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
+64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 21:56各信源进度 →

统计口径与数据新鲜度

统计 2026-11-15 23:30 — 2026-11-16 23:30(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 21:56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6,922 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)
新论文 · v1arXiv:2611.02361

Agent Memory as Rate-Limited Filtering: Explicit State Outlasts Long Context Beyond a Reader-Dependent Horizon

Tianhao Guo, Chongyao Guo

原文摘要

Can a language model that reads the full history of an access log say whether a service may act now? On an executable benchmark of grants, withdrawals, reinstatements, expiries, groups and change windows under a known policy, scored on pairs of histories that differ only in the order or one field of the deciding events, open models of 7 to 32 billion parameters fall from 72 to 86 percent after eight or sixteen events to 48 to 55 percent after 256, while reasoning-trained 32B models reading the current state reach 96 and 98 percent. A reasoning model of 120 billion parameters trails the state by four and seven points at 256 and 512 events and by about ten at 1,024, so the length beyond which explicit state outlasts long context grows with the reader. We treat agent memory as recursive filtering. The decision depends on the history only through a state whose size is set by the number of tracked keys. A store holding fewer component records than the state forgets on a time scale set by its capacity, and a reader model predicts its accuracy within 0.02 on average in a registered test. Retaining log lines without regard to time order errs with probability approaching one half, a bound that is attained. A filter in which a model parses each line and an explicit rule applies the updates answers 80 to 89 percent at 256 events and beats relevance retrieval, model-kept ledgers, a reimplementation of StateMem and the Mem0 library. A retriever that selects at query time the lines the queried state depends on matches it at 256 events on the first domain, so what decides is selection by the state's dependencies, which the filter performs when writing, without the query. Applying parsed updates themselves, the models fall to 58 and 59 percent. A frozen replication on a second domain with fresh wording reproduces the horizon.

cs.LG

版本更新 · v5arXiv:2506.20533

Global Convergence of Iteratively Reweighted Least Squares for Robust Subspace Recovery

Gilad Lerman, Kang Li, Tyler Maunu, Teng Zhang

原文摘要

Robust subspace estimation is fundamental to many machine learning and data analysis tasks. Iteratively Reweighted Least Squares (IRLS) is an elegant and empirically effective approach to this problem, yet its theoretical properties remain poorly understood. This paper establishes that, under deterministic conditions, a variant of IRLS with dynamic smoothing regularization converges linearly to the underlying subspace from any initialization. We extend these guarantees to affine subspace estimation, a setting that lacks prior recovery theory. Additionally, we illustrate the practical benefits of IRLS through an application to low-dimensional neural network training. Our results provide the first global convergence guarantees for IRLS in robust subspace recovery and, more broadly, for nonconvex IRLS on a Riemannian manifold.

stat.ML · cs.LG · math.OC

新论文 · v1arXiv:2611.02353

VisCodeBench: a new benchmark dataset for Text-to-Vis reflecting real-world practice

Hy Nguyen, Thomas James Elton, Ella Wilson, Andrew Reeson, Cecile Paris, Josiah Poon, Jonathan K. Kummerfeld

原文摘要

Generating visualisations from natural language requests (text-to-vis) enables users to explore data more efficiently. While LLMs can complete simple visualisation requests, their performance on challenging ones is not well covered by existing benchmarks. We introduce VisCodeBench, an end-to-end text-to-vis benchmark with 2,672 samples from three different domains: educational, scientific, and general audience scenarios with varying levels of complexity. We benchmark three visualisation systems---Codex, Claude Code, and LIDA---paired with a range of open and commercial LLMs across seven metrics covering execution, chart content, and visual quality. While basic execution is near-solved for strong systems, no system comes close to reproducing the reference data, with the best displayed data similarity reaching only $0.81$ and falling to $0.69$ on our hardest subset for this metric. We further find that frontier models separate on chart correctness and efficiency, with ClaudeCode-Opus-4.7 improving chart correctness by around 5 points over Codex-GPT-5 while running roughly 20 seconds faster. Separately, we find that system design matters as much as the underlying model: despite both running on GPT-5, Codex clearly outperforms LIDA, showing that the surrounding prompting, execution, and refinement pipeline can shape outcomes as much as the model itself. Finally, latency remains a practical obstacle, as even the strongest systems take 39-78 seconds to produce results. VisCodeBench is available at \href{https://github.com/SAIPLL/VisCodeBench}{https://github.com/SAIPLL/VisCodeBench} and accessed via a leaderboard; the full dataset is kept private to prevent LLM training use.

cs.CL · cs.HC

新论文 · v1arXiv:2611.02349

APG: Agentic Static-Analysis Infrastructure on LLVM-IR Graph Databases

Zexin Zhong, Yanjun Zhang, Yulei Sui, Xiao Cheng

原文摘要

LLM-based coding agents increasingly rely on external program knowledge, yet access to static-analysis graphs alone does not ensure appropriate operation selection or evidence interpretation. We present APG, an infrastructure for semantics-grounded static program analysis. Built on SVF and LLVM, it decomposes compiler-derived program analyses into atomic capabilities that LLM agents can select, compose, and orchestrate for goal-directed, cross-graph investigation, with query safeguards and evidence traces preserving inspectable results and provenance. APG provides 38 atomic capabilities. On analysis requests arising during benchmark repair runs, 35 achieve first-attempt query-generation success rates above 50%, including 13 at 100%. Evaluation on 812 C/C++ defects shows higher repair success than the evaluated baselines. Compared with Passive Graph Push, APG improves first-iteration repair success by 17.6 and 18.0 percentage points on the two benchmark collections and reduces average LLM input-token consumption by 90.8%. Our tool is available at https://github.com/JasonZhongZexin/APG-ICSE-Demo and its demonstration video at https://youtu.be/vkuExZUjGnw

cs.SE

新论文 · v1arXiv:2611.02348

Which Words a Model Fears: From a Refusal Direction to a Vocabulary-Wide Fear Lexicon

Boyuan Zhang, Mohammad Zandsalimy, Shanu Sushmita

原文摘要

Whether an aligned language model refuses depends in part on which words a prompt contains, regardless of whether those words carry the request's meaning. Harm words shift the hidden state at the generation boundary along a component shared across prompts and words, which crossed differencing extracts as a single direction. In eight models, writing along this direction raises and lowers refusal monotonically with dose, ahead of random and KL-matched directions. Used as a ruler over the full vocabulary, the direction gives every word a shift, from "feared" words at one end to "courage" words at the other; a word's shift is predictable from its own representation, and appending a feared word to a benign prompt is enough to make the model refuse it. Two held-out tests validate the ranking in all eight models: refusal after inserting an unseen word rises with its table shift, and replacing a prompt's most feared word by its most courageous meaning-preserving synonym significantly raises the answer rate on benign and harmful prompts alike, by an amount that follows the shift the replacement actually causes.

cs.CL

新论文 · v1arXiv:2611.02346

Pointwise Score Evaluation from Succinct Boolean Descriptions: SAT Equivalence for Strongly Log-Concave Gaussian Mixtures

Yinzhen Lv

原文摘要

Score-based generative models rely on access to accurate scores, but the cost of computing those scores depends on how the underlying distribution is represented. Can smoothness and strong log-concavity make score evaluation efficient when a distribution is specified by Boolean constraints? I investigate this question by separating geometric encoding from computational extraction. I construct a succinctly specified Gaussian mixture whose score at a fixed query location encodes whether a Boolean formula is satisfiable. The construction has rational component means and uniform strong log-concavity, yet deterministic polynomial-time approximation of the designated score coordinate to constant absolute accuracy is equivalent to polynomial-time solvability of Boolean satisfiability. This equivalence remains valid throughout a neighborhood whose radius is independent of dimension. I further characterize how numerical precision connects bounded witness enumeration to exact counting, and show that independent sample access turns the same target into a dimension-independent mean-estimation problem. An explicit parity construction also separates low-degree approximation from efficient computation. Together, these results show that favorable geometry alone does not resolve the computational cost of extracting scores from compact descriptions. They identify the precise extraction requirement in this formulation while leaving the existence of a general efficient evaluator open.

cs.CC

新论文 · v1arXiv:2611.02345

MBUndo: Sound, Automatic, and Efficient Measurement-Based Uncomputation for Quantum Circuits

Shengyuan Yang, Charles Yuan

原文摘要

Quantum programs that use temporary values, known as \emph{ancilla} qubits, must explicitly \emph{uncompute} each ancilla back to zero to ensure that the program produces correct outputs. Conventional uncomputation uses unitary quantum gates, whereas leading quantum algorithms now leverage \emph{measurement-based uncomputation} (MBU), which replaces costly unitary $T$ gates with cheaper measurements and phase corrections. Current quantum programming frameworks can use measurement to uncompute specific gates such as Toffoli. No automatic transformation, however, accepts an arbitrary circuit whose ancillas are computed and uncomputed via the standard approach and converts the circuit to use MBU\@. Researchers instead apply MBU to individual circuit designs by hand, which limits algorithm development and compiler optimization. Such a transformation has not existed due to two fundamental unknowns: where MBU applies and which gates realize MBU for an arbitrary computation. To solve these, we first present a static analysis that identifies, for each ancilla qubit, each circuit position where measurement is sound --- subsuming basic assumptions that MBU only applies at the end of a circuit. Next, we present circuit transformations that soundly turn unitary uncomputation gates into efficient phase corrections after measurement --- generalizing rewrite rules from prior literature that are either limited to a single gate or not applicable to a general circuit. We present \textsf{MBUndo}, an integrated compiler pass that automatically converts circuits to measurement-based uncomputation. It requires no user annotations other than which qubits are ancillas, the same as needed by existing uncomputation synthesizers, and can enhance their output circuits. We evaluate \textsf{MBUndo} on 22 benchmarks to reveal \emph{T} gate reductions of ${>}36.5\%$ over uncomputation synthesizers.

cs.PL · quant-ph

新论文 · v1arXiv:2611.02344

Efficient Proper Agnostic Learning of Halfspaces

Anindya De, Elchanan Mossel, Shivam Nadimpalli, Joe Neeman

原文摘要

We give a proper membership-query algorithm for agnostically learning halfspaces under the uniform distribution on $\{\pm1\}^n$. Given query access to an arbitrary function $f:\{\pm1\}^n \to \{\pm1\}$, our algorithm runs in time $\mathrm{poly}(n)+2^{\mathrm{poly}(1/\varepsilon)}$ and outputs a halfspace with error at most $\mathrm{OPT}+\varepsilon$, where $\mathrm{OPT}$ is the error of the best-fitting halfspace for $f$. Previously, even with membership queries, the best known running time was $n^{\mathrm{poly}(1/\varepsilon)}$, achieved by the sample-based algorithm of Kalai, Klivans, Mansour, and Servedio (FOCS 2005). Our algorithm is the first agnostic halfspace learner under the uniform distribution on $\{\pm1\}^n$ whose running time is polynomial in $n$ with an exponent independent of $\varepsilon$. Combining our algorithm with a boosting argument yields a proper agnostic learner using only random examples. It achieves the same $\mathrm{OPT}+\varepsilon$ guarantee, with running time matching that of Kalai et al.\ up to an additive term depending only on $\varepsilon$. To our knowledge, this is the first sample-based proper agnostic learner in the Boolean setting with running time $2^{o(n)}$ for every fixed $\varepsilon>0$.

cs.DS · cs.CC · stat.ML

新论文 · v1arXiv:2611.02341

AgiWorld: An Agentic Video World Model for Building Persistent Worlds

Shaoheng Fang, Yiqun Mei, Hanwen Jiang, Xin Sun, Haitian Zheng, Hao Tan, Kalyan Sunkavalli, Georgios Pavlakos 等 10 位作者

原文摘要

Recent interactive video world models generate visually compelling videos from a single image, following user-provided camera motion and text prompts. However, such low-level control demands substantial user effort to realize a high-level intent, and the generated world struggles to remain persistent as the exploration proceeds. In this work, we present AgiWorld, an agentic video world model that explores and builds a persistent world from a single image and a free-form instruction. For high-level planning, our agent AgiWorld-Agent designs the world and the journey through it in a world state and directs the exploration. For low-level realization, our video world model AgiWorld-WM generates the video chunk by chunk and stores it with its geometry in a spatial memory, from which the agent updates its world state. This closed loop keeps the world faithful to the user's intent and keeps the agent's world state consistent with the generated content throughout the exploration. We co-design the two levels around the shared spatial memory and an interface of metric camera trajectories and structured prompts. In AgiWorld-WM, to handle occlusion and accumulated geometric errors in the spatial memory over long rollouts, we propose occlusion-aware history selection, which retrieves past views by surface visibility, and reliability-based memory maintenance, which prefers reliable geometry and earlier observations. AgiWorld-WM achieves state-of-the-art overall performance on world revisiting and exploration benchmarks, and qualitative results and ablations show that AgiWorld builds persistent worlds from free-form instructions over minute-long journeys. Persistence thus arises not from the video model alone but from its closed loop with the agent.

cs.CV · cs.AI

新论文 · v1arXiv:2611.02339

Achievability of CSS codes performance limits over Pauli channels under separate ML decoding with bounded graphical complexity

Gulnihal Mercanoglu, Achilleas Anastasopoulos, S. Sandeep Pradhan

原文摘要

Capacity-approaching quantum error correction faces a fundamental tension: the required parity-check matrices need to become increasingly dense, while practical decoding and syndrome extraction favor sparse, bounded-degree structures. Reconciling these competing requirements remains an important open problem in quantum code design. We construct Calderbank-Shor-Steane (CSS) code ensembles that approach the CSS rate over memoryless Pauli channels under maximum-likelihood decoding while retaining sparse, bounded-degree graphical representations. The construction separates parity-check density from graphical complexity: although capacity-approaching components may require dense parity-check matrices, they are represented by sparse concatenated factor graphs. We first study a construction combining low-density parity-check (LDPC) and low-density generator-matrix (LDGM) component codes and numerically analyze the average weight distributions of the resulting codes. Motivated by these results, we introduce a second construction based on concatenated LDPC and LDGM codes and generalized irregular repeat-accumulate (GIRA) codes. We derive bounds on the average weight distribution of the resulting sum ensemble and use the Miller-Burshtein framework to prove simultaneous reliability of the two CSS decoding components. Consequently, the proposed ensemble achieves rates arbitrarily close to the CSS hashing bound under separate bit- and phase-error correction, while maintaining bounded graphical complexity, thus providing a foundation for future iterative decoders and indirect syndrome-extraction architectures.

cs.IT · quant-ph

新论文 · v1arXiv:2611.02338

MemShift: Separating Detection from Repair in LLM-Derived Personal Memory

Zichuan Xu, Haochen Zhang, Yuyang Dong, Makoto Onizuka, Chuan Xiao

原文摘要

AI assistants store values that a language model derived from a person's records under the person's own definitions, such as their important contacts. Keeping such a value correct takes two steps: detection, noticing that an input changed, and repair, producing the correct new value. Benchmarks that score only the final answer cannot tell which step failed, and they state every change as new text. Seeing this limitation, we present MemShift, a benchmark of five shifts: source revision, deletion, redefinition, writer switch, and export. The first three fork a synthetic world into branches that differ only in the shift. Gold values and dependencies are fixed before any text, and a deterministic scorer replaces an LLM judge. For a memory that reports its invalidations, MemShift shows for each changed item whether the memory noticed the change and whether its repair was right. Moreover, we build a reference memory with known detection and repair to validate the measure. By evaluating four open-weight LLMs, we find that detection helps only as much as repair allows: the LLMs re-derive few values computed by the person's definitions correctly, and programs that the LLM writes over typed records close most of the gap to exact recomputation. Code and data are available at https://github.com/zichuanxu/memshift.

cs.CL · cs.AI · cs.LG

新论文 · v1arXiv:2611.02337

Embedding a de Rham Complex Structure into Eig-PIELM for Maxwell Eigenvalue Problems

Hao Kang, Liang Li, Xing Li

原文摘要

The perfect electric conductor (PEC) boundary condition alone does not eliminate the gradient nullspace of the Maxwell curl-curl operator. Consequently, a boundary-only Maxwell implementation of Eig-PIELM, an eigenvalue extension of the physics-informed extreme learning machine (PIELM), may retain represented gradient modes in its reduced trial space, producing nonphysical, gradient-contaminated modes. We develop dR-Eig-PIELM, a finite-dimensional structure-preserving extension that retains the linear-coefficient, boundary-null-space, and one-shot multimode architecture of Eig-PIELM. It combines compatible tensor-product grad--curl trial spaces with a constrained null-space construction to restrict the sampled PEC space to the mass-orthogonal complement of the represented gradient image before the eigensolve. Symmetric weak-form Gram assembly yields a symmetric-definite reduced eigenproblem without an eigenvalue cutoff under the stated sampling and material-positivity assumptions. For rectangles and cuboids, we establish fixed-degree exactness at the PEC vector-field space and derive rank and dimension identities. Numerical experiments show that dR-Eig-PIELM excludes the spurious eigenmodes with significant gradient components observed in the boundary-only baseline. The method exhibits rapid convergence with increasing polynomial degree and competitive work--precision performance in the tested configurations. A covariantly mapped waveguide example illustrates the potential of dR-Eig-PIELM for integration into design optimization workflows through a constrained local geometry update driven by computed cutoff frequencies and their sensitivities. This work provides a mathematical and computational framework for separating represented gradient modes from physical Maxwell eigenmodes within Eig-PIELM and supports its use in waveguide spectral analysis and geometry design.

math.NA · cs.LG

新论文 · v1arXiv:2611.02333

When Does Adaptive Halting Win? Selection Rules Matter in Looped Language Models

Mahesh Reddy Pagadala

原文摘要

Looped language models can stop iterating early, and recent work shows that post-hoc readouts of the loop trajectory often rival the learned exit gate of the looped model Ouro. Such comparisons reduce each method to one operating point chosen by a selection rule. We ask whether that rule decides the outcome. On a frozen Ouro-1.4B with four loops, we compared a trained stability probe ($π_1$-x) with Ouro's native gate, adapted to a per-question stopping decision, in pre-registered studies; the confirmatory study was designed after an exploratory finding on development data and evaluated once on a sealed test set of 8,517 questions. In the pre-registered confirmatory comparison, which selected operating points with a bootstrap-screened 1 pp accuracy-loss rule, the probe saved more of the four-loop budget than the native gate (+0.053 gross macro saving over three benchmarks, 95% CI [+0.049, +0.057]), mostly from CommonsenseQA, where the probe's test upper bound on accuracy loss exceeded 1 pp. Under a point-only rule, a pre-specified secondary analysis on the same test set reversed the preference (-0.022 [-0.026, -0.018]), with operating points that exceeded the budget. In pre-specified secondary comparisons, the probe showed no demonstrated advantage over an untrained KL divergence between successive loops (+0.00125 [-0.0041, +0.0065]); we emphasise this result post hoc and claim no equivalence. The probe also broke its budget on one of two out-of-distribution benchmarks (3.26 pp loss on Social IQa). Savings are estimated from stored full-depth traces, not measured inference time, and our screen carries no formal risk-control guarantee. We conclude that halting methods should be compared as signals together with selection procedures, with strong untrained baselines and out-of-distribution checks.

cs.LG · cs.CL

版本更新 · v2arXiv:2608.25654

When Answer Evaluation Misleads Confidence Selection: Fact-Grounded Form Grading for Open-Ended QA

Zhexi Feng, Bingrui Zhang, Wuxi Chen

原文摘要

Many benchmarks now prefer open-ended questions for testing LLM and NLP methods. We find that the choice of a calibration rule can flip when only the correctness labels change: on 141 pilot answers, rules chosen with exact-match labels look 0.239 Brier points better than raw confidence but are 0.103 worse under human labels. How such flips arise remains unclear. We look deeply into their mechanism: reference matching rejects many correct answers (exact match misses 42%), and an exact Brier identity shows these errors favour rules giving such answers lower confidence, which abstain on them at common thresholds. Guided by this, we propose FactForm, a grader that trusts reference matches and otherwise consults evidence retrieved from the question alone. One LLM call per answer fills a fixed-order closed form that a program scores. On 998 fresh EVOUNA answers with 400 fact-based human labels, FactForm reaches a balanced error of 0.077 and an AUROC of 0.985, against 0.112 and 0.930 for the best of eight traditional graders. Its same-model stability is no worse than that of LLM judges (mean absolute run-to-run difference 0.0136 against 0.0159 to 0.0206). We hope this helps benchmarks grade open-ended answers more accurately, with repeatable and discriminative scores.

cs.CL · cs.AI

版本更新 · v2arXiv:2608.29455

Item-Mean Surrogates: Why Richer Personas Fail to Deliver Individual Fidelity in LLMs as Human Surrogates

Daehwan Ahn, Chengfeng Mao, Dokyun Lee

原文摘要

LLMs are increasingly used to simulate and study human behavior. Such studies assume richer personas improve predictions of specific people's answers. We test this premise across four public datasets. First, we split each person's answer to an item into the item mean (the average human answer to that item) and the person's deviation from the item mean. Much of the accuracy that surrogate studies report comes from predicting item means. LLM predictions explain little of the deviations. We call this pattern item-mean surrogacy. Second, we split each deviation into stable signal and noise using a dataset in which the same participants answered the same questions twice. Stable signal is the part of a deviation that repeated answers share. Noise is the part that changes. In this dataset, more than half of the variance in deviations persists across occasions. Most of what the LLM misses is stable signal rather than noise. Third, we split stable signal into the person signal, a person's average deviation across questions, and the person-by-item signal, their remaining question-specific deviations. The person-by-item signal accounts for about 8.5 times as much variance as the person signal. The LLM recovers about a tenth of it. Richer personas, different models, and fine-tuning on human answers still leave most of the person-by-item signal unrecovered. Both fine-tuned models recover less of this signal than their untuned versions. These findings call for closer empirical scrutiny of LLMs as human surrogates.

cs.CL · cs.CY · cs.LG

新论文 · v1arXiv:2611.02332

LEANGUARD: Lean-Verified Guardrails for Tool-Using Agents

Sreehari Sankar, Debargha Ganguly, Vikash Singh, Mohsen Hariri, Weicong Chen, Yanyan Zhang, Shouren Wang, Wang Yang 等 20 位作者

原文摘要

Tool-using language-model agents often execute actions that their domain policies forbid. We present LeanGuard, a runtime that checks every proposed tool call against a policy written in Lean4 before the call reaches the environment. Policies combine state predicates with scoped past-time temporal operators over the execution history, evaluation is fail-closed, and the decision procedure is proved equivalent to a LeanLTL semantics. A denied call can be corrected in two ways. With feedback, the agent receives the violated rule and a hint; with proof-guided repair, Lean metaprogramming searches for a nearby call that preserves the user's intent and proves that the policy admits it, without the model. On Tau-Bench with five Qwen models (0.8B-27B), unguarded agents execute at least 33 calls that the Lean policy forbids, while LeanGuard executes none; with feedback, task success does not change significantly. On DTap, LeanGuard reduces direct prompt-injection success from 57.9% to 5.9%, and in telecom, where benign tasks need no writes, benign success is unchanged. Feedback reduces repeated blocked calls and raises task success for four of five models over a bare denial. On replayed denials, repair proves an admissible edit for 88 of 252 denials caused by argument values, in a median of 2,ms.

cs.AI · cs.CL

版本更新 · v2arXiv:2610.29207

norregel: a preregistered benchmark of dated numeric values in Norwegian law

Eirik Botten Nicolaysen

原文摘要

norregel asks language models for numeric values in Norwegian statutes, regulations and parliamentary decisions (NOK amounts, percentages, multiples of the National Insurance basic amount (G) and age limits), each at a stated date. The 916 items were extracted mechanically from Lovdata's open data, and each carries the dated history of its provision, so an answer can be classed as correct, outdated, wrong or uninterpretable. We report a first and a second preregistered run on the same items. In the first run (five models, listed publicly only after local analysis), H1, that outdated answers are more frequent for values that took effect after a model's cutoff, could not be tested in substance, and H2 and H3 were inconclusive; Opus 5.5 and Sonnet 4.6 answered 46.7% and 40.2% of the 880 primary items correctly. The second run was preregistered and published before any of its calls, and eight models were analysed. Its pooled primary test supported H1 under the preregistered decision rule (d = 17.8 percentage points, 95% interval [7.7, 28.9], p = 0.0001). In the preregistered sensitivity analysis S8, which counts as outdated only the most recent value that differs from the key, d was 3.9 [-3.6, 12.7] (p = 0.1754), an interval that contains 0. H1 was supported in every other preregistered sensitivity analysis, but S8 is the only one that gives both groups one target value, so we do not regard the support for H1 as robust. The share of outdated answers equal to that most recent value (H2) was 56.1% [46.6, 65.0], an interval that contains 0.5, and the pooled test of a lower share correct in Nynorsk than in Bokmål (H3) was inconclusive. Several models, among them Haiku 4.5, gave mostly uninterpretable answers. Items and keys are withheld; item IDs, per-answer classes, code and results are on Zenodo.

cs.CL

新论文 · v1arXiv:2611.02330

Spatial Transfusion: From Structured Spatial Scene Understanding to Generation and Editing

Jaeyeon Kim, Sebastian Braun, Hannes Gamper

原文摘要

Spatial audio conveys not only what sounds are present, but where they originate and how they interact with the environment. Recent audio-language models increasingly understand and reason about such scenes, yet spatial audio understanding, generation, and editing remain largely separate. We present Spatial Transfusion, a unified first-order ambisonics (FOA) framework that extends a pretrained spatial audio understanding model to generation and editing through flow matching over VAE latents. We formulate editing as a structured scene transformation over a taxonomy of atomic edits. Keeping the pretrained backbone frozen, we introduce modality-based LoRA routing and a curriculum that progressively adds generation, editing, and natural-language interaction. Experiments show that Spatial Transfusion largely preserves spatial understanding while outperforming full fine-tuning in generation and editing. The resulting model supports controllable FOA generation and editing through both structured and natural-language interfaces. Audio samples are available at https://microsoft.github.io/spatial-transfusion/

cs.SD · cs.AI · eess.AS

新论文 · v1arXiv:2611.02327

Findings-aware Language-Image Pretraining for CT Volume Interpretation

Takuro Shimaya, Junya Inoh, Masahiro Saiko

原文摘要

Computed tomography (CT) provides diverse patient information and various clinical tasks, for which numerous deep learning models have been investigated. Recent years have seen the development of vision-language foundation models that can comprehensively extract clinical features, driven by representation learning that combines CT volumes with radiology reports. Existing approaches typically adopt Contrastive Language-Image Pre-training (CLIP)-based representation learning, where pairs from different cases are, in principle, treated as negative pairs such that the volume and report embeddings are pushed apart. In the biomedical imaging, however, treating different cases as negative pairs is not always appropriate, particularly when clinical similarity exists. In this study, we propose a supervised representation learning framework grounded in clinical knowledge from language foundation models, rather than the weak supervision used in CLIP. Specifically, we utilized a large language model to extract abnormal finding phrases from reports. We then explicitly constructed the target features that the image encoder should output, as the semantic representation of the extracted findings using a biomedical text embedding model. Our Findings-aware Language-Image Pre-training (FiLIP) outperformed weakly supervised approaches in multi-label zero-shot finding classification in both in-domain and external-domain evaluations, and also proved effective when fine-tuned for the downstream task of radiology report generation.

cs.CV · physics.med-ph

版本更新 · v2arXiv:2608.19172

Cell-Probe Lower Bounds and Complexity-Preserving Reductions for Suffix Array Queries

Dominik Kempa, Tomasz Kociumaka

原文摘要

For a text $T$ of length $n$ over an alphabet of size $σ$, its suffix array lists the starting positions of the suffixes of $T$ in lexicographic order, and its inverse suffix array gives the lexicographic rank of the suffix starting at each position. Since the introduction of the FM-index and the compressed suffix array in 2000, both queries have been supported in $O((\log_σn)^ε)$ time using $O(n\logσ)$ bits, for any constant $ε>0$. Yet no nontrivial time-space lower bound for suffix-array queries in compact space was known. We give the first such lower bound: in the cell-probe model with $Θ(\log n)$-bit words, every deterministic $S$-bit data structure with $S\ge n$ answering suffix-array queries on binary strings of length at most $n$ has worst-case query time $Ω(\log\log n/\log((S/n)\log\log n))$. Hence, structures with $O(n(\log\log n)^{O(1)})$ bits need $Ω(\log\log n/\log\log\log n)$ query time, while constant query time requires $Ω(n\log^εn)$ bits for some constant $ε>0$, matching known upper bounds. In particular, no such $O(n)$-bit structure supports constant-time queries, answering the 25-year-old question of Grossi and Vitter. We also give exact complexity-preserving equivalences between suffix-array access and simpler prefix queries on short strings. For every $2\leqσ\leq n$, suffix-array queries are equivalent to prefix-select queries, and inverse-suffix-array queries are equivalent to prefix-special-rank queries. The reductions in both directions preserve all four standard measures up to constant factors: space, query time, preprocessing time, and preprocessing space. Unlike previous reductions, they incur no additive $O(\log\log n)$ query-time term. Thus, the corresponding prefix-query problems capture suffix-array and inverse-suffix-array access without asymptotic loss in any of the four measures.

cs.DS

新论文 · v1arXiv:2611.03739

Reasoning is topologically simple

Chris Kapulkin, Nathan Kershaw

原文摘要

We formulate and investigate the hypothesis that reasoning text has simpler co-occurrence topology than non-reasoning text, and that higher-quality reasoning has simpler co-occurrence topology than lower-quality reasoning. We discuss and test applications to AI safety and AI evaluation.

cs.AI

新论文 · v1arXiv:2611.02319

Accessed, never disclosed: Silent access to protected health information by medical artificial intelligence agents

Jae Wang Lee, Abhirup Banerjee, Yeong Chan Lee

原文摘要

Medical artificial intelligence agents should retrieve patient identifiers only when a task requires them, the data-minimization principle behind the HIPAA minimum necessary standard for protected health information (PHI). An agent can retrieve identifiers without revealing them. Such silent access remains invisible to evaluations that inspect only the output. We embedded canaries in synthetic patient records and linked each tool invocation to canary exposure in the final answers across five language models, four boundary levels, and seven attack frames for identifier-free clinical tasks. Three models invoked the PHI tools unnecessarily, whereas Claude Opus 4.8 and GPT-5.5 did not. Stronger boundaries reduced this access, and the residual access under the strictest policy was confined to frames in which the user overrode the policy or asserted approval. A pre-execution gate given the task-necessity criterion nearly eliminated this access. Output-only evaluation cannot verify minimum-necessary access; boundaries, gates, and access-log audits are required together.

cs.AI · cs.CL · cs.CR · cs.CY

新论文 · v1arXiv:2611.02318

A Capitulation Direction Emerges With Scale and Is Causally Load-Bearing Where Validation Is Strong: Eight Models, Two Families, and Five Measurement Hazards

Saad Aamir, Muhammad Awais Bin Adil

原文摘要

Sycophancy in language models is well documented as a behaviour but poorly localised as a mechanism. We ask whether capitulation under user pushback, the abandonment of a correct answer when a user objects, corresponds to a linear direction in the residual stream, and whether such a direction emerges with scale. Across eight instruction-tuned models spanning 0.5B to 14B parameters in two families, under a pre-registered protocol with leave-one-question-out cross-validation, within-question permutation nulls, and a positive control, we find no usable direction below 7B and a direction that clears our validation gate at 7B, 8B and 14B. Ablating that direction in Qwen2.5-7B reduces the rate at which the model reverses a correct answer by 4.8 percentage points, roughly a quarter in relative terms, while the identical intervention on a gate-failing direction and on a marginally passing one produces no effect. Probe quality, not statistical significance, predicts causal relevance. Behaviourally, pushback destroys three to five times more correct answers than it repairs at 1B and reaches parity by 7B, falsifying our own pre-registered prediction, and a sign-reversing crossover between pushback types survives an order of magnitude of scale in Qwen and holds at 8B in Llama, with marginalisation over five paraphrases per type. We additionally quantify five measurement hazards, including a numerical-precision effect that changes roughly one episode label in ten.

cs.CL · cs.LG

版本更新 · v3arXiv:2606.00043

Fake Plastic Voters: When Political Parties Can Use AI-Simulated Focus Groups

Claudio Novelli, Javier Argota Sanchez-Vaquerizo, Jennifer Cyr, Giuliano Formisano, Simon McDougall, Giulia Sandri, Luciano Floridi

原文摘要

Political parties strive to understand their electorates, and focus groups are a vital tool in these efforts. AI-enhanced simulation technologies (AESTs) enable synthetic focus groups in a fraction of the time (and cost), raising the question of when and how such simulated evidence can be used in campaign research. This paper develops a decision matrix to help party strategists match research needs to appropriate simulation technologies and to identify when to escalate to hybrid or fully human focus groups. The matrix combines three dimensions: strategic purpose, deployment risk, and empirical grounding of the simulation tool. Strategic purpose is the decisive dimension, as it determines what kind of evidence the focus group is meant to produce: observing how political meanings and identities emerge through interaction (Mode 1) or testing and refining campaign messages (Mode 2). The matrix shows that, given documented failure modes such as sycophancy, persona drift, and the suppression of minority viewpoints, AESTs cannot replace human interaction in Mode 1 at any risk level. Within Mode 2, suitability depends instead on deployment risk and on the empirical grounding. Yet even here, we caution that routine reliance on AESTs may erode the qualitative craft on which sound judgment depends.

cs.CY · cs.MA

新论文 · v1arXiv:2611.02314

GRIPS: An Empirical Study of Pre-SQL Grounding Information for SQL Generation

Ekata Mitra, Yufei Tao, Kamalamalini Nagasundaram, Ameeta Agrawal

原文摘要

Text-to-SQL systems answer natural language questions over structured data, but can fail before SQL generation by linking question mentions to the wrong table, column, value, or relation. Even valid SQL can then return an incorrect answer. This raises a central question: What pre-SQL information helps a SQL generator, and when does additional information cease to help? We investigate this question on TempTabQA-C and BIRD. Among the audited SQL Static failures, grounding errors form the largest category on both benchmarks. We introduce GRIPS (Grounding Representation and Input Processing for SQL generation), a three-stage method that derives structured pre-SQL information to make these decisions explicit before query generation. Across GPT-5-nano and Llama-70B, GRIPS-Adaptive improves over SQL Static in 18 of 20 settings, with an average accuracy gain of 11.8 percentage points, while reducing model-call token usage by 47.8%. Our controlled analysis shows which grounding information can improve SQL generation, whereas adding more pre-SQL information does not consistently help. By separating question-input grounding from SQL generation, GRIPS makes text-to-SQL more accurate, efficient, and easier to diagnose.

cs.CL

新论文 · v1arXiv:2611.02312

Improving Financial Cybersecurity Through Fraud Transaction Detection with Deep Learning

Harish Kant Pathak, Rahul Kumar

原文摘要

Financial fraud detection remains a critical challenge in cybersecurity, with traditional rule-based systems struggling to keep pace with evolving attack patterns. This paper presents a deep learning approach for fraud transaction detection, leveraging neural network architectures to identify anomalous financial behaviors with high accuracy. We evaluate multiple deep learning models including LSTM, CNN, and hybrid architectures on real-world financial transaction datasets, demonstrating significant improvements over conventional machine learning baselines. Our approach achieves robust performance in detecting fraudulent transactions while minimizing false positives, providing a scalable solution for enterprise financial cybersecurity systems

cs.LG

新论文 · v1arXiv:2611.02311

CytoWave: Perturbation-Centric Pretraining for Single-Cell Response Prediction

Yuwei Miao, Jingquan Yan, Hehuan Ma, Thao M. Dang, Lin Xu, Siyuan Zhang, Junzhou Huang

原文摘要

Predicting gene expression under perturbations from single-cell data is a fundamental task for modeling cellular responses, with applications in discovering gene function and designing therapeutics. However, existing methods often struggle to generalize across datasets due to differences in experimental protocols, biological contexts, and species. We present \textbf{CytoWave}, a perturbation-centric pretraining framework for single-cell genetic perturbation response prediction. We leverage large-scale cross-species perturbation datasets unified via orthology and train CytoWave in a two-stage manner. In the first stage, we learn reference control cell states via masked gene expression recovery on unperturbed cells. In the second stage, we pretrain the model to predict perturbed gene expression given control states and perturbation signals, together with a contrastive objective that aligns latent representations of perturbation responses with perturbation signals. We perform strict dataset-held-out evaluation across multiple condition combinations and case studies of pathway activity changes, demonstrating consistent and effective performance of CytoWave in predicting single-cell genetic perturbation responses.

cs.LG

新论文 · v1arXiv:2611.02310

Detect, Repair and Audit: A Hierarchical Multi-Agent Defense Against Adversarial Attacks in Energy Forecasting

Zewei Shi, Ruoxi Sun, Minhui Xue, Feng Liu, Hua Ma, Xingliang Yuan, Surya Nepal

原文摘要

Accurate electricity-demand forecasting is essential for the modern energy infrastructure. Recently, Times-Series Large Language Models (TS-LLMs) have emerged as a promising approach to forecasting more accurate demand from historical consumption data in zero- and few-shot settings, helping energy operators effectively plan supply and allocate energy resources. However, these models are vulnerable to compromised sensor inputs that can distort forecasts and lead to erroneous market orders or unnecessary load shedding. Existing defenses focus primarily on end-device forecasting, overlooking the distinct data-drift, latency, and audit requirements across energy infrastructure. To address this gap, we extract from major energy standards and regulatory frameworks a three-layer conceptual model that captures the security postures for each layer. Guided by this model, we develop a Multi-Agent Reinforcement Learning (MARL) detector coupled with signal repair at local control centers, designed to mitigate distributional data drift through complementary feature-based detection and clean-reference-calibrated repair while meeting latency budgets. We further deploy an evidence-grounded LLM auditing agent with human-in-the-loop validation for audit-critical Virtual Power Plants (VPPs). In our evaluation, the best MARL configuration achieves an F1 score above 0.91 across all dataset-attack pairs, while signal repair reduces adversarial-to-clean input Relative Squared Error (RSE) by up to 83%. The few-shot auditor detects 77.78% of attacks in the Sherlock dataset. On the evaluated hardware, detection and repair require 7.95 ms on average, with a worst-observed latency of 14.50 ms, remaining within the local-center latency budget informed by IEEE 1646.

cs.CR

首次观测 · 记录 #1软件学报:jos:7771软件学报

中文生成式引擎优化

李彤, 吴荣榜, 李清昊, 贾贵, 高翔宇, 范举, 徐恪

原文摘要

大语言模型(large language model, LLM)和生成式引擎正在改变用户获取网络信息的方式. 传统搜索引擎优化主要关注网页能否进入搜索结果并获得较高排名, 而生成式引擎优化则关注网页内容能否被大语言模型采用并呈现在回答中. 然而, 现有研究主要面向英文场景, 面向中文网页和国内生成式平台的系统性生成式引擎优化 (generative engine optimization, GEO)研究仍然缺失. 为此, 构建面向中文场景的生成式引擎优化评测框架, 围绕中文数据集构建、生成式引擎优化策略设计和来源可见度评测开展研究, 并从策略、平台、领域和来源位置等方面分析中文生成式引擎优化的效果. 在此基础上, 进一步比较中文生成式引擎优化与英文生成式引擎优化的差异, 并总结面向中文场景的实践启示.

cs.* · cs.CL

软件学报 · 2026 · 正式发表

新论文 · v1arXiv:2611.02306

Lie Routing Vision Language Models for View-Robust 3D Understanding

Jingjun Yi, Chen Hu, Qi Bi, Hao Zheng, Huimin Huang, Yongkang, Wei Ji, Yawen Huang 等 10 位作者

原文摘要

Multimodal Large Language Models (MLLMs) have made substantial progress in 3D scene understanding from multi-view RGB observations. Yet a fixed object and query can elicit inconsistent target-related responses across camera viewpoints, a phenomenon we characterize as viewpoint-induced semantic drift. We introduce LieVLM, a Lie routing framework for view-robust 3D understanding. Motivated by the rotational component of viewpoint change, Lie routing adapts visual representations by applying local $\mathrm{SO}(3)$ transformations to projected features within learned three-dimensional subspaces. Geometry tokens extracted by VGGT predict local rotation parameters and channel-wise routing coefficients. A closed-form Padé--Cayley map transforms projected visual features, whose identity-relative corrections are projected back and gated into the original visual and geometry features. This design retains the pretrained semantic and direct spatial context while learning view-dependent feature adjustments. Experiments on 3D captioning, grounding, detection, and spatial reasoning show consistent improvements over additive fusion across training scales and backbones. Controlled ablations examine the roles of geometric conditioning and transformation structure. Viewpoint analyses further reveal systematic variation in the predicted routing states alongside more localized target responses and stronger task performance across viewpoint changes.

cs.CV · cs.LG