跳到正文
研踪PaperTrail

论文库

检索计算机领域论文,按研究方向、作者或关键词筛选。

部分采集异常
正在读取账号与研究记录…

全站更新概况

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

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

统计口径与数据新鲜度

统计 2026-11-15 14:16 — 2026-11-16 14:16(北京时间) · 点击数字查看对应简报;AI 相关包含在新增论文内。

最近完成检查:2026-11-16 01:34最近发现新增版本:2026-11-16 13:44库内论文最新更新时间:2026-11-14 02:59

每分钟刷新已保存的数据。空检查不代表采集异常;新收录可能包含历史论文。AI 导读按新增版本计数。

23,827 已收录论文19,941 AI 相关6,921 有修订112 HF 社区发现

浏览论文库

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

找到 34 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:2611.18533

ModHarness-RL: Recasting Harnesses as Learnable and Evolving Programs for Agentic RL

Zelin He, Siyu Zhu, Jacob Helwig, Yujian Liu, Li An, Rohit Jain, Shiyu Chang, Alborz Geramifard

原文摘要

Agentic reinforcement learning (RL) increasingly relies on a harness, the scaffold around the model that decides what it sees of its own interaction and what guidance it receives. Most existing approaches train under a hand-written harness that stays fixed throughout. However, an effective harness depends on the domain, the data and the model, and keeps changing as the model trains. We introduce ModHarness-RL, which recasts the harness as programs that are learned from data and evolve as the model trains with RL. ModHarness-RL records the interaction as structured State, a set of typed blocks, and two core components of the harness are programs over these blocks, a Compression program that assembles the accumulated history into the model's context and a Steering program that supplies guidance when behavior deviates. During RL training, candidate programs are proposed from recent trajectories, verified by execution, and replayed over past trajectories, so the harness tracks how the model's behavior shifts. We study CodeSearch in depth and further analyze ALFWorld and BIRD SQL, with models from 1.5B to 8B. The proposed Compression and Steering mechanisms improve held-out performance over the fixed-harness baseline in most settings. These results suggest that a harness for agentic RL can be learned from the task's data and evolved with the model.

cs.AI · cs.CL · cs.LG

新论文 · v1arXiv:2611.17954

In the Eye of the Beholder: Human Label Variation in Textual and Visual Entailment

Monica Riedler, Jasmin Orth, Raffaella Bernardi, Barbara Plank

原文摘要

Human Label Variation (HLV) is increasingly recognised as an informative signal that reflects meaningful differences in how people interpret the same input. While HLV has been studied extensively in Natural Language Inference (NLI), much less is known about HLV in Visual Entailment (VE; Xie et al., 2019), where the textual premise is replaced by an image, keeping the hypothesis and label space fixed. We introduce VariNLI-VE, a dataset for studying how HLV changes between paired NLI and VE instances. We re-annotate a subset of originally neutral SNLI (Bowman et al., 2015) instances and their visual counterparts, with the same annotators evaluating both the textual and visual versions. Each annotation includes a label, confidence score, and explanation, yielding 2,250 annotations. Labels and explanations are collected jointly, providing ecologically valid explanations. Because the image may reveal information not stated in the text, one might expect disagreement to decrease in VE. Instead, we find no clear overall reduction: disagreement decreases for some instances while it increases for others. Explanations reveal further variation, as annotators can retain the same label while changing their rationale. When evaluating vision-language models (VLMs) on the paired data, we find lower accuracy, weaker alignment with human label distributions, and generally higher uncertainty in VE than in NLI.

cs.CL

版本更新 · v2arXiv:2611.03093

An Experimental Study of Hybrid Retrieval, NLI and Open-source LLMs for Scientific Claim Verification

Lakshmi Narayanan P, Narenkarthik K, Naren Kumar C

原文摘要

Claim verification is the task of deciding whether a source text supports a claim, contradicts it, or lacks sufficient evidence. Natural Language Inference (NLI) models can judge whether a single sentence supports or contradicts a claim, but they struggle when the evidence is spread across several sentences. Conversely, Large Language Models (LLMs) can reason across sentences, but may draw conclusions the text does not support. In this experimental study, we test whether a conservative fusion rule, labeling a claim Support or Contradict only when both models agree and defaulting to Not Enough Evidence otherwise, combines their strengths. We construct a benchmark of 2,035 claims from SciFact and CliniFact (858 Support, 152 Contradict, 1,025 Not Enough Evidence). We retrieve the top five most relevant sentences per claim using a hybrid of semantic and lexical (BM25) retrieval, which slightly outperforms either method alone. These retrieved sentences are then provided to every verification model. We then compare a DeBERTa-v3-base-mnli-fever-anli NLI baseline against Llama 3.1 8B, Qwen3 8B, and Mistral 7B. Qwen3 achieves the highest Macro-F1 (0.7937), followed by Llama 3.1 (0.7204) and Mistral (0.6388), while the NLI baseline reaches only 0.3157. Contrary to our hypothesis, fusion degrades performance, dropping Macro-F1 to 0.7392, 0.6730, and 0.5991, respectively. To understand this decline, we analyze agreements, disagreements, and common error patterns between the NLI and LLM verifiers. Our error analysis shows that NLI and LLM disagreements often result in correct LLM predictions being overturned.

cs.CL · cs.AI · cs.LG

版本更新 · v2arXiv:2605.12147

PrivacySIM: Evaluating LLM Simulation of User Privacy Behavior

James Flemings, Murali Annavaram

原文摘要

Large language models (LLMs) are increasingly used to simulate human behavior, but their ability to simulate individual privacy decisions is not well understood. In this paper, we evaluate whether a core set of user persona attributes can drive LLMs to simulate individual-level privacy behavior across various contexts. We introduce PrivacySIM, an evaluation suite that benchmarks LLM simulation of user privacy behavior against the ground-truth responses of 1,000 real users. These users are drawn from five published user studies on privacy behavior spanning LLM healthcare consultations, conversational agents, and chatbots. Drawing on these user studies and prior work in privacy behavior, we hypothesize three persona facets as plausible predictors of privacy decision-making: demographics, previous experiences, and stated privacy attitudes. We condition nine frontier LLMs on subsets of these three facets and measure how often each model's response to a data-sharing scenario matches the user's actual response. Our findings show that (1) zero-shot conditioning on these facets consistently improves simulation quality, but the performance of the best model (40.4% accuracy) suggests that richer and more nuanced persona facets are needed to reliably align with users' mental models for privacy decisions. (2) A user's stated privacy attitudes alone are often not the best predictor because they can diverge from the user's actual privacy behavior. (3) Users with high AI/chatbot experience are the most challenging to simulate. PrivacySIM is a first step toward understanding and improving the capabilities of LLMs to simulate user privacy decisions. We release PrivacySIM to enable further evaluation of LLM privacy simulation.

cs.CR · cs.LG

版本更新 · v3arXiv:2609.32084

An Asymptotically Tight Bound on the Number of Distinct Squares in Circular Words

Rikuya Hamai

原文摘要

A square is a word $xx$ with $x$ nonempty. We study the maximum number of distinct squares occurring in the rotations of a word of length $n$. Amit and Gawrychowski (SPIRE 2017) obtained an upper bound of $3.14n$. Charalampopoulos et al. (CPM 2026) improved the upper bound to $\lceil9n/5\rceil$ and gave a binary construction with at least $3n/2-O(\sqrt n)$ distinct squares for infinitely many $n$. They conjectured an upper bound of $3n/2$. Li and Song subsequently reported an improved upper bound of $5n/3$. We prove that a circular word of length $n$ containing $σ$ distinct letters has at most $\lfloor3(n-σ)/2\rfloor$ distinct squares. This settles the conjecture and establishes that the leading coefficient $3/2$ is optimal even for binary words. Our proof groups squares by their Lyndon roots and, for primitive words, uses a weighted count of circuits in the Rauzy graphs of the bi-infinite repetition of the word.

math.CO · cs.DS

版本更新 · v2arXiv:2611.13963

Euclidean and Hermitian Self-Orthogonal Codes from Divisible Codes

Ziling Heng, Chenkun Xie

原文摘要

Self-orthogonality and divisibility are two fundamental properties of linear codes, and the relationship between them has long been of interest. Huffman and Pless (2003) showed that every binary \(4\)-divisible code is Euclidean self-orthogonal, that a ternary code is Euclidean self-orthogonal if and only if it is \(3\)-divisible, and that a quaternary code is Hermitian self-orthogonal if and only if it is \(2\)-divisible. The general case remained open for a long time. In 2024, Li and Heng settled the odd-characteristic case, proving that every \(p\)-divisible code over \(\F_{p^m}\), with \(p\) odd, that contains the all-one vector is Euclidean self-orthogonal. In this paper we treat the even-characteristic case: we prove that every \(4\)-divisible code over \(\F_{2^m}\) containing the all-one vector is Euclidean self-orthogonal. We also prove that, for every prime \(p\), every \(p\)-divisible linear code over \(\F_{q^2}\), where \(q=p^m\), that contains the all-one vector is Hermitian self-orthogonal. In addition, we generalize the Hermitian criterion to the Galois self-orthogonal case, proving that any $p$-divisible linear code over $\gf_{q^r}$ that contains the all-one vector is $\ell$-Galois self-orthogonal for \(\ell\in\{1,\dots,r-1\}\). These criteria make the construction of self-orthogonal codes from divisible codes simple and efficient. As applications, we recover the known Euclidean and Hermitian self-orthogonality of generalized Reed--Muller codes and obtain several infinite families of new Euclidean and Hermitian self-orthogonal codes, which give rise to infinite families of quantum codes with excellent parameters.

cs.IT

新论文 · v1arXiv:2611.13630

Machine-learning decoders for fusion-based quantum computing in real time

Daniel Barter, Jacob C. Bridgeman, Chris Dawson, Terry Farrelly, Shane Fleming, Mercedes Gimeno-Segovia, Paul Hartke, Timothy Hein 等 15 位作者

原文摘要

Fault-tolerant quantum computation requires decoders that combine accuracy, throughput, low latency, and scalability to large numbers of physical qubits. In photonic fusion-based architectures, erasures from photon loss and probabilistic fusion measurements are a dominant error mechanism, motivating decoders that remain accurate at high erasure rates. We introduce PRISM, a transformer-based machine-learning decoder designed to exploit the known locations of erasures in protocols subject to both Pauli and erasure errors. We benchmark PRISM across a wide range of fault-tolerant protocols based on surface codes, color codes, and quantum low-density parity-check codes, with distances up to 24. Across these benchmarks, PRISM matches the logical error rates of integer-programming-based decoders while being significantly faster, including at error rates up to 50% of threshold. We further optimize the decoder and its real-time hardware implementation to reduce latency while maintaining accuracy. Our optimized implementation on an AMD VP1902 FPGA can decode logical blocks of distance 8 in both space and time, demonstrating a latency of 6.59 microseconds and an amortized decoding time of 1.56 microseconds per logical block is possible. These results show that machine-learning decoding can provide accurate, low-latency inference for erasure-dominated fault-tolerant quantum architectures.

quant-ph · cs.AR · cs.LG

版本更新 · v3arXiv:2610.27745

Hearsay or Fact? How Framing and Elicitation Shape LLM Probability Updates

An-Sheng Jhang

原文摘要

We measure LLM probability updates on $δ$-SNLI premise-hypothesis pairs across the Multi-Candidate Verbalization (MCV) family. Separate calls elicit priors and posteriors; directional movement s signs changes by the annotated direction. We cross stated vs. reported wording with Direct single estimates (D0) vs. Multi-Candidate Weighted elicitation (MCW; five weighted estimates, V0, weighted mean M_W). Frozen formal tests cover 15 configurations (23,400 planned units): 3 primary (Qwen3-8B thinking off, OLMo-2-7B, gpt-6.1-sol via codex; 6,600), 4 CLI-harnessed configurations (gpt-oss-120b, claude-sonnet-5.5, gemini-3.8-flash low/high; dnli-agy; 5,600), and 8 un-harnessed REST API configurations (dnli-api; 11,200). We add a 20-configuration, 8-provider dev ladder (3B-550B; 6,000 units) and a 2-model OpenAI 1+1+4 ablation of D0, Direct-CoT (D_CoT), MCW (V0), and Multi-Candidate Unweighted (MCU; V2 / M_U; 1,800 calls). All 15 reject H1 (s = +16.1 to +28.9 points) and H4a (endpoint reduction +0.065 to +0.391); 13 reject H2 (+2.0 to +8.9 points), except gpt-oss-120b (+0.9) and sf:qwen3.5-4b (-0.64). In 14 of 15 formal configurations, readouts obey Top-1 > MCW (M_W) > MCU (M_U). On 120 common pairs, API Qwen3 tracks local Qwen3 (H1 +18.72 vs. +18.90; H4a +0.265 vs. +0.254; H2 +3.78 vs. +1.67; H4b -9.03 vs. -6.70); local H2 rejects on all 240 pairs (+1.97), not this subset. API Gemma 31B (formal H4b -2.85, Top-1 +0.64; spread -13.3% / +7.6%) and Groq gpt-oss-120b (dev -4.15, +0.40; -18.4% / +12.2%) show small MCW costs and stated narrowing/reported widening without CLI harnesses. In OpenAI ablations, D_CoT keeps numeric-token entropy <1 bit; MCU (V2) shows smaller movement loss but halves spread and $ΔH_2$ vs. MCW (V0). Because $δ$-NLI annotates only direction, we make no claim about appropriate update magnitude. Code/data: https://github.com/asj252/llm-judgment-studies

cs.CL · cs.AI

新论文 · v1arXiv:2611.13512

Nepali-English Code-Mixed Speech Recognition: Data Preparation and Adaptation of a Speech Foundation Model

Jeevan Neupane, Janardan Bhetwal, Kshitiz Paudel, Anup Aryal, Bipin Bashyal

原文摘要

Nepali speakers regularly mix English words into conversation, but multilingual speech recognisers often struggle with language switches and the writing script used for English words. We study how to adapt SraVaani, a 449M-parameter hybrid TDT-CTC model, using a 172.8-hour corpus covering six Nepali speech domains. Its pretrained tokenizer does not support digits and several common symbols. Rather than replace it or discard these recordings, we generate Nepali and English readings for numbers and use zero-shot ASR output to select a plausible form. Full fine-tuning reduces test word error rate (WER) from 48.6% to 17.1%; a 12-epoch run reaches 16.3%, compared with 16.4% for fine-tuned Whisper large-v3-turbo. Freezing the encoder increases WER to 25.5%, while excluding numeric clips raises WER on numeric test speech from 14.3% to 21.3%. We also evaluate on FLEURS using both its original references and mixed-script alternatives. The comparison shows that transcription conventions explain part, but not all, of the out-of-domain performance loss.

cs.AI

新论文 · v1arXiv:2611.13140

Agreeing on the Answer but Not the Reason: Measurement Limits of LLM Reasoning Failure Taxonomies

Madhu Sudhan Rao Beesetty

原文摘要

Failure taxonomies for LLM reasoning are increasingly reported as measurements: papers state what fraction of a model's errors fall into named categories, then compare those fractions across models. We show that for one such category the reported fraction is largely a property of the measurement procedure. We defined Compositional Collapse (a model notices a required sub-part of a problem, then drops or misapplies it) and measured it across twelve models and eleven languages on a public multilingual math benchmark (2,316 graded records). First, 50.1% of failures carry evidence for more than one category, so the reported plurality depends on the order in which the classifier tests them: reversing two tests moves it in eleven of twelve models, and across all six orderings Collapse's share spans 26.4% to 76.4%. Second, agreement collapses for the category but not for the task. Native speakers of all eleven languages judged 179 transcripts, agreeing with our inspector at kappa = 0.932 on answer correctness but only kappa = 0.093 on Collapse. A second native speaker per language re-rated 142 items; on 121 determinate pairs the two humans agree with each other at kappa = 0.659, while each agrees with the inspector at kappa <= 0.094. Re-running the same inspector model with the human instructions verbatim raises kappa against Human 2 from 0.147 to 0.636, so the gap is largely attributable to the prompt, not the model. Third, the inspector labels Collapse 1.81 times as often as native speakers, but only 1.17 times once a defective benchmark item (a deleted clause in the Bengali and Chinese translations) is excluded. We recommend four cheap checks before comparing failure-mode rates across models: report label co-occurrence, permute the tie-break, validate on a shared independent sample, and audit systematic disagreements.

cs.AI · cs.CL

新论文 · v1arXiv:2611.12599

Consistent Mapping for Fluid Structure Interaction with Geometrically Nonlinear Isogeometric Beams

Juan Ignacio Camarotti, Ricky Aristio, Roland Wüchner

原文摘要

Isogeometric beam formulations provide an efficient reduced-dimensional representation of slender structures, combining accurate geometry and high-order continuity with few structural degrees of freedom. Their application to fluid--structure interaction (FSI) requires transferring motion and loads between the one-dimensional beam centerline and the surrounding fluid interface. While mapping strategies exist for conventional finite element beams, geometrically nonlinear isogeometric formulations require operators consistent with their spline-based kinematics and distinct descriptions of cross-sectional rotation. This work develops consistent beam-to-surface mappings for geometrically nonlinear isogeometric Bernoulli and Timoshenko beams in partitioned FSI. Fluid-interface points are attached to the beam in the reference configuration and reconstructed using the nonlinear kinematics of each formulation. Consistent linearization yields a configuration-dependent mapping operator whose transpose ensures work-conjugate load transfer. The formulations are verified through analytical deformation tests, convergence studies, and virtual-work checks, and evaluated in beam-based adaptations of the Mok and Turek--Hron benchmarks and the Goland wing flutter benchmark. The results confirm accurate interface reconstruction and work-conjugate load transfer under large deformations, while demonstrating the applicability of both beam formulations to coupled FSI problems.

physics.comp-ph · cs.CE

版本更新 · v2arXiv:2509.16801

Sublinear Time Quantum Sensitivity Sampling

Zhao Song, David P. Woodruff, Lichen Zhang

原文摘要

We present a unified quantum sensitivity sampling framework that streamlines coreset construction for many classical approximation problems, significantly speeding up clustering, regression, and low-rank approximation. We assume $n,d\geq\mathrm{poly}(k/ε)$ and, for regression, $n\geq\mathrm{poly}(d)$: - $k$-median/$k$-means: For $n$ points in $\mathbb{R}^d$, we build an $ε$-coreset of input points of size $\tilde O(k^{4/3}ε^{-2})$ (resp. $\tilde O(k^{3/2}ε^{-2})$) in time $\tilde O(ε^{-1}n^{0.5}dk^{1.67})$ (resp. $\tilde O(ε^{-1}n^{0.5}dk^{1.75})$). Our better $d$- and linear $1/ε$-dependence improve on [Xue, Chen, Li and Jiang, ICML'23] for $d\geqΩ(k^{7/3})$ (resp. $Ω(k^{2.5})$). - $\ell_p$ regression: For $\min_x\|Ax-b\|_p$, $A\in\mathbb{R}^{n\times d}$, we build an $ε$-coreset of size $\tilde O_p(d^{\max\{1,p/2\}}ε^{-2})$ in time $\tilde O_p(n^{0.5}d^{1.5}ε^{-1})$ for $p\leq 2$ and $\tilde O_p(n^{0.5}d^{p/4+1}(ε^{-1}+d^{0.5}))$ for $p>2$, matching [Liu and Ji, ICML'26] for $p\leq 2$ and improving on the prior best quantum sampling approach [Apers and Gribling, SICOMP'26] for all $p>2$. - Low-rank approximation: We give the first quantum sublinear-time algorithm for Frobenius-norm rank-$k$ approximation of $A\in\mathbb{R}^{d\times n}$ free of data-dependent parameters, running in $\tilde O(n^{0.5}dk^{0.5}ε^{-0.5})$ time with factored output (plus $\tilde O(nk^2ε^{-2})$ to materialize them). We also give sublinear algorithms for kernel and tensor variants. - Lower bound: For generic sensitivity sampling with total sensitivity $S$, any quantum algorithm outputting an $ε$-coreset of size $o(\sqrt{nS})$ needs $Ω(\sqrt{nS})$ queries, so our query complexity is optimal up to $\mathrm{poly}(ε^{-1},\log n)$ and VC-dimension factors.

cs.DS · cs.LG · quant-ph

新论文 · v1arXiv:2611.10567

On Relative Greedy Weights

Olav Geil

原文摘要

Relative greedy weights were introduced by Li and Liu in [19] as a measure of information leakage in coordinated two-party wire-tap channels of type II when the adversary follows a greedy strategy. Their definition can be viewed as a relative version of the CEZ greedy weights, which are only locally greedy. In this paper, we revisit the concept of greediness and relative greediness introducing new measures, and we discuss other perhaps more relevant applications of relative greedy weights than the coordinated two-party wire-tap channel of type II. We pay special attention to bottom-up greedy weights, which correspond to an adversary with an ultimate greedy approach, and we present families of nested codes for which such weights are larger than the relative CEZ greedy weights.

cs.IT · cs.CR

版本更新 · v2arXiv:2610.31229

Iterative Randomized Rounding for $k$-Median in General and Euclidean Metrics

Jarosław Byrka, Yuhao Guo, Yang S. Hu, Shi Li, Yongho Shin, Phanu Vajanopath, Chengzhang Wan, Zaixuan Wang

原文摘要

We refine the iterative rounding framework of Byrka, Guo, Hu, Li, Wan, and Wang [FOCS'26], which gives a $(2+\varepsilon)$-approximation for $k$-median. Our algorithm opens $k+O(1)$ facilities with expected connection cost at most $1.9999$ times the standard LP value. Despite the small numerical gain, this shows that the LP integrality gap is below $2$ with a constant additive facility allowance, partially answering Li and Svensson's open question [SICOMP'16]. Their reduction yields a $1.9999$-approximation opening at most $k$ facilities. Our preprocessing makes the LP solution $\mathrm{poly}(\varepsilon)$-integral with a $(1+\varepsilon)$ loss in each client's expected connection cost and $O(1)$ extra facilities, without special treatment of clients with concentrated fractional facilities. We replace batch processing with single-facility openings and combine a tighter analysis for clients with unequal distances to their facilities with a complementary algorithm for nearly uniform distances. In Euclidean metrics, we obtain a $1.71955$-approximation for $k$-median and a $1.4561$-approximation for uncapacitated facility location (UFL). For every $p\ge2$ and sufficiently small $\varepsilon>0$, we obtain a $((3^p+2)/3+\varepsilon)$-approximation for $k$-clustering with $p$-th power distance costs. For $p=2$, our $(11/3+\varepsilon)$-approximation for $k$-means improves the $(3+\ln2+\varepsilon)$ bound of Anand, Charikar, Cohen-Addad, Gao, Grandoni, Lee, Sharma, and van Wijland [SODA'27]. The Euclidean $k$-median, $k$-means, and UFL ratios are below the respective general-metric hardness bounds $1+2/e$, $1+8/e$ [STOC'02], and $1.463$ [J. Algorithms'99]. The Euclidean $k$-median bound relies on a geometric inequality proved via a nonnegativity certificate constructed from positive-semidefinite kernels.

cs.DS

新论文 · v1arXiv:2611.07880

Quasi-polynomial-sized vertex multicut-mimicking networks and kernelization for Vertex Multiway Cut

Eduard Eiben, Roohani Sharma, Magnus Wahlström

原文摘要

We give deterministic polynomial-time constructions of vertex multicut-mimicking networks of quasi-polynomial size. Given a graph G, a terminal set T, and a positive integer c, our algorithm constructs a graph with |T|^7 c^{O( log c ) } vertices that preserves every minimum vertex multiway cut value of at most c, for every partition of the terminals. Setting c=|T| gives an exact vertex multicut-mimicking network with |T|^{O( log |T| ) } vertices, resolving an open question of Wahlström [ACM Transactions on Algorithms 2022], who showed a bound of κ^{O( log^3 κ)} for the edge multicut-mimicking variant, where κ:= \sum_{t \in T} |N_G(t)| is the capacity of the terminal set, which was later improved to κ^{O( log^2 κ)} by Anand, Lee, Li and Saranurak [STOC 2024]. As applications of our vertex multicut-mimicking network, we obtain deterministic kernels of size k^{O( log k )} for Vertex Multiway Cut, p^{O(1)} k^{O(log k)} for Vertex Multicut with p requests and |Γ|^{O(1)} k^{O( log k)} for Group Feedback Vertex Set, where k is the deletion budget and Γis a finite group. Our first kernelization result, in particular, resolves an open question by Wahlström [ACM Transactions on Algorithms 2022], which was reiterated in a recent survey on kernelization by Jansen [Computer Science Reviews 2026]. Also, to the best of our knowledge, our kernel for Vertex Multicut is the first one whose size is completely polynomial in p and (quasi-polynomial) super-polynomial only in k. Finally, again using our vertex multicut-mimicking network for graphs, we obtain a hyperedge multicut-mimicking network for hypergraphs, which preserves hypergraph edge-multicuts of size up to c, of size |T|^{O(1)} c^{O(log c)}. Our result removes the dependence on the rank of the hypergraph from the exponent of the size bound, that existed in prior work.

cs.DS

新论文 · v1arXiv:2611.05772

Hotplug Coded Caching: Embedding the YMA and Interference Elimination Schemes via Reed--Muller Codes

Daniela Tuninetti

原文摘要

In the hotplug coded caching model, $K$ users fill their caches of size $M$ as a function of $N$ files, but only an unknown subset of $A<K$ users is active in the delivery phase. We recently showed that (i) under uncoded placement, offline users cause a strict load penalty at intermediate memory sizes, so that coded placement is needed; and (ii) for $AM\leq N$, every classical scheme with row-wise placement, including the scheme of Gomez-Vilardebo (GV), can be embedded in a Reed-Solomon code, which removes the penalty. In this note, we show that every point of the scheme of Yu, Maddah-Ali and Avestimehr (YMA) for a classical system with $A$ users is achievable in the hotplug model for every $K\geq A$. Each user is associated with a hyperplane and each file with a homogeneous polynomial; each user caches the restriction of the polynomial to its hyperplane, and the subfiles seen by the active users are the values of the polynomial at the points where their hyperplanes meet. This is a projective Reed-Muller code, which reduces to the Reed-Solomon code when each subfile is cached by one user. We then show, by an existence argument, that every classical scheme whose cache is a linear compression of the Maddah-Ali-Niesen cache and whose correctness holds for generic coefficients can also be embedded over a sufficiently large field; this includes the interference elimination (IE) schemes of Tian and Chen. These embeddings complement the Reed-Solomon one, which allows any row-wise placement and any delivery. Together with past work, our results show that every point attained by a GV, YMA or IE scheme ($N\leq A$ for IE), and every corner point of the linear tradeoff for $A=N=3$, is achievable in the hotplug model with a load that does not depend on $K$. Whether the optimal hotplug tradeoff depends only on $A$ remains open, but these results support a positive answer.

cs.IT · eess.SP

版本更新 · v2arXiv:2610.21364

CoDeFont: One-Shot Font Generation via Multi-Scale Factor-Specific Representation Learning

Rejoy Chakraborty, Prasun Roy, Saumik Bhattacharya, Umapada Pal

原文摘要

One-shot Font Generation (OFG) aims to synthesize a target glyph by transferring the stylistic characteristics of a single reference glyph while preserving the structure of the target character. However, accurately capturing font style from a single reference image remains challenging due to the limited stylistic information available in a single glyph, and this challenge is further amplified for complex scripts. We propose CoDeFont, a script-agnostic OFG framework that addresses this challenge through multi-level style-content factorization and progressive style-content reconciliation. CoDeFont independently aggregates complementary style and structural information across multiple spatial levels to obtain factor-specific representations, while a contrastive objective explicitly encourages separation between style and content embeddings. During reconstruction, the disentangled representations are progressively injected through multi-level style conditioning, enabling stylistic information to be incorporated at different decoding resolutions while preserving the underlying glyph structure. We evaluate CoDeFont on structurally distinct Bengali and Latin scripts across seen and unseen font styles. Extensive quantitative and qualitative analyses demonstrate consistent improvements over existing state-of-the-art methods, highlighting the effectiveness of the proposed framework for OFG across diverse writing systems.

cs.CV

新论文 · v1arXiv:2611.03397

Online Algorithms Cannot Find Isolated Perceptron Solutions

Shuangping Li, Gabe Schoenbach

原文摘要

The binary perceptron problem asks for a length-$n$ sign vector in the intersection of $\lfloor αn\rfloor$-many independent random Gaussian half-spaces. A striking feature of this model is strong freezing: at every constraint density $α> 0$, all but a vanishing fraction of solutions are isolated -- $Ω(n)$-far from all others, in Hamming distance. Such solutions are believed to be algorithmically inaccessible. Nevertheless, efficient algorithms find solutions at some constraint densities $α$. One possible resolution of this apparent contradiction is that the solution space also contains rare well-connected structures, and that algorithms succeed by locating them. Gong, Huang, Li and Sellke (FOCS 2026) proved this for stable algorithms, by showing that such algorithms cannot locate isolated solutions with probability more than $\approx 0.84$. In this work, we study online algorithms, which receive the columns of the constraint matrix one by one and sign each irrevocably upon arrival. This is a widely used class of algorithms in discrepancy minimization, and it includes the algorithm of Jo and Lee (2026), which has one of the best provable guarantees for the binary perceptron. We prove that at any constraint density, any online algorithm outputs an isolated solution with probability $o(1)$. Our proof is based on the observation that the signs an online algorithm has already fixed can be completed to at most one isolated solution, so a successful algorithm must choose the unique correct sign for each of the last few columns. Using the exchangeability of these columns, we show that such an algorithm must match a suffix-parallel baseline, which signs each of them as if it were the next column to arrive. We then use an anti-concentration argument to show that such a baseline rarely produces a unique completion.

cs.CC · cs.DS · math-ph · math.PR

版本更新 · v3arXiv:2610.20079

Undirected Multiple-Unicast Network Coding and Faster External-Memory Integer Sorting

Baochun Li, Zongpeng Li, Sirui Liu

原文摘要

In undirected multiple-unicast networks, where independent messages are sent between specified source--destination pairs, network coding allows intermediate nodes to combine messages, whereas fractional routing splits and forwards individual messages along paths without mixing them. Proposed in 2004 by Li and Li and independently by Harvey, Kleinberg, and Lehman, the multiple-unicast conjecture---a central conjecture in network coding with deep connections to computational complexity---asserts that network coding and fractional routing have the same achievable rate region. We construct a counterexample refuting this conjecture and use it to reduce block transfers in external-memory data rearrangement, thereby accelerating external-memory integer sorting and answering in the affirmative the longstanding question of Aggarwal and Vitter (1988) on whether integer operations can beat the comparison-sorting input/output (I/O) bound. Specifically, we give a deterministic sorting algorithm: for $N$ integers of $Θ(\log N)$ bits, a memory holding $M$ integers, and block transfers of $B$ integers, where $2B\le M\le N$, the number of block transfers is \[ O\!\left(\frac{N}{B}\left[1+ \frac{\log N}{(\log(M/B))^{1-\varepsilon}(\log M)^\varepsilon} \right]\right), \] where $\varepsilon=1.5\times10^{-5}$ is an absolute constant. This bound asymptotically improves on the classical $Θ\!\left(\frac{N}{B}\left[1+ \frac{\log(N/M)}{\log(M/B)}\right]\right)$ comparison-sorting bound in certain parameter regimes. Under the external-memory model specified in this paper, we have formally verified the algorithm's correctness and I/O complexity bound in Lean~4.

cs.DS · cs.CC · cs.IT

新论文 · v1arXiv:2611.02041

ECGJev: Question-Conditioned Option Scoring for 12-Lead ECG Interpretation

Zhuoyuan Li, Xidong Wu, Hanyu Cui, Jiansong Gao, Rongrong Fu, Mingke Yan, Haotian Tang

原文摘要

Clinical interpretation of a 12-lead electrocardiogram (ECG) often requires answering targeted diagnostic questions with clear, quantifiable decisions. We introduce ECGJev, the first ECG framework built on a decision decoder, a pretrained language model trained to return a probability for every listed option. A question-conditioned bridge compresses frozen waveform and spectrogram features into 16 ECG tokens, and a two-stage procedure adapts the bridge and low-rank adapters on the decoder's native option readout using diagnostic labels alone, scoring all options for a question in one forward pass. Evaluated across four ECG benchmarks, ECGJev achieves the highest macro-F1 on CPSC2018 (80.46%), PTB-XL (71.68%) and Chapman (96.86%), and generalizes competitively to the small-sample PTB cohort. Concept probes and perturbation analyses further show that its predictions rely directly on the ECG input, remain stable under option-order and wording variations, and transfer zero-shot to unseen waveform concepts. Our code is available at https://github.com/zhuoyuanli-lab/ECGJev.

cs.LG · eess.SP

新论文 · v1arXiv:2611.01583

Second-Order Asymptotics for the Scalar Gaussian Broadcast Channel

Vincent Y. F. Tan

原文摘要

This paper characterizes the second-order coding region of the scalar Gaussian broadcast channel along the curved boundary of its capacity region. The characterization holds under both average and maximal error criteria and allows correlation between the receiver noises. At each interior point of this boundary, the second-order region is a halfspace, as in the work of Scarlett and the author on Gaussian multiple-access channels with degraded message sets. Achievability follows from spherical superposition coding with a blocklength-dependent power allocation. The main innovation lies in the converse. Using Ornstein--Uhlenbeck entropy dissipation and Gaussian functional inequalities, we construct an auxiliary observation through an additive Gaussian noise channel and establish quantitative relative entropy bounds for the channel output distributions conditioned on this observation. The strong functional representation lemma of Li and El Gamal then allows us to select a realization of its independent random seed to obtain a deterministic partition of the retained receiver-1 message set. The component label plays a role analogous to the auxiliary random variable in the capacity characterization, while providing the distributional control needed at the second-order scale. These properties guide the construction of auxiliary output measures whose likelihood ratios admit the cancellations and normal approximation needed to evaluate an information spectrum converse.

cs.IT

新论文 · v1arXiv:2610.39747

When the Top Answer Decides: Gender Bias in Decision Models

Yeb Havinga

原文摘要

Decision models are language models built to route, classify and triage: they return option probabilities from one forward pass, and an application acts on the top answer. They are offered off the shelf, trained for the decision format rather than for fairness or any one application, and we found no bias measurement in their cards. We test TypeSafe's closed Jev and five open decision models on gender-swapped yes/no decisions about people (Tamkin et al., 2023), BBQ, StereoSet and the gender-bias studies of Fulgu and Capraro (2024). Five of the six give women significantly more favourable answers than men (+0.05 to +0.10 in log-odds), and a gender swap changes 1-3.5% of top answers (Jev's partly through its call-to-call variation). The makers' recommended confidence gate removes many biased commitments but raises the stereotyped share of those it keeps in ambiguous BBQ gender questions (granite-switch 0.76 to 0.85, pplx 0.81 to 1.00). Probes show gender preferences in the same models: all six rate violence for a greater good as more acceptable against a man than against a woman, and two default to "girl" when forced to guess a child's gender without a cue. Decision-level and probability-level measures differ, so audits should read a model as its application does.

cs.CL · cs.CY

版本更新 · v2arXiv:2602.15493

LEADER: Lightweight End-to-End Attention-Gated Dual Encoder-Decoder Network for Robust Minutiae Extraction

Raffaele Cappelli, Matteo Ferrara

原文摘要

This paper introduces LEADER (Lightweight End-to-end Attention-gated Dual Encoder-decodeR), a neural network that maps raw fingerprint images to minutiae descriptors, including location, direction, and type. The proposed architecture integrates non-maximum suppression and angular decoding to enable end-to-end inference using only 0.9M parameters. It employs a novel ``Castle-Moat-Rampart'' ground-truth encoding and a dual encoder-decoder architecture, interconnected through an attention-gating mechanism. Experimental evaluations demonstrate state-of-the-art accuracy on plain fingerprints and robust cross-domain generalization to latent impressions across two benchmark datasets. Specifically, LEADER outperforms top-tier commercial systems, achieving F1-score gains of up to 34% and 39% over the closest competitor on NIST SD27 and SD302, respectively. Qualitative visual analysis suggests that the internal representations learned by the model align with established fingerprint domain concepts, such as segmentation masks, orientation fields, frequency maps, and skeletons. Inference requires only 15ms on GPU (average latency per image at batch size 32) and 322ms on CPU, outperforming leading commercial software in computational efficiency. The source code and pre-trained weights are publicly released to facilitate reproducibility.

cs.CV

新论文 · v1arXiv:2610.36484

CogArena: Benchmarking Multimodal Agents on Interactive Behavioral Experiments

Kianté Fernandez, Caitlin Chen, Jialu Zhou, Bartek Sadowski, Anthony C. Miceli, Ian Krajbich

原文摘要

Claims about the cognitive capabilities of large language models are typically tested by translating behavioral paradigms into text prompts, bypassing the visual interfaces and time pressure through which human participants actually engage with experiments. At the same time, AI agents can now complete online behavioral tasks and produce human-like data, which raises concerns that online datasets in the social and behavioral sciences are increasingly open to contamination. We introduce CogArena, a benchmark of ten experimental paradigms from the social and behavioral sciences. Agents complete the experiments through a standard web browser, working with the same visual stimuli and response interfaces that online participants use. A three-level scoring procedure evaluates whether the agent completed the task, how accurately it performed, and whether its data show the behavioral signatures each paradigm reliably produces in people. Across four multimodal models, each run ten times per task, we find that finishing an experiment and reproducing its signatures are different things. Many sessions did not finish, and those that did reproduced some signatures reliably and others not at all. CogArena provides the tasks, scoring code, and a public leaderboard for new submissions.

cs.AI · cs.CY · cs.HC · q-bio.NC

新论文 · v1arXiv:2610.34451

Perceived Acceptability of Sensory Augmentation

Michel Witter, Alwin de Rooij, Michel van Dartel, Emiel Krahmer

原文摘要

Sensory augmentation (SA) can extend how people perceive and engage with public environments through the body, offering opportunities to support shared multimodal experiences between deaf and hearing visitors. Designing such systems requires understanding which combinations of tactile stimuli and body locations users perceive as acceptable. We examined the perceived acceptability of fictitious yet feasible SA configurations in a museum use-case. Using a modified technology acceptance model, we surveyed 307 deaf and hearing participants about tactile stimulus types and body locations. The results showed significant effects of stimulus type and body location on acceptance, but no interactions with hearing status. Participants generally preferred temperature-based stimuli, subtle movement, texture, and tickling over vibration, and distal over proximal body locations. These findings suggest that tactile SA design should be guided primarily by embodied and contextual factors rather than hearing status, informing future systems for shared multimodal experience in public cultural settings.

cs.HC

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1210ACL Anthology · ACL 2026

EQUIP: EQUivariant preserving In-Place updates for Efficient Token Pruning

Arun Ramachandran, R Govindarajan, Murali Annavaram University of Southern California, Prakash Raghavendra

原文摘要

Token-pruning has emerged as a primary focus in large language models (LLMs) to enhance model efficiency while preserving accuracy, especially for large sequence lengths. However, the eviction operation of token-pruning methods causes “holes” in KV tensors, posing two major challenges: (1) The shift operation, required to make the KV tensor contiguous, results in significant copy overheads; (2) The changes in position indices due to token eviction lead to increased computational requirements for Rotary Positional Encoding (RoPE). To address these issues, we introduce EQUIP, an EQUivariant preserving in-place token update mechanism that ensures the equivariance property of the operations performed in the attention computation. EQUIP offers two fundamental advantages: First, it combines eviction and a subsequent token insertion into an in-place replacement operation, which reduces the KV cache copy overheads significantly. Second, EQUIP reduces recomputation of rotation operations through a combination of in-place update, caching and a re-indexing strategy. Together, these optimizations enable EQUIP to achieve geomean speedups of 1.62× (or 1.47×) on CPU (GPU) over StreamingLLM, and 3.45× (or 1.86×) on CPU (GPU) over Heavy Hitters (H2O). EQUIP with Paged Attention achieves speedups of 4.18×(2.61×) on CPU (GPU) over auto-regressive baselines. EQUIP matches the model accuracy of baseline pruning methods while delivering superior performance.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-long.1901ACL Anthology · ACL 2026

GenPT: Beyond Self-Report for Reliable LLM Psychometrics via Generative Projective Testing

Ming Wang Singapore Management University and Northeastern University, Shuang Wu Northeastern University, Bixuan Wang, Lu Lin, Yuxin Chen Central University of Finance and Economics, Xiaocui Yang Northeastern University, Daling Wang, Shi Feng Northeastern University, China 等 10 位作者

原文摘要

Self-report questionnaires remain the default tool for probing the psychological characteristics of Large Language Model (LLM) agents, yet classical instruments (BFI, BDI, MBTI, BSS) inherit three well-known threats under LLMs: contamination from training corpora, directional bias under social-desirability framing, and limited responsiveness to context beyond the item text. We ask whether a projective paradigm can be adapted into a usable psychometric tool for LLM agents. We introduce GenPT (Generative Projective Testing), which reformulates TAT, Rorschach, and SCT with newly generated stimuli and organises assessment as a three-stage pipeline (Behavior Collection \to Interpretation \to Diagnosis) grounded in SCORS-G and a Simplified Rorschach Analysis System. On personality traits (Big Five, MBTI) and mental-health risks (depression, suicide ideation), questionnaires exhibit systematic directional shifts under social-desirability framing, most strongly on suicide ideation, whereas GenPT’s collected behavioral patterns stay near the symmetric baseline; under a longitudinal counselling context, GenPT-based depression assessment shifts by roughly an order of magnitude more than its questionnaire counterpart. Questionnaires remain competitive on clean-persona trait tasks where items align lexically with the persona description. Overall, GenPT complements rather than replaces self-report when contamination resistance, bias asymmetry, and context sensitivity matter. Code and stimuli: https://github.com/sci-m-wang/GenPT.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-srw.11ACL Anthology · ACL 2026

Annotation Entropy Predicts Per-Example Learning Dynamics in LoRA Fine-Tuning

Brady Steele Georgia Institute of Technology and Apple

原文摘要

Annotator disagreement on tasks like natural language inference (NLI) reflects genuine linguistic ambiguity, yet most fine-tuning recipes treat every example as equally learnable. We ask whether this external signal of ambiguity predicts per-example learning dynamics under LoRA, the most widely used parameter-efficient fine-tuning method, and find that it does. Correlating annotation entropy (from ChaosNLI’s 100 labels per example) with per-example area under the loss curve (AULC) on SNLI and MNLI, the correlation is positive in all 25conditions tested (Spearman \rho {=} 0.06-0.43), with decoder-only models showing stronger correlations than encoders at matched LoRA rank. More strikingly, under LoRA contested examples exhibit un-learning: their gold-label loss increases during training, a pattern that is largely absent under full fine-tuning and IA^3 in the encoder setting where matched comparisons are available, and that we also observe under LoRA on two decoder-only models. The effect survives partial-correlation controls and replicates across seeds and datasets. A preliminary noise-injection experiment is consistent with these findings.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop) · 2026 · 正式发表

首次观测 · 记录 #1ACL Anthology · ACL 2026:acl:2026.acl-srw.48ACL Anthology · ACL 2026

CAL-Log: Cost-Aware Active Learning with Logarithmic Cognitive Effort Modeling and Online Adaptation to Human Annotation Behavior

Vihanga Supasan Kariyakaranage, Banuka Athuraliya

原文摘要

Active learning (AL) reduces labeled data requirements in NLP, yet most methods optimize label efficiency while ignoring annotation cost. Standard uncertainty sampling assumes uniform effort, leading to suboptimal resource allocation when documents vary in length. Supasan and Athuraliya (2026) introduced CAL-Log, a cost-aware AL variant using logarithmic cost modeling C(x)=α+β log(1+L(x)), where C(x) is the predicted annotation time for document x and L(x) is its token length, grounded in information foraging theory (Pirolli and Card, 1999) and psycholinguistic studies of human skimming (Rayner, 1998). This paper presents CAL-Log in full, extending that preliminary framework with two new contributions: temperature-scaled calibrated entropy and online per-annotator cost adaptation, which together resolve the cold-start calibration bottleneck identified in the prior work. Experiments on ten text classification benchmarks demonstrate a 3.3× speedup over BADGE (Batch Active learning by Diverse Gradient Embeddings; Ash et al., 2020) and 3.9× over Entropy sampling to reach F1=0.80, with large effect sizes (Cohen’s d>0.8). A live annotation deployment with preliminary user evaluation (N=7) confirms that the online cost model produces reading-speed classifications consistent with annotator self-reports, and that a transparency interface successfully communicates the scoring rationale to non-expert users.

cs.CL

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop) · 2026 · 正式发表

首次观测 · 记录 #1CVF · CVPR 2026:cvf:156b15107850857c08dcc424bd5fdba7CVF · CVPR 2026

NeuroFlow: Toward Unified Visual Encoding and Decoding from Neural Activity

Mai, Weijian, Nan, Mu, Zhu, Yu, Cao, Jiahang, Zhang, Rui, Dai, Yuqin, Song, Chunfeng, Luo, Andrew 等 9 位作者

原文摘要

Visual encoding and decoding models act as gateways to understanding the neural mechanisms underlying human visual perception. Typically, visual encoding models that predict brain activity from stimuli and decoding models that reproduce stimuli from brain activity are treated as distinct tasks, requiring separate models and training procedures. This separation is inefficient and fails to model the consistency between encoding and decoding processes. To address this limitation, we propose NeuroFlow, the first unified framework that jointly models visual encoding and decoding from neural activity within a single flow model. NeuroFlow introduces two key components: (i) NeuroVAE is designed as a variational backbone to model neural variability and establish a compact, semantically structured latent space for bidirectional modeling across visual and neural modalities. (ii) Cross-modal Flow Matching (XFM) bypasses the typical paradigm of noise-to-data diffusion guided by a specific modality condition, instead learning a reversibly consistent flow model between visual and neural latent distributions. For the first time, visual encoding and decoding are reformulated as a time-dependent, reversible process within a shared latent space for unified modeling. Empirical results demonstrate that NeuroFlow achieves superior overall performance in visual encoding and decoding tasks with higher computational efficiency compared to any isolated methods. We further analyze principal factors that steer the model toward encoding-decoding consistency and, through brain functional analyses, demonstrate that NeuroFlow captures consistent activation patterns underlying neural variability. NeuroFlow marks a major step toward unified visual encoding and decoding from neural activity, providing mechanistic insights that inform future bidirectional visual brain-computer interfaces. Code and project page are publicly available at https://michaelmaiii.github.io/NeuroFlow-S.

cs.CV

Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition · 2026 · 正式发表