, Dale Schuurmans

原文摘要

Self-supervised learning (SSL) extracts representations from massive unlabeled data that transfer to a variety of downstream tasks. Its empirical success has produced a dizzying array of objectives and procedures, yet a clear understanding of what representation is desired, and why these seemingly different methods work, is still missing. In this paper, we develop a unified spectral framework for representation learning. We first show that a representation is sufficient for arbitrary downstream prediction with a linear head if it spans the spectral decomposition of the pointwise mutual information, which can be extracted purely from unlabeled data. This view organizes representation learning along two axes, the parametrization of the spectral representation (linear, energy-based, denoising, nonlinear, latent-variable, and multi-modal) and the learning objective for fitting it, under which a broad range of existing methods, including SimCLR, MoCo, BYOL, Barlow Twins, VICReg, DINO, MAE, JEPA, BERT, GPT, and CLIP, as well as classic PCA, CCA, Laplacian eigenmaps, LLE, and SNE, are recovered as particular combinations. The framework also clarifies optimization behaviors: the batch-size requirement is determined by whether the objective is unbiased, a consistent surrogate, or biased, rather than by contrastive vs. non-contrastive losses. Finally, the framework inspires new scalable algorithms and extends spectral representations to causal inference and reinforcement learning.

cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2605.10919

Random Access Expectation in DNA Data Storage: Asymptotic Analysis Through Fountain Codes

Christoph Hofmeister, Rawad Bitar, Eitan Yaakobi

原文摘要

The expected number of coded symbols drawn from a linear code until a desired information symbol can be decoded is studied. This quantity is of interest in DNA data storage, where it is called the random access expectation. If the generator matrix of the code is binary and fully symmetric, the process of sampling symbols from the code is equivalent to the encoder of an LT code. Based on this observation, the random access expectation of binary fully symmetric codes under a peeling decoder is analyzed in the large blocklength limit. Asymptotically, a random access expectation, normalized by the number of information symbols, of $0.7869$ is achievable, while for a binary fully symmetric code under a peeling decoder it cannot be below $π/4\approx 0.7854$. The analysis is extended to the case where a set of information symbols needs to be decoded.

cs.IT

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2410.12755

Toward Optimal-Complexity Hash-Based Asynchronous MVBA with Optimal Resilience

Jovan Komatovic, Joachim Neu, Tim Roughgarden

原文摘要

Multi-valued validated Byzantine agreement (MVBA), a fundamental primitive of distributed computing, allows $n$ processes to agree on a valid $\ell$-bit value, despite $t$ faulty processes behaving maliciously. Among hash-based solutions for the asynchronous setting with adaptive faults, the state-of-the-art HMVBA protocol achieves optimal $O(n^2)$ message complexity, (near-)optimal $O(n \ell + n^2 λ\log n)$ bit complexity, and optimal $O(1)$ time complexity. However, it only tolerates $t < \frac15 n$ failures. In contrast, the best-known optimally-resilient protocol, SQ, incurs a higher bit complexity of $O(n^2 \ell + n^3 λ)$. This poses a fundamental question: Can a hash-based protocol be designed for the asynchronous setting with adaptive faults that simultaneously achieves optimal complexity and optimal resilience? This paper takes a significant step toward answering this question. Namely, we introduce Reducer, an MVBA protocol that retains HMVBA's optimal complexity while improving its resilience to $t < \frac14 n$. Like HMVBA and SQ, Reducer relies exclusively on collision-resistant hash functions. A key innovation in Reducer's design is its internal use of strong multi-valued Byzantine agreement (SMBA), a new variant of Byzantine agreement we introduce and construct, which ensures that the decided value was proposed by a correct process. To further advance resilience toward the optimal one-third bound, we then propose Reducer++, an MVBA protocol that tolerates up to $t < (\frac13 - ε)n$ adaptive failures, for any fixed constant $ε> 0$. Unlike Reducer, Reducer++ does not rely on SMBA. Instead, it employs a novel approach involving hash functions modeled as random oracles to ensure termination. Reducer++ maintains constant time complexity, quadratic message complexity, and quasi-quadratic bit complexity, with constants dependent on $ε$.

cs.DC

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2504.00001

HistogramTools for Efficient Data Analysis and Distribution Representation in Large Data Sets

Shubham Malhotra

原文摘要

Histograms provide a powerful means of summarizing large data sets by representing their distribution in a compact, binned form. The HistogramTools R package enhances R built-in histogram functionality, offering advanced methods for manipulating and analyzing histograms, especially in large-scale data environments. Key features include the ability to serialize histograms using Protocol Buffers for distributed computing tasks, tools for merging and modifying histograms, and techniques for measuring and visualizing information loss in histogram representations. The package is particularly suited for environments utilizing MapReduce, where efficient storage and data sharing are critical. This paper presents various methods of histogram bin manipulation, distance measures, quantile approximation, and error estimation in cumulative distribution functions (CDFs) derived from histograms. Visualization techniques and efficient storage representations are also discussed alongside applications for large data processing and distributed computing tasks.

cs.DB · cs.PF

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2508.11682

Re-evaluating Heart Rate Variability Biomarkers for Glucose Sensing: The Impact of Age Normalisation and Subject-Independent Validation

Md Basit Azam, Sarangthem Ibotombi Singh

原文摘要

Heart rate variability (HRV) from electrocardiogram (ECG) signals may indicate glycemic status, but prior studies often combine distinct glucose targets and use validation strategies vulnerable to data leakage. Since HRV declines with age, aging may be confounded with diabetes-related autonomic dysfunction. We tested whether age-dependent HRV normalisation improves glycemic estimation. We analysed ECG features from 43 male patients with type 2 diabetes. Outcomes were glycated hemoglobin (HbA1c; n=29) and fasting blood glucose (FBG; n=38). We used leave-one-subject-out cross-validation, with feature selection and standardisation within each training fold. We compared 20 machine-learning algorithms, six age-adjustment methods, and 20 normalisation parameter combinations. Validation included 500 permutations and bootstrap confidence intervals. Extra Trees regression performed best: R2=0.222 (r=0.476, p=0.009) for HbA1c and R2=0.086 (r=0.344, p=0.034) for FBG. Mean absolute errors were 1.18 percentage points and 2.27 mmol/L, respectively. Permutation testing supported both associations (p=0.002). Contrary to our hypothesis, no age-adjustment method or parameter combination improved performance. Tree ensembles retained positive performance under strict validation, whereas linear models had negative R2 values; neural networks performed poorly. Strict within-fold preprocessing materially changed model performance and revealed potential inflation from conventional validation. Bootstrap 95% confidence intervals were [0.13, 0.82] for HbA1c R2 and [0.10, 0.72] for FBG R2. Although findings support HRV-glycemic associations, performance remains insufficient for standalone clinical use. The study highlights the importance of separating glycemic targets, subject-independent validation, within-fold preprocessing, and comprehensive baselines in non-invasive glycemic monitoring research.

eess.SP · cs.AI · cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2506.10239

A Unified Framework for Probabilistic Dynamic-, Trajectory- and Vision-based Virtual Fixtures

Maximilian Mühlbauer, Bernhard Weber, Sylvain Calinon, Freek Stulp, Alin Albu-Schäffer, João Silvério

原文摘要

Probabilistic Virtual Fixtures (VFs) enable the adaptive selection of the most suitable haptic feedback for each phase of a task, based on learned or perceived uncertainty. While keeping the human in the loop remains essential, for instance, to ensure high precision, partial automation of certain task phases is critical for productivity. We present a unified framework for probabilistic VFs that seamlessly switches between manual fixtures, semi-automated fixtures (with the human handling precise tasks), and full autonomy. We introduce a novel probabilistic Dynamical System-based VF for coarse guidance, enabling the robot to autonomously complete certain task phases while keeping the human operator in the loop. For tasks requiring precise guidance, we extend probabilistic position-based trajectory fixtures with automation, allowing for seamless human interaction, geometry-awareness and optimal impedance gains. For manual tasks requiring very precise guidance, we also extend visual servoing fixtures with the same geometry-awareness and impedance behavior. We validate our approach on different robots, including an evaluation with expert users, showcasing operation modes, the ease of programming fixtures and lower interaction forces and favorable usability compared to a baseline.

cs.RO

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2511.04556

Flow Reconstruction from Sparse Measurements in Urban Drainage Networks: A Simulation-Based Evaluation of Data-Driven Sparse Sensing

Zihang Ding, Amit Kumar, Imran Md. Azizul Islam, Mila Avellar Montezuma, Ruihang Zhang, Kun Zhang

原文摘要

Urbanization and increasingly frequent intense storms are placing stress on urban drainage networks. While dense monitoring is desirable, practical constraints in time, budget, and technology hinder its full implementation. How to monitor and predict flow conditions across the entire network under constrained resources is a major challenge. To address this, we evaluated data-driven sparse sensing (DSS), which selects sensor locations and reconstructs network-wide flow from the same reduced basis, in a 76-conduit urban drainage network. Simulations of a calibrated SWMM model were used to construct an event-weighted spatial basis using singular value decomposition (SVD), select rank-specific layouts using pivoted QR, and define the reconstruction decoder. Applied to 225 held-out simulations combining 25 plausible calibrated parameter sets with nine rainfall events, monitoring 5 of the 76 conduits (6.6%) achieved a median system-level Nash-Sutcliffe efficiency (NSE) of 0.992 (10th percentile 0.973) for observed storms, and a median NSE of 0.967 at the active unmonitored conduits. Pivoted QR performed comparably to genetic-algorithm layouts without requiring iterative layout searches, whereas largest-upstream-area placement failed with the DSS decoder and trailed QR even with a weighted regression decoder. The DSS flow reconstruction performance was insensitive to prescribed multiplicative Gaussian measurement noise, additive noise up to about 2 L/s and a 10 L/s detection floor, but showed location-dependent responses to loss of a monitored conduit, which a refitted regression decoder largely removed. This study supports the use of DSS, driven by physics-based simulations, for deciding how many flow sensors to install and where.

cs.AI · cs.CE

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.31613

Progress Probes $P^2$: Lightweight Runtime Monitoring for Vision-Language-Action Models

Atiksh Bhardwaj, Edward W. Duan, Junwon Seo, Prithwish Dan, Wei-Chiu Ma, Andrea Bajscy, Preston Culbertson

原文摘要

Vision-Language-Action (VLA) models can perform diverse manipulation tasks directly from visual observations and natural language instructions, but they remain prone to mistakes and unpredictable failures. Deploying these policies without continuous human supervision will require robust runtime monitors that detect when the policy has gone off-course. Existing monitors use external foundation models, such as vision-language models (VLMs) to supervise the policy, introducing significant complexity and computational cost for monitoring at runtime. In this work, we investigate an alternate approach: \textit{can a VLA's own internal activations provide the information needed to monitor its execution?} Our method trains linear probes on policy activations from demonstrations to estimate progress, then flags executions that deviate from expected progress. In experiments with the open-source VLA $π_{0.5}$ and across simulated and real-world manipulation tasks, this approach is competitive with or outperforms VLM-based monitors in task failure detection, as well as methods which require failure data for training. These findings suggest that simple probes of VLA internals can be used to develop reliable runtime monitors, opening a path toward building policies that can act reliably and monitor their execution without external foundation models.

cs.RO

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2604.26020

Training Computer Use Agents to Assess the Usability of Graphical User Interfaces

Alice Gao, Weixi Tong, Rishab Vempati, Katharina Reinecke, R. Benjamin Shapiro, Tianyi Zhang, Jason Wu

原文摘要

Recent HCI research leverages AI agents to assess graphical user interfaces (GUIs) usability, but current agents still struggle. Computer use agents (CUAs) that are used in these assessments are predominantly developed using objective task trajectories with verifiable outcomes and lack an operationalized definition of usability and necessary training data. This work presents a novel training objective that operationalizes usability to fine-tune CUAs to assess GUI usability through relative comparisons between interface variants. We apply our method to CUA models using a large-scale dataset of fully interactive UIs paired with usability labels mixed with human preferences.Two empirical evaluations across real and synthetic GUIs show that our method i) improved relative usability assessment accuracy across three off-the-shelf CUA models, ii) enabled our best model to outperform proprietary baselines, and iii) produced grounded usability assessments. Our work lays a data-driven foundation for agent-based usability assessment, enabling exploration of new forms of scalable usability evaluation and interface design.

cs.CL · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2609.39118

Diagnosing On-Policy Self-Distillation for Reasoning Language Models

Yang Li, Gongle Xue, Yuheng Yuan, Yijia Guo, Shizhe Zhang, Liwen Hu, Lei Ma

原文摘要

On-policy self-distillation (OPSD) has attracted growing interest as a promising approach to improve the reasoning ability of language models. Without external rewards nor a separate stronger teacher, the self-teacher with privileged information could provide dense signals on student's trajectories. However, its behavior in language reasoning remains unclear, with reported outcomes ranging from modest gains to behavioral collapse. In this work, we diagnose OPSD for mathematical reasoning across models spanning 0.6B--8B parameters. We conduct controlled experiments and token-level analyses to fully delve into OPSD. We point out that teacher's signal is shaped by reasoning-mode alignment and the complete teacher prefix, rather than by privileged semantics alone. OPSD improves reasoning only in narrow compatibility regimes. Otherwise, it produces ineffective length growth, stable degradation, or behavioral collapse. Token-level analysis shows that teacher's signal is not stable and does not predict downstream performance. Based on these results, we argue that OPSD is a sensitive algorithm rather than a generally reliable reasoning-improvement post-training method.

cs.CL

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2606.24177

Agon: An Autonomous Large-Scale Omnidisciplinary Research System Built on Prompt Economy

Youran Sun, Xingyu Ren, Chugang Yi, Jiaxuan Guo, Kejia Zhang, Jianda Du, Haizhao Yang

原文摘要

Large language models are making research scalable. We present \textsc{Agon}, a research orchestrator designed to probe the limits of automated research. \textsc{Agon} is built on six design principles: Prompt Economy, Future-Facing, Minimal Prompts, OmniDisciplinary, Massive Parallelism, and Zero-Code. We deployed \textsc{Agon} across domains using only brief topic descriptions as starting points and no human-written code. These deployments demonstrate scalability but also reveal previously unseen classes of failure that are difficult to address through prompt engineering. We organize these failures into a taxonomy along four dimensions: severity, fixability, visibility, and capability locus. Together, these results show that \textsc{Agon} is pushing research towards a new paradigm: machines scale, humans steer. Our code is publicly available at \href{https://github.com/AutoResearch-Factory/Agon/tree/lite}{this GitHub repository}.

cs.SE · cs.AI · cs.CL · cs.MA

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2511.17044

Parametric Retrieval-Augmented Generation using Latent Routing of LoRA Adapters

Zhan Su, Fengran Mo, Jinghan Zhang, Yuchen Hui, Jiaao Sun, Jian-yun Nie

原文摘要

Parametric Retrieval-Augmented Generation (PRAG) is a RAG approach that integrates external knowledge directly into model parameters using a LoRA adapter, aiming at reducing the inference cost compared to traditional RAG. However, current PRAG approaches adopt a \textit{one-to-one} document encoding scheme, using a dedicated LoRA adapter for each individual document. This scheme introduces two major limitations: 1) As the number of documents increases, there will be a prohibitive cost for training and storage. 2) The LoRA adapters may largely overlap due to the shared knowledge across documents, making the approach highly inefficient. To overcome these challenges, we propose the Poly-PRAG approach, which uses a small set of LoRA adapters that are able to encode more general knowledge. Each document can be encoded using a combination of them through a latent routing function. By jointly training the LoRA adapters and the latent routing function, each LoRA adapter is able to encode a shared part of the knowledge across documents, and the routing function can select the best combination of adapters for a document. Experimental results on four benchmarks demonstrate the effectiveness of the Poly-PRAG compared to other strong PRAG baselines. In addition, this approach reduces the storage requirement by avoiding the need to store a large number of LoRA adapters and offers a more efficient way to encode external knowledge into LLMs.

cs.IR

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2609.21439

People escalate against a competitor labelled human and hold back against one labelled an optimising machine

Vinicius Ferraz, Leon Houf

原文摘要

People increasingly compete against AI agents, and organisations decide what to tell them about these opponents. We distinguish an opponent effect, specific to the system faced, from an information effect, a property of the disclosure. In a preregistered, deception-free experiment (N = 1,395), we crossed what participants were told about their opponent with whom they faced in a live all-pay auction in which escalation arises from the incentives. Holding the opponent fixed, being told a human was or might be the opponent raised the median price by 6.7% of the prize value, and being told an optimising machine might be lowered it by 8.8%. Facing an AI agent lowered prices and made it rarer that both players finished with positive earnings. Exploratory analyses indicate that the labels changed how often people raised again while already ahead, not how often they responded when behind. Describing an AI competitor is not behaviourally neutral.

cs.HC · econ.GN

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2611.10380

Quantitative Assessment of Crystal Structure Prediction

Sergio Rincón, Gabriel González, Nicolás Andrade, Rafael Velásquez, Sebastián Ojeda, Juanita Puentes, Nicolás Aparicio, Paula Cárdenas 等 9 位作者

原文摘要

Crystal Structure Prediction (CSP) aims to predict a material's 3D structure from its chemical composition, with applications from energy storage to biomedical devices. Despite its importance, existing evaluation frameworks suffer from key limitations: they rely on threshold-sensitive or incomparable metrics, focus predominantly on geometric performance while overlooking physical plausibility, and fail to account for polymorphism, where a chemical composition can crystallize into multiple stable structures. In parallel, recent Symmetry-Informed CSP (SICSP) methods often report results alongside standard CSP baselines while receiving crystallographic templates at inference, which constrain the search space and make SICSP a fundamentally different task. To address these limitations, we introduce an Orientation-invariant, Polymorph-Aware, Large Benchmark (OPAL-Bench) that evaluates CSP and SICSP as separate tasks. We evaluate 7 CSP and 2 SICSP methods under OPAL-Bench across 4 datasets using threshold-free geometric and physical distance metrics. Our results show that no method dominates across all metrics, and that geometric performance is not strongly correlated with physical plausibility. We also release OPAL-CSP, an SE(3)-equivariant flow-matching baseline for the CSP track that performs competitively on geometric metrics and ranks among the best on physical ones. By unifying geometric and physical metrics under a polymorph-aware protocol, OPAL-Bench enables a more comprehensive assessment of generative quality in CSP.

cs.LG · cond-mat.mtrl-sci · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2602.02605

Fine-Tuning Language Models to Know What They Know

Sangjun Park, Elliot Meyerson, Xin Qiu, Risto Miikkulainen

原文摘要

Evaluating true metacognition in Large Language Models (LLMs) is difficult due to biases and heuristics. This paper presents a framework to measure and enhance LLM metacognition while controlling for these biases. A measurement method using the $d'_{\rm type2}$ metric is established to isolate metacognitive ability. The Evolution Strategy for Metacognitive Alignment (ESMA) is proposed, demonstrating robust generalization across unseen datasets, languages, and newly acquired knowledge. Gradual parameter updates reveal that these improvements are driven by a sparse set of parameters, opening a potential avenue for targeted metacognitive optimization.

cs.NE · cs.AI · cs.CL · q-bio.NC

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2601.12716

CellularS2-Bench: A Staged, Evidence-Grounded Benchmark for Cellular Network Security Reasoning

Ke Xie, Xingyi Zhao, Yiwen Hu, Haotian Lin, Shuhan Yuan, Wen Li, Tian Xie

原文摘要

Cellular networks are critical infrastructure supporting billions of worldwide users and safety- and mission-critical services. Vulnerabilities in cellular networks can therefore cause service disruption, privacy breaches, and broad societal harm, motivating growing efforts to analyze 3GPP specifications that define required device and operator behavior. However, existing benchmarks for this domain share three systematic deficiencies: they rely on LLM-generated entries without expert verification, they do not require faithful interpretation of normative language, reasoning across cross-referenced clauses, and verifiable conclusions grounded in multimodal evidence such as tables and figures, and they do not evaluate security reasoning over end-to-end protocol behavior. To address these gaps, we present CellularS2-Bench, a staged benchmark. CellularS2-Bench covers core control-plane specifications, processing over 1,866 paragraphs, 22,033 sentences, 424 figures, and 248 tables. It comprises 13,851 questions and instances across three progressive stages, of which 10,269 (74.14%) are newly constructed from 3GPP specifications and 43 real-world vulnerabilities. Beyond newly constructed tasks, CellularS2-Bench integrates and corrects 3,582 (25.86%) instances from 7 prior datasets to broaden task coverage. To establish a reproducible baseline on CellularS2-Bench, we introduce CellularS2-ARI, a unified Adapt-Retrieve-Integrate framework for systematic understanding and specification-driven security analysis of 3GPP specifications. Together, they establish an accessible and reproducible foundation for quantifying progress in specification understanding and security reasoning in cellular network security.

cs.CR

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2611.07289

Patch-Based Generalized Neural Texture Upscaling

Shin Fujieda, Takahiro Harada

原文摘要

We propose a lightweight neural texture upscaling method. A single generalized MLP is trained on a diverse collection of texture patches and integrated into the texture sampling stage, upscaling low-resolution textures on-the-fly at rendering time. Because the model is generalized rather than overfit to specific content, no per-asset training is required - the same model applies to any existing scene without modifying texture data or the asset pipeline. Storing lower-resolution base textures reduces disk footprint, upscaling at sampling time eliminates the need to store high-resolution textures in VRAM, and sharing a single set of model weights across all textures ensures non-divergent GPU execution. We evaluate our method across multiple texture classes, demonstrating consistent improvement over analytic upscaling methods in perceptual quality.

cs.GR

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2611.08698

VICTR: Value-Informed In-Context Learning for Vision-Language-Action Models

Luca Macesanu, Tianluo Zhang, Aditya Patil, Archit Sharma, Kevin Angers, Christopher M. Clark, Chen Feng, Juexiao Zhang

原文摘要

Recent advances in robot foundation models have highlighted the potential of in-context learning (ICL) for adapting robot behavior from demonstrations without further fine-tuning. Previous work has used retrieval-based post-training to equip pretrained vision-language-action (VLA) models with the capability for ICL by teaching them to condition action predictions on retrieved visually-relevant demonstration chunks. However, the effectiveness of retrieval hinges on selecting context relevant to the current execution stage, which visual similarity alone may not capture. We introduce VICTR, a framework enabling in-context learning in VLAs via progress-informed context retrieval built around an In-Context Value Function Estimator (IC-VFE). IC-VFE conditions on a reference demonstration video, a language instruction and the current observation to estimate task-relative progress. By conditioning progress estimation on a reference demonstration, IC-VFE can align the current state with a demonstrated execution even for tasks unseen during training. VICTR uses the estimated progress to select demonstration chunks at corresponding stages of execution and supplies them to a $π_{0.5}$ policy post-trained for in-context action generation. On unseen out-of-domain real-world manipulation tasks, VICTR raises the success rate from 6.2% for a fine-tuned $π_{0.5}$ baseline to 23.8% and more than doubles the task progress rate, from 28.9% to 55.5%. Project page with code, data and videos: https://ai4ce.github.io/VICTR/

cs.RO

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2611.09422

Robust & Scalable Website Fingerprinting via Minimal DRAM Traffic

Shubham Roy, Samar Perwez, Biswabandan Panda

原文摘要

Website fingerprinting attacks infer websites visited by users and threaten their privacy. These attacks are classified as on-path, which monitor network traffic, and co-located, which exploit shared system resources. Existing co-located attacks, such as sweep-counting and loop-counting, achieve over 90% accuracy but suffer significant performance degradation on many-core systems. Loop-counting accuracy drops from 93% on 8 cores to 50% on 24 cores, while sweep-counting accuracy decreases from 84% to 45% across multiple Core Complex Dies (CCDs). To address these limitations, we propose memory-chasing, a scalable website fingerprinting attack that exploits DRAM traffic interference through pointer-chasing traversals with minimal interference. By counting DRAM accesses over fixed intervals, the attack generates distinctive traces for website classification. Evaluated on the top 50 websites, memory-chasing achieves 96% accuracy on a 24-core system, 86% across multiple CCDs, 80% across multiple sockets, and 97% in an open-world setting. Extensive evaluation across hardware configurations, browsers, operating systems, isolation mechanisms, system noise, and countermeasures demonstrates its robustness and scalability.

cs.CR · cs.AR

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.15903

BRQGen: Towards Adaptive Completeness Assessment and Follow-Up Question Generation for Bug Reports

Tevfik Emre Sungur, Akif Emre Köşüş, Uğurcan Doğruçam, Eray Tüzün

原文摘要

Many bug reports lack the necessary information required to reproduce issues. In early studies, large language models have been employed to generate follow-up questions to enhance incomplete reports. However, they ask a fixed number of questions regardless of whether the report is incomplete. We propose BRQGen, an iterative framework that identifies missing or ambiguous information in bug reports and generates targeted follow-up questions, only if necessary. Reports with no identified gaps are classified as sufficient. When the author responds, BRQGen evaluates whether the answers provide the requested information and generates further questions for any exposed gaps. We curated a novel dataset of 429 incomplete bug reports from Home Assistant and Brave Browser, each containing at least one maintainer follow-up question. Each question is mapped to the author's answer and assigned a label indicating whether the requested information could have been provided directly through the issue template. Earlier maintainer requests were associated with author responses, whereas delayed requests were associated with no response. Among responding authors, requests within one hour received median replies after 0.52 hours in Home Assistant and 1.45 hours in Brave, versus 9.76 and 13.49 hours after four hours. BRQGen covered 35.4% (Qwen 3.8 Flash) to 51.8% (DeepSeek V4.1 Flash) of the 1,073 ground-truth follow-up questions. DeepSeek V4.1 Flash managed to ask 45.7% of all detailed questions and 63.9% of all generic questions. Gemini 3 Flash Preview reached the highest macro-F1 for completeness assessment of 49.3%, but with a post-hoc analysis by getting only the gaps with high severities, Qwen 3.8 Flash reaches 73.6% macro-F1 score. Completeness assessment improved when at least one identified gap had high severity, and nearly all ground-truth questions were captured within the first two iterations.

cs.SE

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2605.31460

On-Device Robotic Planning: Eliminating Inference Redundancy for Efficient Decision-Making

Joonhee Lee, Sungmin Lee, Hyunseung Shin, Hyunmi Kim, Pei Zhang, Jeonggil Ko

原文摘要

Reasoning-based robotic policies using large language and vision-language models achieve strong semantic planning capabilities but mostly suffer from a high inference latency that limits practical real-time deployment. In this work, we observe that robotic reasoning workloads contain substantial temporal redundancy, where consecutive observations frequently produce identical actions and subgoals. Based on this insight, we present REIS, a human cognition inspired robotic decision-making framework that minimizes unnecessary reasoning while preserving semantic adaptability. REIS combines lightweight scene gating, KV-steered affordance routing, and deliberative reasoning to accelerate robotic control under embodied constraints. Experiments on ALFRED, and real-world robotic tasks demonstrate that REIS significantly suppresses reasoning overhead while maintaining competitive task performance

cs.RO · eess.SY

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2606.17413

Amortized Probabilistic Retrieval of Atmospheric CO2 from OCO-2 Spectra Using Deep Learning with Laplace Approximations and Normalizing Flows

Alejandro Calle-Saldarriaga, Felix Jimenez, Jack Grosskreuz, Jiazheng Wang, Jonathan Hobbs, Matthias Katzfuss

原文摘要

Space-based monitoring of atmospheric carbon dioxide (CO$_2$) constrains the global carbon budget. NASA's Orbiting Carbon Observatory-2 (OCO-2) estimates column-averaged dry-air mole fractions of CO$_2$ (XCO$_2$) from high-resolution spectra, but operational retrievals are computationally expensive and impose stringent Gaussianity assumptions on the retrieved posterior. We present a deep learning framework that addresses both through amortized probabilistic inference. Lacking ground truth for real observations, we train and evaluate on a high-fidelity OCO-2 simulation ensemble with calibrated forward-model errors, comparing against the version-10 operational full-physics retrieval, without its bias correction, on the same radiances. Our architecture encodes each spectral band separately and estimates posteriors of the full CO$_2$ column, or summaries thereof, with Laplace approximations and conditional normalizing flows. Once trained, inference costs milliseconds per sounding rather than minutes. Trained on simulations that explicitly include forward-model discrepancy, our retrievals are more accurate than the operational one for XCO$_2$ on both data partitions we consider, and competitive on profiles. The flow represents asymmetric posteriors that a Gaussian cannot, a gain attributable to shape rather than scale, and its advantage in predictive density persists where its accuracy advantage does not. These results are established on a land-only ensemble against one configuration of the operational algorithm. On held-out reference soundings and two unseen months the XCO$_2$ and density advantages persist. For XCO$_2$, coverage is near nominal on February data and, though below nominal on the unseen months, remains far closer to it than the operational retrieval's.

cs.LG · stat.AP

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2609.39532

Tight Liquid Welfare Guarantees for Auctions with Budgets

Twan Kroll, Guido Schäfer, Artem Tsikiridis

原文摘要

We study the efficiency of auctions with budget-constrained bidders and derive tight price of anarchy (PoA) bounds for coarse correlated equilibria (CCE). To this end, we develop a general framework that reduces proving PoA guarantees to bounding the optimal value of a linear program. The LP is formulated over a small number of per-bidder equilibrium statistics, incorporates constraints implied by the auction setting and equilibrium behavior, and captures a novel smoothness notion tailored to budget-constrained environments; we provide a reduction from classical smoothness to this new notion. Our approach enables simple and tight PoA analyses across a broad range of auction formats. Using our framework, we obtain tight liquid welfare guarantees for simultaneous first-price, second-price, and all-pay auctions, simultaneous first-price auctions in which every bidder scales her values by a single bid multiplier (Deng et al., 2021), discriminatory and uniform price multi-unit auctions, and generalized first-price and second-price position auctions. Beyond budget-constrained settings, we also prove tight bounds for the budget-free uniform price auction and the budget-free generalized second-price auction.

cs.GT

查看论文
arXiv ↗PDF ↗