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

+758新增论文
+578其中 AI 相关
+94已有论文修订
+95新增 AI 导读

arXiv 最近完成扫描:2026-11-17 04:58 · 全站最近入库:2026-11-17 04:58各信源进度 →

统计口径与数据新鲜度

统计 2026-11-16 05:34 — 2026-11-17 05:34(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:2026-11-17 04:58最近发现新增版本:2026-11-17 04:58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。

24,106 已收录论文20,158 AI 相关7,049 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 24,106 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-17 04:57(北京时间)
新论文 · v1arXiv:2611.10474

Do links to retracted papers predict future retractions? An out-of-sample test on the open biomedical citation graph

Hossein Mohasel Arjomandi

原文摘要

Tools that screen manuscripts for research-integrity problems flag papers that cite retracted work, and retrospective studies report that such papers are retracted more often. This study tests the signal prospectively on open data. Predictors were computed on the January 2024 OpenCitations graph using the retractions known at that date, frozen, and compared with retractions in the Retraction Watch database from February 2024 to October 2026. The design was pre-specified. Among 10.5 million biomedical papers published 2015-2023, 2.1% cited a paper already retracted, and 4,323 were retracted later (0.041%). Citing a retracted paper raised the later retraction rate from 0.035% to 0.27% with one such reference; excluding reviews (post hoc), it rose to 2.3% with five or more. Within journal and publication year the odds ratio was 2.99 (95% CI 2.73-3.27), and 3.03 when reference-list length was also held fixed (post hoc), so the signal is not explained by retraction waves in journals or by long reference lists. As a screen it is weak: 0.9% of the 1,000 highest-ranked papers were retracted later (2.2% with reviews set aside, post hoc), and 84% of later retractions cite no retracted paper. When the time point was moved into the late-2023 retraction wave, the journal's own retraction rate became the far better screen. NIH-funded papers cited already-retracted work as often as other papers in the same journal and year. Links to retracted papers are a useful flag but not a screen on their own.

cs.SI

版本更新 · v3arXiv:2602.09969

Causal Multi-Task Demand Learning

Varun Gupta, Vijay Kamble

原文摘要

We study multi-task demand learning for retail pricing, where firms seek to estimate heterogeneous linear price-response functions across many decision contexts. Each context has rich covariates but limited price variation, motivating transfer across tasks. However, such transfer is complicated by endogeneity: prices may be arbitrarily correlated with unobserved task-level demand determinants. We propose a meta-learning framework that identifies the conditional mean of task-specific causal demand parameters despite this confounding, provided each task contains at least two distinct, locally exogenous price points. The key is to carefully design the information available to the learner. We condition on the full price history to address cross-task confounding, but withhold the demand outcomes associated with two locally exogenous price points to preserve identifiability. These withheld outcomes provide supervision through a randomized query. We establish that this information design is maximally uniformly valid for the squared-loss objective: any refinement of the conditioning set that reveals information about the withheld outcomes or the query index admits a data-generating process under which the objective no longer identifies the conditional mean causal target. Synthetic experiments demonstrate improved estimation of causal demand parameters under strong confounding, while retail experiments show improved held-out demand prediction relative to standard transfer-learning baselines.

cs.LG · econ.EM · stat.ML

版本更新 · v4arXiv:2610.25113

A learner-free swarm approximates the explore-exploit equilibrium allocation: A self-organized attractor in place of reinforcement-learning machinery

Galen J. Wilkerson

原文摘要

Reinforcement learning (RL) solves the explore-exploit problem by training a learner on reward; its multi-agent, resource-sharing form has an equilibrium allocation as its optimum -- the ideal-free distribution (IFD), the Nash equilibrium of a congestion game. We show a minimal swarm approximates it with no learner at all. We present a decentralized, learner-free method for this problem: agents explore by heavy-tailed (Levy) flights, recruit neighbours to a discovery through a degree-normalized threshold cascade, and disperse as a patch crowds. The method drives the collective to the shape of the equilibrium allocation (pooled spatial matching slope 1.14, R^2=0.61) with no training, no stored parameters, and no reward in its dynamics; it reaches this as a self-organized attractor and holds it, re-finding it when the resources move -- though it reaches the shape, not the exact equilibrium, and stays measurably exploitable (Nash-gap 1.46). Coupling recruitment to movement causes the allocation: removing it collapses the matching (slope 0.006). Against Proximal Policy Optimization (PPO) learners (feedforward, centralized-critic, graph message-passing, and recurrent) trained in the same sensing-limited environment, the method at zero training is the only arm left unbiased w.r.t. the IFD gradient (slope ~1, CI containing 1, vs learners 0.69-0.97, all under-concentrating) and ties the best learner's harvest where resource is clustered, while the learner collects more where resource is scarce (0.33 vs 0.22) and holds a tighter per-capita intake. Learning buys precision and transfer across reward structures, not a better allocation shape. A companion paper develops the mechanism.

cs.LG · cs.AI · cs.SI · math-ph · nlin.AO · physics.bio-ph

新论文 · v1arXiv:2611.10472

Group-Conditional Conformal Risk Control: Exact Guarantees, Limits, and Overlapping Extensions

Tareq Aldirawi, Wenge Guo

原文摘要

We study group-conditional conformal risk control (CRC) for disjoint and overlapping groups under bounded monotone losses. For disjoint groups, we establish exact finite-sample conditional risk control, two-sided bounds on the group-restricted risk, and fundamental limitations arising from local support and calibration-rank granularity. For overlapping groups, we study two exactly valid approaches: group-restricted calibration followed by union deployment, and direct calibration of the deployed prediction-set family. We characterize the union's limiting reconciliation gap, identify conditions for its disappearance, and establish sharp lower bounds for structured losses in specified product-form models. We also demonstrate a strict size advantage for direct calibration in an ordinal model. Experiments show that direct calibration can reduce prediction-set size, but its efficiency depends on group structure, loss type, and shape fitting. Results on historical diabetic-retinopathy scores further illustrate these tradeoffs. Overall, local support, reconciliation slack, and threshold heterogeneity determine efficiency, with neither approach uniformly dominating.

stat.ML · cs.LG · math.ST

新论文 · v1arXiv:2611.10471

Can open citation data detect citation stacking? An audit against Clarivate's journal suppressions

Hossein Mohasel Arjomandi

原文摘要

Clarivate suppresses journals from the Journal Citation Reports when self-citations or concentrated citations from another journal ("citation stacking") distort their Journal Impact Factor. Its data are closed. Yearly journal citation networks were built from the OpenCitations snapshot of January 2024 (1.96 billion citations; 85,437 journals) and compared with Clarivate's suppression lists for data years 2015-2023, collected from primary sources. Where reference lists are present, open data reproduce Clarivate's published self-citation shares within ten points for 36 of 50 journals. The failures follow publishers that do not deposit references with Crossref, which also hides several stacking donors. Biodiversity data downloads registered with DataCite inflate naive impact-factor numerators of taxonomy journals. The leading open method, CIDRE, detects none of 15 named Clarivate pairs: it drops large donor journals, and lowering its donor threshold merges the pairs into large groups. A recipient-centric reading of CIDRE's own test flags stacking-listed journals 6 to 14 times more often than other journals, depending on consensus level, and 5 to 7 times more often than size-matched controls, sometimes years before the listing. At equal flag rate it finds more listed journals than a plain 43% donor-share rule (10 vs 3 of 28), and about as many as a rule with a citation floor. A prospective check on the 2024-2025 lists agrees in direction but is not significant (one of six). Flagged pairs show more citations per citing paper and weaker topical similarity than matched pairs. Open audits are feasible if coverage gaps are reported.

cs.SI

新论文 · v1arXiv:2611.10469

Hibana: Fast Agentic Inference on Unified-Memory Devices

Yahya Emara, Chi-Chih Chang, Anthony Fei, Ahmed F. AbouElhamayed, Chong-Yan Chen, Hung-Yueh Chiang, Kai-Chiang Wu, Mohamed S. Abdelfattah

原文摘要

Local AI promises capable personal assistants that run on users' own devices, keeping data and computation under their control. Open-weight models that fit on local devices can now handle practical tasks such as coding. Yet current inference engines fall short. Popular engines such as vLLM and SGLang target datacenter GPUs with dedicated host memory, while local engines underuse the memory hierarchy of unified-memory devices. We present Hibana, an inference engine for unified-memory devices that co-optimizes model representation, memory management, and decoding strategy. Model surgery, the process of editing a trained model's architecture to suit its hardware, changes how the model computes, shrinking its memory footprint and data movement. Unified-memory management preserves each agent's state across turns, so long-running agents can resume instantly instead of recomputing their state. Adaptive speculation generates several tokens per step. Hibana codesigns these features around the shared memory capacity and limited bandwidth of unified-memory systems-on-chip to support local multi-agent workloads. On a single DGX Spark serving Qwen and Nemotron models of up to 125B parameters, Hibana completes up to 3.4$\times$ more agentic tasks per hour and delivers up to 3.9$\times$ higher decoding throughput than other established frameworks such as vLLM, NVIDIA Flare, llama.cpp, and TensorFold. Code is available at https://github.com/sharar-project/hibana.

cs.CL · cs.AI

版本更新 · v3arXiv:2605.15491

Ghosted Layers: Unconstrained Activation Alignment for Recovering Layer-Pruned LLMs

Vincent-Daniel Yun, Junhyuk Jo, Sai Praneeth Karimireddy, Sunwoo Lee

原文摘要

Layer pruning removes entire Transformer decoder blocks from large language models, but introduces a mismatch between the hidden state received by the next surviving layer and the distribution it was trained to process, leading to significant performance degradation. We propose Ghosted Layers, a training-free recovery module that addresses this issue by solving a boundary activation alignment problem. Our method derives a closed-form optimal linear operator from a small calibration set to reconstruct the activation discrepancy introduced by the pruned layers. We show that this solution corresponds to the unconstrained optimum of the alignment objective, whereas LinearPatch is restricted to a fixed basis symmetric operator family. Across multiple LLM backbones, pruning strategies, and pruning depths, Ghosted Layers outperforms prior training-free recovery methods in accuracy and perplexity in most settings while preserving the efficiency benefits of layer pruning.

cs.LG · cs.AI · cs.PF

新论文 · v1arXiv:2611.10468

Dynamic State Estimation from Sparse PMUs via Physics-Informed Graph Neural Operators

Huy Hoang Le, Haoguang Wang, Christian Moya, Marcos Netto, Guang Lin

原文摘要

This paper proposes a physics-informed graph neural operator for power system dynamic state estimation. From a sparse, intermittent measurement window at a few generator buses, the operator estimates the rotor-angle and speed trajectories of all machines, unmeasured ones included, in one forward pass. A graph-attention encoder processes the measurements, and the swing equations are imposed as a hard constraint or as a soft one with a learned correction that absorbs model errors. It is trained on simulated measurements, without state labels. With several phasor measurement units, masked training hides one from the input but keeps it in the loss. On four transmission systems, simulations show that the operator is the most accurate estimator that neither optimizes per scenario nor uses state labels, except on the smallest system at large model errors, where tuned Kalman smoothers are better. Networks trained on simulated states are more accurate on the two smallest systems but not on the 118-bus system, where the operator is faster and more accurate than hard-constrained moving-horizon estimation. Under model errors, it matches or outperforms an optimization-based benchmark at a fraction of its cost. The results hold for a detailed machine model with exact parameters seen only through voltage phasors.

cs.LG · eess.SY

新论文 · v1arXiv:2611.10467

Who Benefits, and When? Trajectories of Academic Achievement and Interest in AI-Supported Learning

Hyejeong Lee

原文摘要

AI-supported learning environments are increasingly used, yet relatively little is known about how students' achievement and interest develop over time or whether these trajectories vary across learners. This study examined longitudinal trajectories of mathematics achievement and interest in a semester-long AI-supported mathematics learning. Using a mixed-methods longitudinal design, repeated measures of mathematics achievement were collected at four time points and mathematics interest at three time points. Linear mixed models (LMMs) were used to examine overall changes over time, while linear quantile mixed models (LQMMs) were used to examine differences in trajectories across the lower, median, and upper portions of the outcome distributions. Mathematics achievement increased substantially during the intervention, with the largest gains emerging between the second and third measurement occasions. However, the trajectory differed across the achievement distribution. Mathematics interest increased significantly after the first month, but this increase was not sustained through the end of the semester. Qualitative findings further indicated that students experienced the AI platform differently. The study highlights the importance of examining both cognitive and motivational trajectories.

cs.HC

新论文 · v1arXiv:2611.10466

The Fast Tree Transform

Paul Burchard

原文摘要

The Fast Tree Transform is a locally-aggregated transform defined for arbitrary graph geometries that generalizes both the Hadamard frequency transform and the Haar wavelet transform on the circle. The pure frequency part of the transform approximately diagonalizes the graph Laplacian; this fact makes it useful as a preconditioner in geometric optimization problems. The full transform provides an only slightly-redundant representation of signals on the graph that allows sparse representation of signals that include both edges and textures.

cs.LG · math.DG · math.SP

版本更新 · v2arXiv:2610.32069

DriveMind: Mind-Evolving Belief Tracking for Closed-Loop Autonomous Driving

Yan Shi, Kehua Chen, Jiarui Qi, Yihong Tang, Haicheng Liao, Yinhai Wang

原文摘要

Recent multimodal large language models (MLLMs) can perceive a traffic scene, reason about it, and propose a driving action within a single model. However, existing MLLM-based drivers reason from scratch at each frame and commit to a single explanation, discarding earlier hypotheses about the scene. When the scene is ambiguous, the agent locks in on the wrong interpretation and acts on it before evidence accumulates. In this paper, we propose DriveMind, an MLLM-based driving framework that keeps several hypotheses about the scene alive at the same time and updates their weights as new frames arrive. Inspired by particle filtering, DriveMind propagates each hypothesis through a VLM, scores it against the current observation, and resamples to maintain diversity. On Bench2Drive, adding belief tracking to the same driving pipeline improves the Driving Score (DS) from 68.50 to 81.37 and the Success Rate (SR) from 41.36% to 54.09%. The largest gains occur on ambiguous scenarios such as unsignalized intersections (+20.0 DS) and oncoming-traffic interactions (+20.5 DS). Beyond the overall metrics, DriveMind also improves fine-grained interactive driving abilities such as merging, emergency braking, and give-way.

cs.AI

新论文 · v1arXiv:2611.10464

Beyond Discrete Ancestry Groups: Distribution-Aware In-Context Learning for Cancer Outcome Prediction Across the Genetic Ancestry Continuum

Weizhi Li, Yan Cui

原文摘要

Machine learning models for omics-based prediction of cancer outcomes are commonly built on aggregated data without distinguishing ancestry groups, or under an ancestry-discrete design, which partitions samples into ancestry groups and fits a separate model within each group. Recent evidence shows that this approach is inherently limited: predictive accuracy declines with genetic distance from the training samples, even within a single ancestry group, because the data distribution shifts continuously rather than only between predefined categories. In this paper, we propose distribution-aware in-context learning (DA-ICL) for ancestry-continuum modeling. Our method extends conventional ancestry-discrete modeling to ancestry-continuum modeling, capturing the nonstationary omics-endpoint relationship across the genetic continuum while enabling efficient modeling through an in-context learning foundation model. We provide theoretical analysis demonstrating the superiority of the ancestry-continuum over ancestry-discrete modeling, along with empirical results supporting the proposed framework. Source code is available at https://github.com/ai4pm/DA-ICL-Continuum.

stat.ML · cs.AI · cs.LG

新论文 · v1arXiv:2611.10463

LongCapHarness: Capability-Guided Auto-Harnessing for Long-Context Reasoning

Tao Zhang, Yuhang He

原文摘要

Existing methods enhancing an LLM's long-context reasoning capability often rely on data synthesis and follow-up LLM post-training. However, the data synthesis strategy and the post-training recipes are empirically driven and computationally intensive. In this work, we propose a new framework, dubbed LongCapHarness, to instead harness the underlying capabilities for long-context reasoning. The harnessing process recursively revises and updates capability primitive implementation, capability state and control logic through recursive self-improvement on the development set. With neither data synthesis nor LLM post-training, LongCapHarness is applied to Llama3-8B and transferred to Qwen2.5-3B. LongCapHarness substantially outperforms recent data-synthesis-based baselines on HELMET and RULER benchmarks; its estimated development cost is less than 50% of the estimated training costs of NExtLong and PolicyLong. We hope these results motivate further research into harnessing the latent capabilities of LLMs to solve tasks beyond their explicit training.

cs.CL

新论文 · v1arXiv:2611.10461

Zeroth-Order Langevin Dynamics for Stochastic Nonconvex Optimization: Curvature Regularization and Non-Asymptotic Guarantees

Stefano Bruno, Dong-Young Lim

原文摘要

Zeroth-order methods provide a way to optimize stochastic nonconvex objectives using only noisy function evaluations. However, the smoothing inherent in their gradient estimators effectively modifies the objective being optimized. In particular, Gaussian smoothing with smoothing parameter $η>0$ induces, to leading order, an implicit Hessian-trace regularization, which motivates the study of the curvature-regularized objective of the form $f(x)+(η^2/2) tr(\nabla^2 f(x))$. We develop a non-asymptotic theory in the nonconvex setting for a class of algorithms called stochastic zeroth-order Langevin dynamics (ZOLD), which use only noisy function evaluations and do not have access to first- or higher-order derivative information, and derive excess-risk guarantees for the curvature-regularized objective. We first control the finite-time distribution of the ZOLD iterates in Wasserstein distance with respect to the Gibbs measure associated with the Gaussian-smoothed objective. An explicit coupling between the smoothing and inverse temperature parameters then quantifies the proximity of this measure to the Gibbs measure associated with the corresponding Hessian-trace-regularized objective. These distributional estimates in Wasserstein distance lead to excess-risk bounds for the curvature-regularized objective. Our analysis accommodates assumptions under which the stochastic function evaluations are allowed to have non-uniform dependence on the data stream. In addition, the constants in the non-asymptotic bounds do not rely on log-Sobolev or Poincaré inequalities as in prior zeroth-order Langevin analyses. To the best of our knowledge, this is the first non-asymptotic analysis of a zeroth-order Langevin method that simultaneously establishes excess-risk guarantees for a curvature-regularized objective and Wasserstein guarantees for the associated Gibbs measure in the nonconvex setting.

math.OC · cs.LG · math.PR · stat.ML

版本更新 · v4arXiv:2606.00995

Subliminal Learning Is Steering Vector Distillation

Camila Blank, Agam Bhatia, Senthooran Rajamanoharan, Arthur Conmy, Neel Nanda

原文摘要

Subliminal learning refers to a student language model acquiring a teacher's traits (e.g. a system-prompted preference for owls) when fine-tuned on the teacher's outputs, despite the outputs being semantically unrelated to those traits. It remains poorly understood how data without semantic meaning can transfer specific semantic traits. In this work, we show that subliminal learning is mediated by a single steering vector, i.e. a vector added to the model's activations. Across two open-source models, we find that the teacher's system prompt is well approximated by a steering vector, and that the student's behavior is driven by learning an aligned vector over fine-tuning. System prompts that are not well approximated by steering vectors are not subliminally learned. This is a special case of steering vector distillation, in which a student trained on the outputs of a steered teacher learns to imitate that steering. We demonstrate steering vector distillation on a range of semantic and random vectors. Adding a semantic vector to a model's activations can have both model-independent and model-specific (i.e. non-semantic) effects on its behavior, so generated data that is non-semantic can transmit a vector with semantic effects, enabling subliminal learning. This also explains why subliminal learning does not transfer between models. We find that adaptive optimizers are necessary for subliminal learning in language models: activation gradients on steered data carry a small but consistent component along the steering direction, and non-adaptive optimizers impede this by allowing outlier gradients to dominate.

cs.AI

版本更新 · v2arXiv:2603.12873

TRACE: Structure-Aware Character Encoding for Robust and Generalizable Document Watermarking

Jiale Meng, Jie Zhang, Runyi Hu, Zhe-Ming Lu, Tianwei Zhang, Yiming Li

原文摘要

We propose TRACE, a structure-aware framework leveraging diffusion models for localized character encoding to embed data. Unlike existing methods that rely on edge features or pre-defined codebooks, TRACE exploits character structures that provide inherent resistance to noise interference due to their stability and unified representation across diverse characters. Our framework comprises three key components: (1) adaptive diffusion initialization that automatically identifies handle points, target points, and editing regions through specialized algorithms including movement probability estimator (MPE), target point estimation (TPE) and mask drawing model (MDM), (2) guided diffusion encoding for precise movement of selected point, and (3) masked region replacement with a specialized loss function to minimize feature alterations after the diffusion process. Comprehensive experiments demonstrate \name{}'s superior performance over state-of-the-art methods, achieving more than 5 dB improvement in PSNR and 5\% higher extraction accuracy following cross-media transmission. \name{} achieves broad generalizability across multiple languages and fonts, making it particularly suitable for practical document security applications.

cs.CV

版本更新 · v3arXiv:2601.15295

Elsewise: Authoring Open-ended Interactive Narrative with Possibility Space Visualization

Yi Wang, John Joon Young Chung, Melissa Roemmele, Yuqian Sun, Tiffany Wang, Shm Garanganao Almeda, Brett A. Halperin, Yuwen Lu 等 9 位作者

原文摘要

Interactive narrative (IN) authors craft spaces of divergent narrative possibilities for players to explore, with the player's input determining which narrative possibilities they actually experience. Generative AI can enable new forms of IN by improvisationally expanding on pre-authored content in response to open-ended player input. However, this extrapolation risks widening the gap between author-envisioned and player-experienced stories, potentially limiting the strength of plot progression and the communication of the author's narrative intent. To bridge the gap, we introduce Elsewise: an authoring tool for LLM-based INs that implements a novel Bundled Storyline concept to enhance author's perception and understanding of the narrative possibility space, allowing authors to explore similarities and differences between possible playthroughs of their IN in terms of open-ended, user-configurable narrative dimensions. A user study (n=12) shows that our approach improves author anticipation of player-experienced narrative, leading to more effective control and exploration of the narrative possibility spaces.

cs.HC · cs.AI · cs.CL

版本更新 · v2arXiv:2606.29955

SpreadsheetBench 2: Evaluating Agents on End-to-End Business Spreadsheet Workflows

Jian Zhu, Yuzheng Zhang, Zeyao Ma, Bohan Zhang, Armin Schoepf, Daniel Woloch, Peter Yiliu Wang, Guangyu Robert Yang 等 14 位作者

原文摘要

Spreadsheets are widely used for business analysis, financial modeling, reporting, and decision-making. However, most existing spreadsheet benchmarks evaluate isolated operations such as single-formula generation or local cell edits, and therefore fail to capture end-to-end workflows in realistic business settings. We introduce \textsc{SpreadsheetBench 2}, a workflow-level benchmark for spreadsheet agents that covers three task categories: generation, debugging, and visualization. The benchmark is constructed from authentic business data, including financial reports and corporate filings, and is annotated and validated by domain experts. The benchmark contains 321 tasks; each instance averages 11.8 worksheets and requires 593.5 cell modifications, reflecting large multi-sheet workbooks with cross-sheet dependencies. We evaluate eight frontier large language models under a unified multi-turn agent scaffold, and additionally include several LLM-based spreadsheet products as complementary baselines. Results show that current systems remain far from reliable on real-world workflows: the best model achieves 34.89\% overall task accuracy, and debugging accuracy is as low as 12.00\%. Trajectory analysis and a failure taxonomy further indicate that insufficient spreadsheet inspection and incorrect target-cell selection are the dominant bottlenecks. Together, these findings position \textsc{SpreadsheetBench 2} as a challenging testbed for advancing reliable spreadsheet automation. Project page: https://spreadsheetbench.github.io/

cs.SE · cs.AI

版本更新 · v2arXiv:2602.21215

Inference-time Alignment via Sparse Junction Steering

Runyi Hu, Jie Zhang, Shiqian Zhao, Jiale Meng, Jiwei Li, Jason Zeng, Ming Wu, Michael Heinrich 等 10 位作者

原文摘要

Token-level steering has emerged as a pivotal approach for inference-time alignment, enabling fine-grained control over large language models (LLMs) by modulating their output distributions without parameter updates. While effective, existing methods rely on dense intervention at every decoding step. This persistent manipulation not only incurs substantial computational overhead but also risks compromising generation quality by excessively drifting from the model's intrinsic distribution. In this work, we show that dense intervention is unnecessary and propose Sparse Inference-time Alignment (SIA), which performs sparse junction steering, intervening only at critical decision points along the generation trajectory. Our key insight is that high-entropy junctions are empirically consistent with pivotal decision points in the generation trajectory and are particularly susceptible to misalignment, suggesting that these points benefit from alignment-related reward signal. Extensive experiments across different model families and alignment objectives show that steering only 20%-80% of tokens achieves the superior alignment-efficiency trade-offs. For strong base models such as Qwen3, intervening on as few as 20% of tokens matches or even surpasses heavily post-trained instruct models. This sparsity enables stronger guidance while better preserving the model's native distribution, integrates seamlessly with search-based methods (e.g., Best-of-N), and improves efficiency by reducing the required sampling budget while yielding practical wall-clock speedups under matched-quality comparisons.

cs.CL · cs.AI

版本更新 · v2arXiv:2606.00813

Non-Monotonic Safety Alignment Across Gemma Generations: A Quality-Diversity Red-Teaming Study

Subhadip Mitra

原文摘要

Note on v2 (Nov 2026): attack success now requires a response judged non-refusing and harmful (v1 counted every non-refusal), so all rates are lower; v1's figures are reported as the non-refusal rate, and with 3 seeds no significance claims are made. Safety alignment in LLMs need not improve monotonically across model generations. Studying four generations of Google's Gemma family (7B-31B) with quality-diversity evolution (MAP-Elites) as an automated red-teaming probe, we find that Gemma 3 (12B) has the highest attack success rate (ASR; response judged non-refusing and harmful): 33.7% +/- 1.3% (mean +/- std, 3 seeds), against 19.0% +/- 10.6% for Gemma 2 and 12.3% +/- 4.2% for Gemma 4. Every Gemma 3 seed exceeds every Gemma 2 and Gemma 4 seed, which we report as directional evidence. Harmful successes rarely transfer between generations (1.4-7.9% ASR when archives are replayed on other generations), lowest on Gemma 4 (1.4-2.8%). Cybercrime is the most exploited category in Gemma 1-3 (47-83%) but falls to 2% in Gemma 4, while misinformation rises from 17% (Gemma 2) to 80% (Gemma 3) and is still 53% in Gemma 4. A second-judge audit (Llama-3.1-70B) confirms 74% of the 8B judge's harmful labels. The non-refusal rate shows the same ordering at 2-3x higher levels and overstates harmful transfer by 6-16x. Code and artifacts: https://github.com/bassrehab/red-queen

cs.CR · cs.CL · cs.ET · cs.LG · cs.NE

新论文 · v1arXiv:2611.10457

Represented but Not Used: Value-Policy Dissociation in LLM Agents

Yuzhang Luo, Mohamed Aghzal, Liangming Pan, Ziyu Yao

原文摘要

Large language model (LLM) agents need to choose actions whose consequences unfold over multiple future steps, yet poor decisions do not reveal whether the model failed to represent those consequences or failed to use them. Mechanistic interpretability has studied internal representations of environment state and plans, but sequential decision making also requires evaluating how candidate actions change future outcomes. Reinforcement learning formalizes this as action value. We study this computation in Qwen2.5-7B-Instruct, Qwen2.5-32B-Instruct, and GPT-OSS-20B across a partially observed Grid environment and WebShop, a simulated web shopping environment. We find that the relative values of candidate actions are linearly decodable from intermediate hidden states, including at states where the policy prefers a worse action. We call this gap between value representation and policy choice value-policy dissociation. To investigate how represented value relates to action selection, we compare direct interventions along the value readout direction with Natural Value Transport (NVT), a transformation learned from hidden state changes that naturally result from differences in action value. While readout-direction interventions have limited and inconsistent effects, NVT more consistently influences action preferences, particularly at later layers. We also test a closed-loop controller that uses the value readout as a sensor and NVT as an actuator, observing improvements in task outcomes. These results suggest that LLM agents can represent useful prospective value without reliably translating it into policy choice.

cs.AI

版本更新 · v2arXiv:2609.36056

GeoWind2Plan: Mission-Time 3D Urban Wind Prediction for Energy-Efficient UAV Planning

Shaoxiang Qin, Yucheng Zhao, Fuyuan Lyu, Di Zhou, Jiachen Yao, Xue Liu, Anima Anandkumar, Liangzhu Leon Wang 等 9 位作者

原文摘要

In urban low-altitude flight, buildings reshape ambient wind into spatially varying 3D flow, making unmanned aerial vehicle (UAV) energy depend on local wind exposure as well as path length. However, building-resolved wind is rarely available when a mission must be planned. Computational fluid dynamics (CFD) produces high-fidelity urban flow fields, but each simulation is tied to a fixed inflow boundary condition and can take hours to days, which is incompatible with urban UAV missions that typically last minutes to tens of minutes. We present GeoWind2Plan, a geometry-to-wind-to-planning framework for mission-time 3D urban wind prediction and energy-efficient UAV planning. Given only a background wind vector, 3D building geometry, and a start-goal pair, GeoWind2Plan transforms the building geometry into a reference-wind frame, predicts mission-relevant 3D wind patches with a localized geometry-conditioned neural operator, stitches them into a local wind field, and optimizes a feasible 3D path and speed profile using a physically grounded UAV energy model. Rather than pursuing CFD-perfect reconstruction, GeoWind2Plan targets decision-useful wind prediction: trajectories are planned with predicted wind and evaluated under high-fidelity CFD wind. Across held-out urban domains, wind speeds, and mission wind-angle regimes, GeoWind2Plan performs corridor-localized wind inference in about 3 s, compared with roughly 8 h for CFD. Under CFD evaluation, trajectories planned with GeoWind2Plan reduce energy by 6.9%, 12.7%, and 4.5% in tailwind, headwind, and crosswind missions relative to wind-agnostic planning, recovering 87.9%, 85.7%, and 75.0% of CFD-reference savings. These results show that fast, corridor-localized 3D urban wind prediction can make wind-aware UAV energy planning practical at mission time. Code/dataset: https://github.com/DUAL-Xiao/GeoWind2Plan/ . Demo: https://geowind2plan.com .

cs.AI · cs.RO

版本更新 · v2arXiv:2606.29441

Closing the Activation-Cone Blind Spot: Response-Time Probing and Unified Defense

Subhadip Mitra

原文摘要

Note on v2 (Nov 2026): errors found in the evaluation code mean the empirical results reported in v1 are not supported; the theoretical corollary is unaffected. Details are in the notice on page 1 and in KNOWN_ISSUES.md in the linked repository. We compare five inference-time safety paradigms (no defense, static steering, CAST, AlphaSteer, probe-gated steering) across seven instruction-tuned models (7-31B) and five attack types (GCG, AutoDAN, DeepInception, prefilling, intent laundering). We derive a corollary: any defense that gates intervention on a single layer's activation alignment with a benign reference (cone, subspace, or null-space) is blind to attacks that craft activations to lie inside that reference, whether checked at prompt time or per token. For prefilling attacks we propose response-time probing, a linear probe on the hidden state at the first generated tokens combined with a halt, and compose this halt with AlphaSteer's null-space steering. We also examine whether MMLU captures steering's utility cost and how the diversity of negative training sets affects probe false-positive rates. Code, attacks, and per-sample results are released.

cs.CR · cs.AI · cs.CL · cs.ET · cs.LG

新论文 · v1arXiv:2611.10455

Bayesian Low-Rank Posteriors for Scalable Membership Inference

Khanh Nguyen, Antonio Emanuele Cinà, Andrey Barsky, Alejandra Reinares, Fabio Roli, Dimosthenis Karatzas

原文摘要

Membership inference attacks (MIAs) aim to determine whether a sample was used during the training of a target model. Strong MIAs rely on reference (or shadow) models to estimate target-conditional score distributions, but training such ensembles is computationally expensive for modern large-scale Vision--Language Models (VLMs). In this work, we propose a scalable alternative that replaces explicit reference-model training with a Bayesian approximation over low-rank adaptation parameters. Specifically, we model uncertainty only within the LoRA subspace and construct a posterior from a single training run, from which we sample a diverse set of \emph{virtual reference models}. We instantiate this approach using Stochastic Weight Averaging Gaussian (SWAG), enabling efficient approximation of target-conditional score distributions at a fraction of the cost of conventional shadow-model ensembles. We evaluate the resulting attack on three downstream-adapted VLMs across two privacy-sensitive visual question answering tasks, namely Document VQA and Medical VQA. To isolate the effect of pretraining knowledge, we additionally introduce a controlled synthetic dataset. Our approach outperforms standard reference-based attacks while requiring only a single reference-model fine-tuning run, making membership inference more practical for large VLMs.

cs.LG

新论文 · v1arXiv:2611.10454

Decentralised Drone Swarm Formation under Silent Failures with ACO Pheromone Claims and PSO Tuning

Md. Abu Horaira

原文摘要

We study decentralised drone formation, without a central planner, under the conditions that make it hard: nobody knows how many drones are alive, failures are silent, radio is short-ranged and lossy, and actuation is noisy. In our controller, ant colony optimisation (ACO) allocates drones to picture positions through claims that behave as reinforced, evaporating pheromone and are gossiped between neighbours, so evaporation doubles as failure detection. A particle swarm optimisation (PSO) law flies the drones, a noise-buffered Voronoi filter keeps them apart, and PSO also tunes the controller offline. We prove that claims of destroyed drones vanish within bounded time, that conflicts between drones that exchange tables are resolved consistently, that survivors cover the picture to within a factor of two, and that drones cannot collide while their safety cells are feasible. On held-out seeds and with two tuned configurations, the controller formed the picture in 95-99% of runs in the hardest regime, restored it after a silent loss of half the swarm in 96-100% of runs at normal difficulty, transferred unchanged to eight unseen shapes and to 300 drones, and overlapped in 5 of 21,088 runs. A CBBA-style baseline that shares the evaporating claims but uses argmax without swaps was at least as reliable as ours once tuned with the same budget, and repaired faster, whereas ours generalised better to unseen shapes. The benefit thus comes from the pheromone dynamics and from tuning, not from ACO's stochastic construction or PSO's social terms. A centralised oracle is faster.

cs.AI

版本更新 · v2arXiv:2609.04134

Prospective Coding Improves Learning in Deep Continuous-Time Recurrent Networks

Shivang Rawat, Mirko Morello, Flaviano Morone, David J. Heeger

原文摘要

Temporal integration gives continuous-time recurrent networks memory, but in deep stacks it also delays bottom-up signals and attenuates top-down errors. We compensate for this delay by making each layer's bottom-up input prospective, and we develop Recursive Quadrature Filters (RQFs), biologically motivated temporal band-pass filters that are a special case of diagonal state-space models (SSMs). Starting from an energy model, we derive the RQF dynamics and show that each RQF is a band-pass filter whose learnable parameters control its tuning frequency and bandwidth. We then make each layer's bottom-up input prospective using a parameter-free two-tap update that leaves the recurrent transition and parallel scan unchanged. We extend this correction to general diagonal SSMs and show that it mitigates depth-dependent gradient attenuation when temporal gradients are truncated, i.e., spatial-only backpropagation. We evaluate the performance of RQFs, S5, and ORGaNICs (a nonlinear gated RNN with built-in attentional modulation and recurrent normalization) trained using full backpropagation through time (BPTT) and spatial-only backpropagation. Under full BPTT, the prospective variants of these models match or outperform their non-prospective controls in every configuration. A non-residual width-32 six-layer RQF achieves 96.09% accuracy on raw-audio Speech Commands with 31.9k parameters; a width-64 six-layer RQF achieves 83.36% on the 16,384-step Path-X task. We also introduce selective RQFs (sRQFs), which make each filter's input gain and, optionally, its bandwidth and tuning frequency input-dependent while retaining the parallel scan. With the input gain alone, the same six-layer model achieves 86.31% on Path-X; with all three input-dependent, a four-layer prospective sRQF solves Selective Copying (99.83%).

cs.LG · cs.NE · q-bio.NC

新论文 · v1arXiv:2611.10451

Grip, Trace, Write: Force-Aware Imitation and Reinforcement Learning for High-DoF Robotic Handwriting

Rongqian Chen, Yiding Chen, Jijie Zhou, Tom Z. Jiahao

原文摘要

Most robotic handwriting rigidly mounts the pen and reduces writing to trajectory tracking. A freely held pen instead couples grasp stability, tip motion, and contact force. We present Grip, Trace, Write, a force-aware simulation environment and learning framework for high-DoF handwriting. A two-branch actor separates pen retention from arm-wrist writing while preserving their closed-loop coupling. Training learns a grasp controller, initializes writing from inverse-kinematics demonstrations by behavior cloning, and improves writing through reinforcement learning with rewards on ink, force, time, and retention. Across diverse tasks, the policy improves completion, precision, force regulation, and efficiency over behavior cloning and inverse kinematics, and generalizes without fine-tuning to unseen English and Chinese content. These results establish force-aware handwriting as a testbed for dexterous tool use.

cs.RO

新论文 · v1arXiv:2611.10450

The Truth Beneath the Tokens: Prompt-Space Recovery of Knowledge Edits in Language Models

Manit Baser, Rachel Yeo Hui Min, Aditya Nawal, Dinil Mon Divakaran, Mohan Gurusamy

原文摘要

Model editing provides an efficient way to update factual knowledge in transformer models. Recent approaches reverse edits by modifying, masking, or decomposing model weights, but make recovery dependent on weight-space interventions. We instead show that pre-edit knowledge can be recovered through input-space interventions alone, without modifying the edited model's weights. We introduce Memento, a prompt-space mechanism for auditing residual knowledge that learns a reusable recovery signal to steer an edited model toward its pre-edit response. Memento supports both continuous recovery prefixes and discrete textual prompts for standard text interfaces. Across three LLMs and five editing techniques, Memento recovers 80.07% of held-out edited facts, achieving a 9.58% relative improvement over baselines, suggesting that editing may suppress the facts without fully eliminating their recoverability. We further investigate whether prompt-space recovery extends beyond model editing. On pristine, unedited LLMs, Memento improves average factual recall by 53.9% relative to standard inference. These results demonstrate the potential of prompt-space recovery for auditing residual knowledge in edited models and investigating the gap between factual knowledge and its accessibility during inference. We will release our source code.

cs.CL · cs.AI

新论文 · v1arXiv:2611.10448

ESTRA: Enabling Extreme Attention Sparsity for Training-Free Video Diffusion Acceleration

Enhuai Liu, Yunke Wang, Yutong Wang, Changming Sun

原文摘要

Video diffusion transformers generate high-quality videos, but attention over long spatiotemporal sequences makes inference computationally expensive. Sparse attention reduces this cost by computing only a subset of query-key interactions exactly, and recent training-free methods further approximate the remaining interactions. These methods, however, degrade quickly at high sparsity, which limits the achievable speedup. We present ESTRA (Extreme Sparsity Through Refined Attention), a training-free acceleration method that preserves generation quality even at extreme sparsity. ESTRA concentrates exact attention computation on a small set of important key-value tokens and approximates the remaining contributions with cluster representatives. To identify these tokens efficiently, it groups similar queries and uses their representatives to score individual keys. A variance-based correction further refines the attention weights of the cluster representatives, improving the accuracy of the approximation. Experiments on Wan2.1-14B and HunyuanVideo-13B show that ESTRA establishes a new quality-efficiency frontier for training-free sparse attention, achieving $3.26\times$ and $4.71\times$ end-to-end speedups at 5% attention density while surpassing Sol-Attn and SVG-EAR in fidelity to dense generation and on most VBench dimensions. Even at 1% density, ESTRA remains more faithful to dense generation than Sol-Attn at 12.5%, while reaching up to $5.39\times$ speedup.

cs.CV

版本更新 · v2arXiv:2605.09121

A Communication-Theoretic Framework for LLM Agents: Cost-Aware Adaptive Reliability

Hamed Omidvar, Vahideh Akhlaghi

原文摘要

Agents built on large language models (LLMs) rely on reliability techniques such as retry, majority voting, and self-consistency developed in parallel rather than within a common analytical framework. We observe that an LLM sampled at temperature $T$ is a discrete stochastic channel $p(y \mid x)$ in the sense of Shannon's coding theory, and use this identity as the entry point for such a framework. Each of these techniques is a special case of one of six classical reliability operators: diversity combining, hybrid retransmission, iterative generator-critic decoding, rateless sampling, structured redundant verification, and difficulty-adaptive routing. Within the framework we give two closed-form results, a noise-variance threshold above which uniform averaging beats quality-weighted averaging and a contractivity criterion for generator-critic refinement, and turn each into a design rule checkable before committing compute. We further introduce a cost-aware semantic-nearest-neighbor router whose single Lagrangian knob traverses the quality-cost frontier without retraining. Across six local and cloud channel configurations, no fixed model-technique-budget choice dominates. On a 300-item hard split of MMLU, GSM8K, and HumanEval, the router occupies the empirical Pareto frontier: at matched quality its normalized cost is $\approx$56% lower than the best fixed technique, and at matched cost it improves quality by $\approx$7% (26% over single-shot decoding). Judge-free grading of 6,900 outputs confirms that the judge tracks correctness at the item level and that the router's advantage over a fixed policy persists at most cost settings; with internal reasoning enabled, the gain of external refinement shrinks but remains significant for our incremental-redundancy operator. These results argue for consolidating reliability techniques into a single tunable layer informed by channel coding.

cs.LG · cs.AI · cs.CL · cs.IT