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

+479新增论文
+361其中 AI 相关
+63已有论文修订
+93新增 AI 导读

arXiv 最近完成扫描:2026-11-16 01:34 · 全站最近入库:2026-11-16 13:44各信源进度 →

统计口径与数据新鲜度

统计 2026-11-15 20:18 — 2026-11-16 20:18(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:2026-11-16 01:34最近发现新增版本:2026-11-16 13:44库内论文最新更新时间:2026-11-14 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。

23,827 已收录论文19,941 AI 相关6,921 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 23,827 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-16 01:33(北京时间)
版本更新 · v4arXiv:2606.04746

CADENCE: Predicting Realized MAPF Execution Time Beyond Sum of Costs

Abhishek Seetharamu, Badrikanath Praharaj, Sreeram MV, Mohan Prakash Reddy Vakati

原文摘要

Multi-Agent Path Finding (MAPF) algorithms are increasingly used to plan motion for robot teams in industrial warehouses and robotic shared workspaces, but standard MAPF algorithm evaluation metrics, such as Sum of Costs (SoC), makespan, and planner runtime, can obscure how planner choices translate into realistic execution performance. We present CADENCE (Coordination and Action-Driven Estimation for Networked Continuous Execution), a hardware study of this evaluation gap on a fixed 7 by 7 workcell with seven differential drive robots, asking which features available before execution can best predict final wall-clock completion time. We compare SoC, total planned travel cost, primitive motion burden (how much basic motion the plan requires, such as makespan, turns, consecutive moves, and start-stop transitions), and interaction aware coordination structure (how much inter-robot coordination the plan induces, such as dependency links, interacting robot pairs, dependency depth, and crowding exposure). To test this, we generate 120 plans across 15 scenarios -- 5 Empty, 5 Medium Random, and 5 Bottleneck and execute each plan four times, yielding a 480 trial hardware corpus. Using both a scenario-held -- out ridge model and a trial-level mixed-effects model, we find that SoC alone is informative but incomplete, while primitive motion burden gives the strongest improvement, reducing held out error by about 48.6%-59.8% in MAE and 44.2%-61.4% in RMSE relative to SoC-only models. Interaction-aware coordination features add smaller, less uniform gains, most clearly in the mixed-effects analysis. Across both models and uncertainty checks, primitive motion burden is the most reliable additional signal beyond SoC, suggesting that much of the execution time gap is already visible in the offline plan before any robot starts moving.

cs.RO

版本更新 · v2arXiv:2606.04569

MineXplore: An Open-Source Reinforcement Learning Exploration Benchmark for GNSS-Denied Underground Environment

Abhishek Seetharamu, Badrikanath Praharaj, Sreeram MV

原文摘要

Underground mines present extreme conditions for autonomous robot navigation: GPS is denied, lighting is degraded, and tunnel topology is loop-rich and non-convex. Simulation benchmarks grounded in real production-mine geometry and compatible with GPU-accelerated learning pipelines do not yet exist in the open-source ecosystem. We present MineXplore, an open-source MuJoCo-based navigation benchmark derived from the Leung et al. 2017 Chilean underground copper mine dataset. The environment reconstructs a 104,423 sq.m tunnel network through an six-stage contour-to-MJCF pipeline incorporating octagonal wall cross-sections, LiDAR-sourced jagged wall geometry, three terrain friction zones, a global 5 degree incline, and periodic spot lighting. Geometric fidelity is validated at an Intersection over Union (IoU) of 0.9538 against the source survey map, and surface texture similarity scores 79.4% across six structural dimensions. A single-agent PPO baseline trained via RLlib across five independent random seeds achieves a best rolling coverage of 88.89% (3 of 5 seeds reaching the 90% coverage target), confirming that MineXplore supports stable and reproducible policy learning under realistic underground sensing and topology.

cs.RO

新论文 · v1arXiv:2611.12271

Public, Identifiable, and Effectively Immutable: Ethical Lessons from Research on arXiv

Jan Pennekamp, Johannes Lohmöller, Peter Mayer, Arjun Arunasalam

原文摘要

Ethical decision-making is central to safe cybersecurity research. Recent studies, one of which is our own, analyzing the source files of arXiv papers bring this challenge to light - arXiv publicly distributes the source files, to the surprise of many authors. Unique aspects reveal ethical challenges and dilemmas that extend beyond established ethical frameworks, institutional processes, and community best practices. In this work, we examine this case and provide insights for similar research on other public live resources, such as Zenodo or the Wayback Machine. While such empirical studies are particularly realistic, they must account for personally identifiable information, leaked secrets, and other sensitive details that could expose, shame, or otherwise negatively affect individuals, possibly leading to both reputational and practical harm. sDrawing on our collective experience and discussions among researchers, affected authors, and readers, we make three contributions: First, we define the ethical setting and highlight that even intentionally-public data with deliberate real-name attribution can raise ethical concerns when repurposed for empirical analysis. Second, we discuss the ethical issues surrounding the use of arXiv data and derive four higher-level dilemmas that researchers may face when working with similar data. Third, we derive practical guidance for ethically navigating such settings and, in doing so, distill four lessons for research on arXiv and similar public, live datasets: public availability should serve as the starting point of ethical reasoning; research should minimize the amplification it induces on exposed information; disclosure should be calibrated to the harm, confidence, and actionability of a finding; and effective immutability must be accounted for. Together, these contribute to community resources for ethically-sound cybersecurity experiments.

cs.CR

新论文 · v1arXiv:2611.13042

Trust, but verify, audit and attribute: a human-in-the-loop framework for checking AI-produced research results

György Eigner, Patrik P. Süli, Roland Hollós

原文摘要

Artificial intelligence (AI) systems now produce research results in bulk: in October 2026 a single laboratory released more than 700 AI-generated mathematics manuscripts, only part of them with machine-checked proofs. Such volumes will be checked mostly by AI agents that call specialised software. A passing check certifies only the artefact that was checked. For proofs translated into the proof assistant Lean, it has been shown that this artefact may state something other than the author's text, and that deciding whether a translation is faithful is undecidable in general. The human researcher therefore stays in the loop, and human and AI contributions must be recorded. We propose a domain-independent, human-in-the-loop framework for checking AI-produced results. Validated, agent-usable tools with different grounds of trust check the results; a fidelity audit asks whether each check addressed the claim as stated; the human decides at fixed points; and a dialog report attributes the work to the human and to the AI. We describe two automatic implementations of the tool layer, for mathematics (Wolfram Language, MATLAB, Lean) and statistics (Python and R), and run the mathematical one on a two-claim example. An ambiguous claim was split into two readings, one refuted and one proved.

cs.AI · cs.CY · cs.HC · cs.LO

版本更新 · v4arXiv:2607.23111

Leaf-discriminating trees for multiparameter counting of numerical semigroups

Maria Bras-Amorós, Vicenç Torra

原文摘要

A numerical semigroup is a cofinite additive submonoid of ${\mathbb N}$. The largest element of the complement is its {\em Frobenius number} and the cardinality of its complement is the {\em genus}. We propose a new approach to counting semigroups by the Frobenius number and by the genus, by introducing two ad-hoc trees. Those are leaf-discriminating trees in the sense that their leaves correspond exactly to the objects we want to count and, so, exploring these trees instead of general trees is optimal. Our approach reduces the search space for genus to 23\% and for Frobenius to 0.03\% of that required by previous approaches and this allowed to obtain an important number of unkown terms from both sequences. We implemented a recursive descending algorithm based on the so-called seeds structure, trimming the general semigroup tree exactly at those nodes with no descendants with a given genus, in the first case, or with no descendants with a given Frobenius number, in the second case. We refined the parallelizing strategies and we overcame the previous limitation of the length of integers in the bitwise representation of the gap sequence and the seed sequence. We computed multiparameter decomposition of the numbers of semigroups of each Frobenius number up to $135$, the number of irreducible semigroups of Frobenius number up to $135$, and of the numbers of semigroups of each genus up to $80$.

cs.DM · math.AC · math.CO

版本更新 · v6arXiv:2509.21190

Towards Foundation Models for Zero-Shot Time Series Anomaly Detection: Leveraging Synthetic Data and Relative Context Discrepancy

Tian Lan, Hao Duong Le, Jinbo Li, Wenjun He, Meng Wang, Chenghao Liu, Chen Zhang

原文摘要

Time series anomaly detection (TSAD) is a critical task, but developing models that generalize to unseen data in a zero-shot manner remains challenging. Existing foundation models for TSAD often rely on reconstruction-error scoring at inference time, which can miss subtle anomalies that are well reconstructed and can falsely flag complex but normal patterns in unseen domains. We introduce TimeRCD, a foundation model for TSAD built on Relative Context Discrepancy (RCD), a pre-training paradigm that trains the model to detect anomalies by comparing a query pattern with its surrounding context. This relational formulation, implemented with a standard Transformer architecture, enables the model to infer normality from the input context rather than relying on fixed global normal patterns. We further construct a large-scale synthetic corpus with context-dependent anomaly labels to provide supervised pre-training signals for RCD. Experiments across diverse benchmarks show that TimeRCD outperforms existing general-purpose and anomaly-specific foundation models in most zero-shot TSAD settings, while remaining competitive with dataset-specific full-shot baselines. These results provide empirical evidence that RCD is an effective direction for building robust and generalizable TSAD models.

cs.LG · cs.AI

新论文 · v1arXiv:2611.12270

Modelling socio-economic effects of the pandemic in a metropolitan area

Jan E. Snellman, Rafael A. Barrio, Cecilia I. Ventura, Tzipe Govezensky, Kimmo K. Kaski, Maarit J. Korpi-Lagg

原文摘要

One of the most notorious measures taken by authorities in every country trying to prevent the uncontrollable spread of the Covid 19 pandemic was to impose severe measures to diminish the mobility of people. In recent publications we proposed a mathematical model to address the consequences of these measures and to predict the infections under different scenarios, for various different countries. In this paper we extend the model, called the BTH--SEIRS-model, to deal with the reduction of mobility in an urban area. The chosen region was the densely-populated metropolitan area around Helsinki, the capital of Finland. This choice obeyed to the fact that there are precise data related to mobility from a mobile telephone company, and that this region was completely isolated from the rest of the country for a short period during the pandemic. We also used public transport data from railways and the accommodation occupation rates in the area of interest. We show that the BTH--SEIRS model is capable of simulating the real infection data as well as the mobility-related economic activity, exhibiting the main trends present in the three indicators used for comparison. Our calculations allow us to highlight the difference of the behaviour of the pandemic in a totally different spatial scale, and to discover the hidden factors that govern the behaviour of people who are forced to commute every day from home to work.

physics.soc-ph · cs.MA

新论文 · v1arXiv:2611.12269

Agent4Crypto: A Dual-Reflective and Market-Feedback-Aligned LLM-Based Multi-Agent System for Cryptocurrency Trading

Guangyan Gan, Jiahao Li, Jiashun Wang, Lin William Cong

原文摘要

Cryptocurrency markets provide a natural testbed for autonomous AI agents because they are digital-native, high-frequency, and machine-speed environments, yet existing methods often struggle with non-stationary market regimes, opaque decisions, and fixed aggregation rules. In this paper, we propose Agent4Crypto, a dual-reflective and market-feedback-aligned multi-agent system. The framework coordinates specialized agents for data collection, analysis, risk assessment, reflection, and trading, synthesizing heterogeneous signals from prices, indicators, news, and charts in an interpretable decision pipeline. Our central design is a trader-centric decision mechanism: three style-specific trader agents, Aggressive, Neutral, and Conservative, generate candidate actions and natural-language rationales from the same market context, while their influence is learned rather than manually fixed. A PPO-based gating policy dynamically allocates trader weights through a dual-reflective feedback process. One reflection stream evaluates post-trade decision quality from realized portfolio outcomes and distills reusable lessons into a Financial ReasoningBank for retrieval in subsequent trader prompts, while the other estimates trader-level counterfactual contribution, indicating which style has recently aligned better with market movements. The PPO state combines Counterfactual Contribution Memory with market features, previous trader weights, and current trader actions, so trader influence adapts across bull, sideways, and bear regimes. Experiments on BTC, ETH, and XRP show that Agent4Crypto improves total return and Sharpe ratio over traditional strategies, machine learning and reinforcement learning baselines, and existing LLM-based trading systems, demonstrating the value of interpretable trader competition and market-feedback-driven adaptation in volatile cryptocurrency markets.

cs.MA · cs.CL

新论文 · v1arXiv:2611.12268

Odds-Structured Distributional Regression for Interpretable Time-to-Event Prediction

Jakob Versnjak, Manuel Welte, Titus Kuehne, Marcus Kelm, Grégoire Montavon

原文摘要

Time-to-event prediction is fundamentally challenged by partial observability and aleatoric uncertainty. In clinical medicine, for instance, the timing of disease progression or acute readmission is contingent on highly dynamic and largely latent health trajectories. Rather than forcing a deterministic point forecast, distributional regression offers a principled means to address this complexity by predicting a probability distribution over time. However, explaining these distributional predictions presents a key obstacle: raw distribution parameters output by the model are abstract mathematical quantities that do not align well with domain concepts. To gain interpretability, we propose a novel paradigm where output parameters are designed to be additive in log-odds---a widely established measure in clinical risk modeling. Our approach, called ODDS-DR, moves beyond technical convenience to deliver clear domain alignment as well as the layer-wise computational structure required for robust, time-dependent feature attribution. Validated across two real-world clinical cohorts, our framework matches state-of-the-art predictive performance while resolving distributional outputs into interpretable risk markers across flexible prognostic horizons.

cs.LG · stat.ML

版本更新 · v3arXiv:2610.20108

Exhaustive Canonical Skeletons Generation: A Formal Dataset for Decomposable Symbolic Regression

Giorgio Morales, Anikó Ekárt

原文摘要

Neural symbolic regression (SR) models require diverse datasets of mathematical expressions for robust training. However, existing datasets rely predominantly on stochastic sampling methods, which suffer from high structural redundancy and incomplete functional coverage. To address this, we introduce an exhaustive dataset of univariate canonical skeletons generated via a deterministic constrained context-free grammar. A canonical skeleton acts as a generic structural representation that subsumes more specific functional forms, eliminating redundant expressions and avoiding repeated computational efforts. Although our dataset consists of univariate skeletons, it is designed to support the decomposable SR paradigm to solve multivariate SR problems without assuming linear separability. Furthermore, we pair this dataset with a sampling strategy that generates clean numerical data without triggering numerical instabilities or relying on artificial protected operators. Canonicalizing two existing neural SR datasets reveals that our dataset is substantially more compact, eliminates structural repetitions, and achieves a uniform operator distribution free of simplicity bias. Finally, we establish a benchmark of 150 SR problems across three tiers, on which state-of-the-art SR methods are shown to struggle significantly.

cs.LG

版本更新 · v3arXiv:2504.02421

Balanced spanning forests of graphs: Hardness and solving methods

Morteza Davari, Phablo F. S. Moura, Hande Yaman

原文摘要

The balanced connected $k$-partition problem (\textsc{bcp}) is a classic problem, which consists in partitioning the set of vertices of a vertex-weighted connected graph into a collection of~$k$ classes such that each class induces a connected subgraph of \emph{roughly} the same weight. In this study, we investigate edge-weighted variants of $\textsc{bcp}$, where, given a connected graph $G$, a positive integer $k$, and edge weights, the goal is to compute a spanning $k$-forest~$\mathcal{T}$ of $G$ (i.e., a forest with exactly $k$ trees) that minimizes the weight of the heaviest tree in~$\mathcal{T}$ in the min-max version, or maximizes the weight of the lightest tree in~$\mathcal{T}$ in the max-min version. We show that both versions of this problem are $\NP$-hard on complete graphs with $k=2$, unweighted split graphs, and unweighted bipartite graphs with $k\geq 2$ fixed. Moreover, we prove that these problems do not admit subexponential-time algorithms, unless the Exponential-Time Hypothesis fails. We focus on the min-max version and devise a tight $k$-approximation algorithm, compact and non-compact integer {linear programming} formulations, branch-and-cut, and branch-and-price algorithms. Finally, we present the outcomes of an experimental study on the performance of different solution methods. The source code of the complete implementation of the proposed algorithms is also available.

cs.DS · cs.DM

版本更新 · v3arXiv:2510.09034

Convergence of optimizers implies eigenvalues filtering at equilibrium

Jerome Bolte, Quoc-Tung Le, Edouard Pauwels

原文摘要

Ample empirical evidence in deep neural network training suggests that a variety of optimizers tend to find nearly global optima. In this article, we adopt the reversed perspective that convergence to an arbitrary point is assumed rather than proven, focusing on the consequences of this assumption. From this viewpoint, in line with recent advances on the edge-of-stability phenomenon, we argue that different optimizers effectively act as eigenvalue filters determined by their hyperparameters. Specifically, the standard gradient descent method inherently avoids the sharpest minima, whereas Sharpness-Aware Minimization (SAM) algorithms go even further by actively favoring wider basins. Inspired by these insights, we propose two novel algorithms that exhibit enhanced eigenvalue filtering, effectively promoting wider minima. Our theoretical analysis leverages a generalized Hadamard--Perron stable manifold theorem and applies to general semialgebraic $C^2$ functions, without requiring additional non-degeneracy conditions or global Lipschitz bound assumptions. We support our conclusions with numerical experiments on feed-forward neural networks.

cs.LG · math.DS · math.OC

新论文 · v1arXiv:2611.12261

Hierarchical Multi-Channel Tip-Over Prediction via Isolated Temporal Models for Planetary Rover Safety

Siddhant Shete, Raul Dominguez, Frank Kirchner

原文摘要

Reliable tip-over prevention is critical for planetary rovers operating in unstructured environments where communication delays prevent real-time human intervention. Existing learning-based approaches typically optimize a single prediction objective, limiting their ability to capture both gradual instability evolution and abrupt terrain disturbances. This work presents a hierarchical multi-channel tip-over prediction framework based on isolated temporal learning. A multi-horizon Temporal Convolutional Network (TCN) learns progressive instability patterns from inertial measurements, while a Gated Recurrent Unit (GRU)-based forward dynamics model predicts nominal rover motion and detects unexpected dynamic deviations through prediction residuals. These complementary signals are integrated through decision-level fusion and priority-based arbitration to generate graded safety actions, including early warning and emergency braking. The proposed system is deployed as an Orocos Real-Time Toolkit (RTT)-based safety component within the Rock robotics framework, operating at 100 Hz on a planetary rover software stack. Experiments on Inertial Measurement Unit (IMU) samples collected across heterogeneous terrains demonstrate 97.58% event prediction coverage with a 1.82s median lead time, outperforming individual prediction channels. The results validate that isolated temporal models combined through hierarchical decision fusion provide a robust solution for real-time rover safety in uncertain terrain.

cs.RO · cs.LG

新论文 · v1arXiv:2611.12259

A forensic evaluation framework for egocentric foundation models: an EFM3D case study for crime scene reconstruction

Davidson Chisom Onwubiko, Shahana Bano, Carlos Francisco Moreno-García

原文摘要

Foundation models increasingly combine geometry, semantics, and spatial context for three-dimensional scene understanding, but their suitability for forensic crime scene reconstruction remains unclear. This study presents a structured forensic evaluation framework for egocentric foundation models and demonstrates it using the Egocentric Foundation Model for 3D (EFM3D). Four simulated indoor crime scenes were recorded with Meta Project Aria Gen~1 at Robert Gordon University's East Lodge House forensic training facility. The framework evaluates room-scale reconstruction, semantic scene understanding, computational performance, and forensic suitability. Across the four scenes, EFM3D generated meshes averaging $12{,}668\pm2{,}430$ vertices and $24{,}838\pm4{,}903$ faces, with 39,529 recorded semantic predictions spanning 21--24 categories at a mean confidence of 0.40. Because reference meshes and expert-labelled 3D ground truth were unavailable, these statistics are treated as descriptive outputs rather than measures of geometric or detection accuracy. Final inference runs processed $1{,}388\pm203$ snippets per scene at $0.69\pm0.01$ samples/s and required $50.77\pm6.12$ minutes per recording. A seven-criterion capability profile indicated stronger support for room geometry, furniture localisation, scene context, and spatial relationships than for fine-scale forensic evidence. The results position EFM3D as a complementary tool for contextual documentation and scene review rather than a replacement for high-fidelity forensic capture. The framework provides a reproducible basis for evaluating future egocentric foundation models in specialised forensic settings.

cs.CV

新论文 · v1arXiv:2611.12252

Lessons Not Learned: Why Cross-Chain Bridges Keep Losing Billions the Same Way

Mohsen Ahmadvand, Piotr Dobaczewski, Camille Bouzerand

原文摘要

Bridge hacks have cost billions, often without exploiting a code bug. In the cross-chain bridge hacks we catalogue from 2021 to 2026, failures of security posture, meaning how a bridge's people, code, controls and off-chain infrastructure are set up, account for 88% of reported losses since 2023, and nearly all repeat earlier failures. Code audits cannot catch these failures, so we propose a bridge security assessment rubric. It scores each bridge deployment from on-chain evidence, and a hard cap drops the score to the lowest value when a fatal risk is present. A token is only as safe as its weakest deployment if all its deployments share one backing. For attacks nobody anticipated, we show that rate limits and an emergency stop still bound the loss. We build an analyzer that applies the rubric to deployed bridges and monitors them for changes, and an agentic analyzer that runs the same checks on any bridge. Operators learn what to fix, and users can compare bridges. Our rubric's controls cut bridge hack losses 105-fold, from \$3.52B to \$33M.

cs.CR · cs.DC

版本更新 · v2arXiv:2606.08833

CSFlow: Aligning Flow Matching with Human Contrast Sensitivity

Malgorzata Galinska, Bart Pogodzinski, Jan Eric Lenssen

原文摘要

We introduce Contrast Sensitive Flow (CSFlow), a weighting scheme that connects the human eye's Contrast Sensitivity Function (CSF) to the iterative denoising steps of flow matching. Because real-world images concentrate signal at low spatial frequencies, these components reach high signal-to-noise ratio earlier during continuous diffusion than high-frequency components. When generating images with diffusion or flow matching models, this induces a soft autoregressive structure in Fourier space, where coarse image content stabilizes before fine detail. Meanwhile, the human visual system is unequally sensitive to spatial frequencies: very low and very high frequencies require significantly higher contrast to be perceived. We for the first time merge these observations through two contributions: (1) a metric that estimates which frequencies are generated at each reverse flow interval and (2) timestep weights obtained by aligning the frequencies generated at each noise level with human contrast sensitivity. We validate our contributions experimentally showing that these weights can improve generative performance by lowering FID by up to 12.15%, increasing Inception Score by up to 3.2% and improving GenEval scores by up to 3.0% using inference-only timestep modification or short fine-tuning. Qualitatively, we find that our CSFlow weights lead to better visual realism and less cartoonish appearance of generated images. In a two-alternative forced-choice study, participants preferred CSFlow generations over the baseline in 65% of comparisons. Additional visualisations, code, models and the example CSFlow weights are available on our project website.

cs.CV

新论文 · v1arXiv:2611.12251

RAIL: Representation-Aligned Imitation Learning for Student-Compatible Teacher Policies

Meraj Mammadov, Pedro Zuidberg Dos Martires, Johannes Andreas Stork

原文摘要

Reinforcement learning (RL) from raw sensory inputs such as images or onboard sensors can be challenging due to high-dimensional observations and sparse rewards. A common strategy is to train a teacher policy with access to privileged state information and then distill it into a student policy that acts from raw inputs alone. However, when the teacher relies on information unavailable to the student, exact imitation may be impossible, creating an irreducible imitation gap. Existing approaches typically mitigate this mismatch through careful reward shaping or additional reinforcement learning on the student policy after the teacher has been trained. Instead, we introduce Representation-Aligned Imitation Learning (RAIL), which addresses the mismatch during teacher training itself. RAIL learns a latent representation shared across teacher and student observations using contrastive learning, and trains the teacher policy directly in this space. Because the teacher policy is constrained to operate on this shared space, it learns behaviors that are reproducible from student observations. This substantially reduces the imitation gap while preserving task performance. Across multiple environments, RAIL outperforms strong baselines without reward modification or post-hoc student fine-tuning, and surpasses direct reinforcement learning from raw observations. The learned representations further enable zero-shot transfer to new tasks through teacher-only training. Project site: https://merajmammadov.com/rail/.

cs.LG · cs.AI · cs.RO

新论文 · v1arXiv:2611.13041

Code-Grounded Agentic Explanations for Static Analysis Vulnerability Findings

Oshando Johnson, Ranjith Krishnamurthy, Rohith Kumar Shanmuganathan, Goran Piskachev, Eric Bodden

原文摘要

Static application security testing (SAST) tools surface vulnerabilities early and their verdicts are rule-based and auditable. Yet developers routinely dismiss findings they cannot interpret. Large language models can supply the missing explanation, but designs that also hand the model the detection give up the auditability that makes static analysis defensible. We keep detection with the analyser and give the model only the explaining. A preliminary user study with 20 participants and a workshop with 19 practitioners yielded five requirements. Guided by these we redesigned SAFE, an open-source IntelliJ IDEA plugin that explains findings from any analyser reporting in the Static Analysis Results Interchange Format. It tiers explanations by demand, ties them to the reported data flow, and replaces the single-prompt backend with an agent that must consult the repository and the rule that fired before answering. A within-subjects crossover experiment with 24 participants compared the redesign against its predecessor. Perceived usability rose from 77.3 to 85.0 on the System Usability Scale. Actionability gained most among the explanation-quality items, the aspect both earlier studies flagged as weakest, and the odds of a correct remediation were 2.6 times higher. Triage accuracy and time per finding were indistinguishable, which we read as uninformative rather than as equivalence. Trust was the one measure that did not move, and the closing interviews indicate that grounding made findings easier to understand rather than easier to believe. Participants also drew a boundary no requirement anticipated, wanting the tool to find and explain while they judge and apply.

cs.SE · cs.AI · cs.HC

新论文 · v1arXiv:2611.12248

A method to automatically discover symbolic local learning rules

Andrea Perin, Fabio Anselmi, Stephane Deny

原文摘要

We present a general method for discovering symbolic local learning rules that can replace backpropagation, applicable, in principle, to any deep learning architecture and task. The method automatically discovers local loss functions specific to each layer by (1) constructing a dictionary of monomials of local tensor variables, inspired by Penrose diagram notation, (2) regressing over coefficients weighting these monomials to build layerwise polynomial losses whose gradients together minimize a target global loss. We test the method on classification, reconstruction, and self-supervised learning tasks with small multilayer networks and minimal datasets. Across all tasks, the discovered rules induce non-trivial interactions between layers, which lead to significant reductions in the corresponding global losses. Furthermore, the method selects sparse combinations of monomials, finding local rules that are more interpretable than those generated by MLP-based alternatives. The potential implications of our method range from energy-efficient deep learning to the discovery of biologically plausible learning mechanisms.

cs.LG · cs.NE · stat.ML

版本更新 · v2arXiv:2603.25780

A Judge Agent Closes the Reliability Gap in AI-Generated Scientific Simulation

Chengshuai Yang

原文摘要

Large language models can generate scientific simulation code, but the generated code silently fails on most non-textbook problems. We show that classical mathematical validation -- well-posedness, convergence, and error certification -- can be fully automated by a Judge Agent, reducing the silent-failure rate from 42% to 1.5% across 134 test cases spanning 12 scientific domains. The headline result comes from a prospective benchmark: 72 blinded tasks submitted by 12 independent scientists yield an 89% success rate (95% CI: [80%, 95%]) with automated error bounds, versus 53% without the Judge. On clinical CT (the only powered experiment, n = 200), the pipeline reaches 99% of expert quality. The residual 1.5% concentrates at bifurcation points where certifiability breaks down. We formalize this boundary through the simulability class S and introduce spec.md, a structured specification format that makes any scientific computation problem machine-readable and solver-independent. Code, data, and all 72 benchmark tasks are publicly archived.

cs.SE · cs.LG

新论文 · v1arXiv:2611.12247

Fundamental Outage and Diversity Limits of Fluid Antenna Multiple Access Under Correlated Nakagami-$m$ Fading

Anastasios Papazafeiropoulos

原文摘要

Existing analytical studies of fluid antenna multiple access (FAMA) are largely confined to Rayleigh fading, independent-port assumptions, or asymptotic selection metrics, leaving the fundamental outage behavior under realistic fading and spatial correlation conditions uncharacterized. This paper fills this gap by presenting the first exact and asymptotically tight outage analysis of interference-limited FAMA systems under spatially correlated Nakagami-$m$ fading. By explicitly modeling the joint statistics of the desired and interference powers across fluid-antenna ports, \textcolor{black}{we first formulate the outage probability in terms of the joint desired-signal and interference distributions and then derive an explicit integral expression under a tractable single-reference-port spatial correlation model.} Building on this result, we establish a rigorous diversity-order characterization revealing that the achievable diversity is jointly dictated by the fading severity and the spatial correlation rank imposed by the finite antenna aperture, rather than by the nominal number of ports. To enable accurate and tractable performance evaluation beyond the asymptotic regime, we further introduce a Gamma-matched outage approximation that preserves both the exact diversity order and the array-gain constant. The resulting closed-form expression provides an accurate characterization over a wide range of signal-to-interference ratio (SIR) thresholds and offers clear physical insight into the role of spatial correlation in FAMA systems. Extensive Monte Carlo (MC) simulations confirm the tightness of the exact analysis, the sharpness of the asymptotic results, and the accuracy of the proposed approximation.

cs.IT

版本更新 · v5arXiv:2405.15647

A Logic of Knowledge and Justifications, with an Application to Computational Trust

Francesco A. Genco

原文摘要

We present a logical framework that enables us to define a formal theory of computational trust in which this notion is analysed in terms of epistemic attitudes towards the possible objects of trust and in relation to existing evidence in favour of the trustworthiness of these objects. The framework is based on a quantified epistemic and justification logic featuring a non-standard handling of identities. Thus, the theory is able to account for the hyperintensional nature of computational trust. We present a proof system and a frame semantics for the logic, we prove soundness and completeness results and we introduce the syntactical machinery required to define a theory of trust.

cs.LO

版本更新 · v4arXiv:2607.17469

Local Search with Correlated Randomness

Yunbei Xu

原文摘要

The running-time distribution of a randomized algorithm is computed under independent randomness, but independence is rarely what a source guarantees. We study sources $ν$ with $ν[w]\le DP[w]^s$ for every finite prefix $w$ of the tape, where $P$ is an independent reference law, $0<s\le1$, and $D\ge1$; individual choices may be predictable, even determined by the past. The worst-case probability of any closed tape event $E$ is exactly $\min\{1,DC_s(E)\}$, where $C_s(E)$ is the infimum cost of covering $E$ by prefixes of varying depth. This tree max-flow/min-cut identity makes worst-case probabilities exact and optimizable. Three applications follow. In scheduling, $n$ tasks on one machine are each retried until the next random symbol lies in its success set; every schedule has the same makespan distribution under independent randomness, yet under our constraint two fixed orders miss a linear deadline with worst-case probabilities $1$ and $\exp[-Θ(n)]$, and a static priority rule is optimal at every deadline. For the standard local walk on a $k$-CNF with an assignment satisfying a majority of each clause, we find the sharp source threshold $s_*$: below it, on some instances one admissible source forces nontermination for every clause selection; at it, with fixed $D$, the expected number of flips is at most cubic in the initial distance, against linear under independence, with matching instances at distances up to a constant fraction of the number of variables; a fixed margin above it restores linear dependence. For noisy predecessor search with prefix-bounded comparison errors, verified backtracking attains any prescribed error probability at or above a source threshold whose sharpness we leave open, with the same cubic-to-linear query bounds. Restarting on the shared tape restores exponential tails exactly when the nontermination event has a prefix cover of cost below one.

cs.CC · cond-mat.stat-mech · cs.LG · math.CO · math.PR

新论文 · v1arXiv:2611.13040

On the Concavity of Rényi Entropy and Rényi Entropy Power along Heat Flows on Manifolds

Hao Wu, Lei Yu

原文摘要

In this paper, we establish the concavity of Rényi entropy and Rényi entropy power along heat flow on Euclidean space of arbitrary dimension, for Rényi orders $α\in[1,2]$. More generally, we prove the corresponding result along the heat flow generated by the Witten Laplacian on complete Riemannian manifolds under a suitable curvature-dimension condition. Our argument builds on novel functional inequalities that were recently developed for the study of Tsallis entropy concavity along the heat flow. We extend both these inequalities and the resulting concavity theorem from the Euclidean setting to the Riemannian manifold setting. In addition, we establish the complete monotonicity of Tsallis entropy of order $2$ along the heat flow and derive its spectral decomposition.

math.DG · cs.IT · math.PR

版本更新 · v2arXiv:2611.07998

Towards Multiscale Molecular Representation Learning through Geometry-Based Fragmentation for Polymer Property Prediction

Makoto Takamoto, Chaitanya Sampara, Viktor Zaverkin, Daniel Oñoro-Rubio, Mathias Niepert

原文摘要

It is well established that molecular substructures can contribute unequally to chemical and physical properties, with specific motifs or side chains often playing dominant roles. Fragment-based representations have therefore emerged as an effective strategy for incorporating chemically meaningful inductive biases, as exemplified by molecular fingerprints and fragment-based attention mechanisms. However, many existing fragmentation-based machine learning methods rely on predefined chemical rules, which may yield suboptimal fragmentation for previously unseen or chemically unfamiliar structures and often overlook the importance of multiscale molecular representations. To address these limitations, we formulate molecular fragmentation as a radius-constrained optimization problem inspired by the k-center formulation, where the number of clusters is not fixed a priori but emerges from the specified radius, and propose a geometry-based fragmentation method that partitions arbitrary molecular or material structures using largely geometric information, without relying on predefined chemical fragmentation rules. The method provides flexible control over fragment size, enabling the construction of multiscale fragment representations for scale-adaptive modeling. Building on this framework, we develop a geometric fragment-based attention mechanism that integrates multiscale structural information and incorporate it into a multimodal machine learning model combining p-SMILES representations with three-dimensional (3D) conformational information. Experiments on a benchmark polymer dataset show that incorporating 3D conformational information improves prediction accuracy for several properties over models based exclusively on p-SMILES representations. Explainable artificial intelligence analyses further provide insights into the physical relevance of the learned representations.

cs.LG · cond-mat.mtrl-sci · physics.chem-ph

新论文 · v1arXiv:2611.12243

CADMA: Capacity-Aware Coverage Decomposition for Generative Model Assessment

Debin Meng, Zheng Gao, Ziquan Liu, Yanran Li, Ioannis Patras, Georgios Tzimiropoulos

原文摘要

Evaluating generative models requires measuring not only sample fidelity but also how well a generated set covers the real data distribution. Recent work has therefore introduced recall-oriented metrics for distribution-level evaluation. However, existing recall metrics often reduce distributional coverage to geometry-based coverage. This simplifying assumption weakens their diagnostic value and interpretability in two ways: these metrics provide only a single aggregate score that cannot distinguish essential diversity failures such as mode dropping and mode collapse, and geometry-based coverage can provide unreliable estimates by assigning high scores to poorly covered generative distributions. We therefore introduce CADMA, a capacity-aware framework for recall-oriented evaluation and diagnosis. CADMA evaluates distributional coverage through local capacity matching between real and generated samples, separating geometry-based coverage from effective coverage. This yields multi-dimensional coverage diagnostics that quantify mode dropping and mode collapse while reducing the overly optimistic estimates produced by geometry-based metrics. Through extensive experiments on toy data, real-world datasets, and generative models, we show that CADMA provides more reliable coverage estimates and more fine-grained, interpretable diagnostic signals of diversity failures than existing metrics. Our code is available at https://github.com/Open-Debin/CADMA_NeurIPS2026.

cs.CV · cs.LG

版本更新 · v3arXiv:2606.24389

BRAVR: An AP-Assisted Online DRL Mechanism for Interactive VR Bitrate Adaptation over Wi-Fi

Miguel Casasnovas, Francesc Wilhelmi, Boris Bellalta

原文摘要

Interactive virtual reality (VR) streaming over Wi-Fi requires stringent latency and reliability guarantees, which become increasingly difficult to achieve under dynamic channel conditions and shared-medium contention. These challenges make real-time bitrate adaptation a critical yet fundamentally difficult control problem, particularly under limited visibility of network conditions. This paper formulates VR bitrate adaptation as a network-aware, online decision-making problem and proposes BRAVR, a decentralized deep reinforcement learning (DRL) algorithm that combines application-layer observations with lightweight wireless statistics collected at the Wi-Fi access point (AP) to balance visual quality, streaming performance, and multi-user airtime fairness. We implement BRAVR in a real VR streaming system and evaluate it on a physical Wi-Fi testbed against the VR-tailored NeSt-VR algorithm and an ablated BRAVR variant without AP telemetry. Experimental results show that, in the evaluated scenarios, BRAVR consistently achieves its design objectives, delivering robust quality of service (QoS) while avoiding sustained airtime overutilization. It also outperforms the AP-blind ablation, demonstrating the value of incorporating network-level information into the bitrate adaptation control loop. Overall, these results demonstrate the potential of AP-assisted online learning for decentralized interactive VR streaming over commodity Wi-Fi and provide practical insights into bitrate adaptation in shared wireless environments.

cs.NI

新论文 · v1arXiv:2611.12242

Controlling Temporal Pseudo-Label Marginals for Stable Online Test-Time Adaptation

Rani Amar, Ofir Lindenbaum

原文摘要

Continual Test-Time Adaptation (CTTA) methods update models online using a stream of unlabeled data to overcome distribution shifts encountered at test time. However, CTTA is often unstable, as it can reinforce biases in early predictions and increasingly predict only a few classes, resulting in drift or even collapse. In this paper, we identify narrowing pseudo-label class usage as an indicator of CTTA instability and quantify this behavior. Most existing methods have attempted to address this issue by either selecting reliable adaptation samples or re-weighting the loss to encourage diverse pseudo-labels, but they do not explicitly enforce a class marginal over a temporal window. Instead, we propose Sinkhorn Class-Marginal Regularization (SCALAR), a simple plug-and-play module that controls the pseudo-label distribution over a short temporal window. SCALAR works by maintaining recent predictions in a buffer, balancing their class distribution via the Sinkhorn algorithm to match a specified target distribution, and using the balanced soft targets for adaptation. Without changing the model architecture, SCALAR maintains the per-step runtime and increases the peak allocated GPU memory by only 10 MB on ImageNet-C. Our method is particularly effective on skewed, non-stationary data with severe distribution shifts, where SCALAR significantly alleviates drift and collapse when combined with existing stabilization methods based on entropy or diversity.

cs.LG · cs.AI · cs.CV

新论文 · v1arXiv:2611.12241

SNAPRAG: Cost-Efficient Knowledge Extraction Attack on RAG Systems via Adversarial Perturbations

Melih Coşğun, Mehmet Emre Gürsoy, Alptekin Küpçü, Sinem Sav

原文摘要

Retrieval-Augmented Generation (RAG) systems improve Large Language Model (LLM) responses by incorporating information from external knowledge bases. However, recent studies have shown that attackers can extract substantial portions of private knowledge bases from RAG systems through repeated querying. Yet, state-of-the-art extraction attacks either rely heavily on external LLMs to generate adversarial queries or require a large number of queries. In this paper, we propose SNAPRAG, a novel and cost-efficient attack for extracting RAG knowledge bases through adversarial perturbations. Our approach algorithmically generates adversarial queries without relying on external LLMs and allows for a strictly configurable query budget. In particular, SNAPRAG uses a discrete token-mutation algorithm to generate novel queries in a strict black-box setting. Our adaptive dissimilarity threshold calibration systematically forces queries into unexplored semantic regions of the target database, enabling extraction with fewer queries. Specifically, to extract half of the target knowledge base, our method requires up to 6.77$\times$ fewer queries than existing approaches. Furthermore, within a fixed query budget, our method recovers up to 96.9%, outperforming the closest approach by 7.7-19.9%. We further evaluate SNAPRAG against active defenses, where it remains the leading attack on four of five defended targets, achieving 50.7-96.3% extraction rate.

cs.CR

新论文 · v1arXiv:2611.12240

Prune, Don't Merge: Expert Compression of Mixture-of-Experts LLMs on Multilingual Production Tasks

Tomas Rasymas, Povilas Kvedaras

原文摘要

Mixture-of-Experts language models are cheap to run per token but expensive to hold in GPU memory, which drives the cost of serving them in production. One-shot expert compression shrinks them after a single calibration pass. Its two current state-of-the-art methods are REAP for expert pruning, which deletes the least salient experts, and REAM for expert merging, which merges them into the most salient ones. Both have been evaluated only on academic benchmarks. We compare the two on five multilingual production tasks of a web-hosting company, from classification to the generation of marketing HTML email templates, with two 30B-parameter models (Qwen3-30B-A3B-Instruct-2507 and the hybrid Mamba-attention Nemotron-3.5-Lightning) at 25% and 50% of experts removed and the same calibration data. Pruning wins: on the same rows, REAP is significantly better in 16 of 20 comparisons and REAM in 2. At 25%, REAP keeps 93.6% of Qwen3-30B's generative quality and REAM 62.5%, although on benchmarks the two are within a point for the same model. The damage concentrates in the output contract rather than the content: 89.8-95.2% of each significant loss on email templates comes from hard-gate failures such as invalid JSON, broken mobile layouts and runaway generation, and most failed non-English domain descriptions copy the prompt's English opening phrase. Cross-perplexity and logit divergence, two diagnostics from the REAP paper, also track production quality, as does the rate of repetition loops; n-gram diversity does not. We close with practical guidance for compressing MoE models for production.

cs.CL · cs.LG