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 19:20 — 2026-11-16 19:20(北京时间) · 点击数字查看对应简报;AI 相关包含在新增论文内。

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

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

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

浏览论文库

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

找到 23,827 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-16 01:33(北京时间)
新论文 · v1arXiv:2611.12567

LawKAN: Group-Law Layers for Composition -- When Imposing a Known Law Pays Off

Anna Mishchenko, Nikolai Gavrilin

原文摘要

Many tasks compose transformations by a known rule: chained rotations in robotics and navigation. Learned models relearn it, compounding error; classical estimators hardcode it and miss hardware deviations. We impose the known structure and learn only the remainder: an exact non-commutative group law sits in the node of a self-contained layer, its edges learn local increments, and a zero-initialized correction psi learns unknown deviations of the law. A pre-training criterion predicts exact representability. On fixed-angle SU(2) composition, a 54-parameter dense edge is the most accurate dense model, at 27x fewer parameters than the strongest fixed-matrix baseline, and lies 0.05 deg above the Bayes floor. On real robot arms, an isolated 306-parameter psi reaches 1.12-1.14 mm versus 1.29-1.33 mm for Gauss-Newton at N >= 60, and the recipe transfers across three vendors and two protocols. The recipe matches classical fits where the data obey the law and loses where the imposed law is not the task's.

cs.LG · cs.RO

新论文 · v1arXiv:2611.12564

GHiML: VLM-Guided Hierarchical Multi-Label Learning with Applications to Benthic Imagery

Isaac Xu, Devi Ayyagari, Martin Gillis, Benjamin Misiuk, Craig J. Brown, Thomas Trappenberg

原文摘要

Vision language models (VLMs) encode substantial expert knowledge within their weights, yet their direct application to specialised classification domains such as benthic imagery remains limited. Meanwhile, vision transformers trained on these tasks rely predominantly on low-level visual latents (colours, textures, and shapes) that lack the higher-level reasoning humans employ for identification. We propose Guided Hierarchical Multi-Label (GHiML) learning, a pipeline that extracts conceptual latent variables from a VLM conditioned on training data, generates sample-relevant captions grounded in those variables, and fuses the resulting textual embeddings with image embeddings via cross-attention and a per-node gated router. We also introduce the inverted HML score ($d_{\mathrm{inv}}$), a similarity score that evaluates hierarchical multi-label predictions through graph traversal distance and a cardinality penalty, providing an interpretable alternative to existing measures, which are insensitive to realistic degradations in sparse, deep hierarchies. Experiments on BenthicNet, FathomNet and iNaturalist show that the router improves macro $F_1$ over the vision-only baseline by up to 6.7 pp and $d_{\mathrm{inv}}$ by up to 21.1 pp on in-domain benthic data, with gains persisting under distribution shift. Ablations locate the contribution of the text itself: on the multi-label biotic hierarchies, removing the per-sample text signal degrades $d_{\mathrm{inv}}$ while leaving macro $F_1$ unchanged, consistent with text acting through prediction cardinality rather than node discrimination, whereas iNaturalist, whose annotations are single-label, loses 18.7% of macro $F_1$. Our results support the hypothesis that VLMs contain exploitable specialised knowledge that, when appropriately elicited, can augment classification in domains where direct visual learning is insufficient.

cs.CV

版本更新 · v2arXiv:2603.15189

The Sampling Complexity of Condorcet Winner Identification in Dueling Bandits

El Mehdi Saad, Victor Thuot, Nicolas Verzelen

原文摘要

We study best-arm identification in large-scale stochastic dueling bandits under the sole assumption that a Condorcet winner exists, i.e., an arm that wins each noisy pairwise comparison with probability at least $1/2$. We introduce a new identification procedure that exploits the full gap matrix $Δ_{i,j}=q_{i,j}-\tfrac12$ (where $q_{i,j}$ is the probability that arm $i$ beats arm $j$), rather than only the gaps between the Condorcet winner and the other arms. We derive high-probability, instance-dependent sample-complexity guarantees that (up to logarithmic factors) improve the best known ones by leveraging informative comparisons beyond those involving the winner. We complement these results with matching lower bounds that establish the optimality of our procedures in all regimes. Overall, our results reveal the general form of the sampling complexity, characterized by a trade-off between the cost of locating informative entries and the verification cost required to achieve the desired confidence. In particular, this complexity drastically differs from what is suggested by pure asymptotic results or by procedures that are tailored to Strongly Stochastic Transitive models.

stat.ML · cs.LG

新论文 · v1arXiv:2611.12557

WanderBench: Can World Models Explore Dynamic Worlds from Video?

Yuhang Chen, Jiheng Li, Jiale Wang, Zai Wang, Guangyue Liu

原文摘要

Video world models conditioned on an input video make it possible to explore real, recorded dynamic scenes, enabling re-shooting of footage, free-viewpoint replay, and real-to-sim environments for embodied agents. Such uses require the generated rollout to stay faithful to the recorded world, yet existing benchmarks start from a single image or a text prompt, and video-conditioned generation is evaluated mainly on perceptual quality and camera following. These evaluations do not directly assess whether exploration preserves subject identity, event progression, and scene consistency upon revisiting a location. We introduce WanderBench, a benchmark for exploring dynamic worlds from video. It comprises 745 single-shot source videos of real and AI-generated content in three difficulty tiers (3-25 s), paired with camera routes given as keyboard actions, and evaluates six dimensions with 24 human-validated sub-metrics. Benchmarking eight models, we find that as clips grow longer, models keep the static scene and visual quality but lose camera control and the source dynamics, and that none keeps revisited places consistent. We further propose WanderLand, a training-free framework that conditions each frame of a frozen model on the same-instant source and, on revisits, on its own earlier static content. WanderLand improves five models on every dimension, by up to 8.3 points in source-dynamics fidelity.

cs.CV · cs.AI

版本更新 · v2arXiv:2605.28751

Correctness vs Efficiency: Behavior Control with Weight-Space Extrapolation in Code RL

Kunhao Zheng, Pierre Chambon, Juliette Decugis, Jonas Gehring, Taco Cohen, Benjamin Negrevergne, Gabriel Synnaeve

原文摘要

A model trained with an imperfect reward in RL may learn to exploit the reward with undesired behavior. We study a controlled form of this problem in competitive programming. A program may pass small tests while violating the intended efficiency requirement on larger inputs. We construct a family of rewards that progressively relax the efficiency constraint, making the reward increasingly easy to exploit, and train separate policies. This sweep reveals a correctness-efficiency frontier. Relaxing the reward leads to less efficient programs and more resource-limit failures. Stricter rewards suppress these failures but increase correctness failures near the model's capability boundary. This frontier is also reflected in weight space. Interpolating between the resulting models reproduces the same behavioral change. More surprisingly, extrapolation in both directions further amplifies or suppresses it beyond the trained checkpoints. Extensive experiments show that this pattern holds across pure reasoning, tool use, and agentic coding, and at both 32B and 7B scales. This suggests a simple intervention for this form of reward hacking: use a loose reward to amplify the undesired behavior, use the weight-space difference to a strict-reward model as its steering vector, and then extrapolate past the strict model to suppress it.

cs.LG · cs.AI · cs.CL

版本更新 · v4arXiv:2511.19208

Leader Election via Unique Sink Orientation

Jérémie Chalopin, Maria Kokkou

原文摘要

A proof-labeling scheme (PLS) certifies a global property using only local checks. That is, a prover assigns a certificate to every node and each node accepts or rejects by inspecting only the certificates in its neighborhood, so that every graph satisfying some global property admits a certificate assignment accepted by all nodes (existence) and every certificate assignment accepted by all nodes gives a legal configuration (soundness). In this work, we design PLSs for two classical problems, leader certification and spanning tree certification, assuming that all nodes are anonymous, for a specific class of graphs. Certifying a unique leader is impossible in general anonymous graphs for any certificate size due to a well-known result for rings by [Korman, Kutten and Peleg, 2010]. This raises the question of which graph classes admit a PLS for leader certification with anonymous nodes, ideally with size independent of the size of the graph. Our work contributes to a growing list of such classes, which includes chordal graphs, meshed graphs and simply connected subsets of the triangular grid. We consider dismantlable graphs, a class which contains trees, chordal, bridged and Helly graphs, as well as the $k$-th powers of $δ$-hyperbolic graphs for $k \geq 2δ$. To the best of our knowledge, these are the first PLSs tailored to dismantlable graphs, potentially highlighting structural properties useful for certifying additional problems. Finally, we present a generic transformation that converts any labeling scheme with certificates of finite type into a silent self-stabilizing algorithm by adding only one extra state, assuming a Gouda fair scheduler, whereas known generic transformations require $O(\ell + \log n)$ bits for $\ell$-bit certificates. This transformation may be of independent interest.

cs.DC

版本更新 · v2arXiv:2508.13288

Hierarchical Conformal Classification

Floris den Hengst, Inès Blin, Majid Mohammadi, Syed Ihtesham Hussain Shah, Taraneh Younesian

原文摘要

Conformal prediction (CP) is a powerful framework for quantifying uncertainty in machine learning models, offering reliable predictions with finite-sample coverage guarantees. When applied to classification, CP produces a prediction set of possible labels that is guaranteed to contain the true label with high probability, regardless of the underlying classifier. However, standard CP treats classes as flat and unstructured, ignoring domain knowledge such as semantic relationships or hierarchical structure among class labels. This paper presents hierarchical conformal classification (HCC), an extension of CP that incorporates class hierarchies into both the structure and semantics of prediction sets. We formulate HCC as a constrained optimization problem whose solutions yield prediction sets composed of nodes at different levels of the hierarchy, while maintaining coverage guarantees. To address the combinatorial nature of the problem, we formally show that a much smaller, well-structured subset of candidate solutions suffices to ensure coverage while upholding optimality. An empirical evaluation on three new benchmarks consisting of audio, image, and text data highlights the advantages of our approach, and a user study shows that annotators significantly prefer hierarchical over flat prediction sets.

cs.LG · cs.AI · stat.ML

版本更新 · v4arXiv:2509.00832

Challenges in Non-Polymeric Crystal Structure Prediction: Why a Geometric, Permutation-Invariant Loss is Needed

Emmanuel Jehanno, Romain Menegaux, Julien Mairal, Sergei Grudinin

原文摘要

Crystalline structure prediction is an essential prerequisite for designing materials with targeted properties. Yet, it is still an open challenge in materials design and drug discovery. Despite recent advances in computational materials science, accurately predicting three-dimensional non-polymeric crystal structures remains elusive. In this work, we focus on the molecular assembly problem, where a set~$\mathcal{S}$ of identical rigid molecules is packed to form a crystalline structure. Such a simplified formulation provides a useful approximation to the actual problem. However, while recent state-of-the-art methods have increasingly adopted sophisticated techniques, the underlying learning objective remains ill-posed. We propose a better formulation that introduces a loss function capturing key geometric molecular properties while ensuring permutation invariance over $\mathcal{S}$. Remarkably, we demonstrate that within this framework, a simple regression model already outperforms prior approaches, including flow matching techniques, on the COD-Cluster17 benchmark, a curated non-polymeric subset of the Crystallography Open Database (COD). Our code is available at https://github.com/inria-thoth/SinkFast.

cs.LG · cond-mat.mtrl-sci

版本更新 · v2arXiv:2604.22787

Conformal PM2.5 Mapping Under Spatial Covariate Shift: Satellite-Reanalysis Fusion for Africa's Green Industrial Transition

Yaw Osei Adjei, Davis Opoku, Ephraim Abotsi, Kwadwo Owusu Amanqua, Oliver Kornyo, Elisha Soglo-Ahianyo, Cephas Anertey Abbey, Ransford Oppong

原文摘要

Africa's green industrialization imperative demands reliable infrastructure for monitoring air quality. We present a satellite-reanalysis PM2.5 fusion system trained on 2,068,901 records from 404 monitoring locations in 29 African countries (OpenAQ, 2017-2022), combining LightGBM with leakage-resistant spatial cross-validation and conformal prediction to quantify predictions and their geographic applicability limits. Under 5-fold location-grouped spatial cross-validation, LightGBM achieves RMSE = 30.83 +/- 5.07 ug/m3, MAE = 14.54 +/- 1.66 ug/m3, R2 = 0.134 +/- 0.023, and macro F1 = 0.336 +/- 0.018. This R2 is substantially below random-split benchmarks (>0.90) but reflects true geographic generalisation difficulty rather than model failure. Split conformal prediction targeting 90% marginal coverage reveals severe East Africa degradation (actual PICP = 65.3% vs. nominal 90%), consistent with medium-strength covariate shift (humidity KS = 0.2237, sat_pblh KS = 0.2558). We operationalise these findings through regional reliability flags (High/Medium/Low/Unreliable) and a monitor prioritisation score directing infrastructure expansion toward highest-burden unmonitored populations, directly supporting Africa's green industrial transition and SDGs 3.9, 7.1.2, 9, 11.6.2, and 13.

cs.LG · cs.AI

版本更新 · v2arXiv:2609.27149

Temporally Ordered Region-Token Mamba with Logit-Space Diffusion for Remote Sensing Change Detection

Maneet Chatterjee, Aparup Ghosh, Anuvab Sen, Udayon Sen, Arnav Aditya, Yixin Zhang

原文摘要

Change Detection (CD) in remote sensing requires jointly modeling bitemporal context and fine-grained spatial details to accurately localize changed regions. While dense attention effectively captures bitemporal relationships, its computational cost limits scalability to high-resolution imagery. In contrast, conventional feature fusion and coarse decoding often struggle to distinguish true changes from appearance variations and preserve object boundaries. To overcome these limitations, we propose Bitemporal Mamba-Diffusion for Change Detection (BMD-CD), a framework that integrates temporally structured state-space modeling with logit-space diffusion refinement. BMD-CD first converts deep bitemporal features into region tokens and organizes them into explicit temporal partitions for bidirectional state-space propagation. It then employs a Learnable Trigonometric Fusion (LTF) module to integrate temporal information and a Conditional Diffusion Decoder that refines coarse predictions in logit space with a short DDIM schedule to generate fine-grained, high-resolution change maps. Extensive experiments on LEVIR-CD, WHU-CD, DSIFN-CD, CDD, and S2Looking demonstrate the effectiveness of the proposed approach. BMD-CD achieves IoU scores of 0.923, 0.957, 0.882, and 0.979 on WHU-CD, DSIFN-CD, LEVIR-CD, and CDD, respectively, and further exhibits strong zero-shot robustness on ValaisCD and B-FLAIR-test.

cs.CV

新论文 · v1arXiv:2611.12552

SciMARS: Multi-Agent Recursive Self-Improvement for Closed-Loop Scientific Discovery

Weimin Wu, Han Liu

原文摘要

We propose SciMARS, a framework for recursive task-local adaptation of multi-agent harnesses in closed-loop scientific discovery. Multi-agent systems coordinate analysis, experimental design, and verification as scientific evidence evolves. However, two problems limit their effectiveness: (i) without shared memory, agents lose observations and repeat work; (ii) fixed role procedures, handoffs, and memory policies perpetuate mistakes despite retained evidence. We address these problems with a shared evidence graph and separate component editors. SciMARS adapts roles and coordination; SciMARS + Memory adds a third editor for how agents record, retrieve, and receive evidence. Candidate probes guide composition and selection of the next harness. We evaluate scientific discovery through experimental-data replay and a digital twin. SciMARS + Memory resolves 12/18 GPT-6-Astra and 2/18 GLM-5.3 attempts on Autonomous Science Bench, compared with 8/18 and 1/18 for the base multi-agent team. On ScienceAgentBench and AppWorld, it solves 14/38 and 11/20 tasks, compared with 11/38 and 5/20 for the base team. These results support combining shared memory with harness adaptation across scientific and interactive tasks.

cs.AI

新论文 · v1arXiv:2611.12550

A Strict Upper Bound for Three-Point Weighted Selection in Planar Vector-Valued Regression

Sergey Volkov

原文摘要

We prove $F_{\mathrm{weighted}}(2,2,3)<11/6$ for minimum-Frobenius-norm least-squares regression with two input and two output coordinates, improving the upper bound $15/8$. More precisely, a universal $\varepsilon_0>0$ satisfies $\mathcal T_3\le(5/6-\varepsilon_0)E$ for every finite residual frame. The proof combines an exact scalar certificate, rigidity of equality in planar selection, and a compactness argument that controls disappearing weights. Together with the known lower bound, this gives $13/8\le F_{\mathrm{weighted}}(2,2,3)<11/6$. No numerical value of $\varepsilon_0$ is claimed.

math.ST · cs.AI · cs.LG

新论文 · v1arXiv:2611.13055

Beyond the Document: Compiled Context and Governed Agents for Any Work Product

James Rhodes, George Kang

原文摘要

A document preserves the visible result of professional work but often discards the research, calculations, assumptions, evidence, and authorization boundaries behind it. We study a different unit of delivery: a Portable Agent Format (PAF) package that carries a work product together with compiled context, evidence, provenance, and rules for subsequent interaction. To test whether this context remains useful beyond the process that produced it, we evaluate four new packages, created after the hosted and offline readers were fixed: an investment proposal, fund research, financial-statement analysis, and an engineering incident review. Two packages were produced by specialized agents and two through a general-purpose assistant's packaging skill. Across 400 frozen questions and 12,000 evaluated responses, a hosted governed reader succeeds on 96% of questions versus 86% with instructions plus raw-file retrieval (Holm-adjusted p < 10^-7). On the 260 answerable questions per condition, success is 94% versus 79%; the gap varies by package. Source-material chat and instructions alone reach 48% and 74% overall. A separate 0.8-billion-parameter offline reader succeeds on 69% overall, but 54% on answerable questions. Its -1 response rate is 2.9%, combining materially wrong and unauthorized answers, versus 56% with the source and instructions alone. Ablations suggest advantages for the integrated selector and package-adapted writer, without isolating representation from training effects. The same run exposes a failure: offline retrieval resolves none of the 40 historical-record questions in one financial-data package. These findings support treating compiled knowledge as part of the delivered work product, while leaving independent standards interoperability and end-to-end governance conformance unproven.

cs.SE · cs.AI · cs.IR

版本更新 · v2arXiv:2512.17414

A decomposition approach for large virtual network embedding

Amal Benhamiche, Pierre Fouilhoux, Lucas Létocart, Nancy Perrot, Alexis Schneider

原文摘要

Virtual Network Embedding (VNE) is the core combinatorial problem of Network Slicing, a 5G technology which enables telecommunication operators to propose diverse service-dedicated virtual networks, embedded onto a common substrate network. VNE asks for a minimum-cost mapping of a virtual network on a substrate network, encompassing simultaneous node placement and edge routing decisions. On a benchmark of large virtual networks with realistic topologies we compiled, the state-of-the art heuristics often provide expensive solutions, or even fail to find a solution when resources are sparse. We introduce a new integer linear formulation together with a decomposition scheme based on an automatic partition of the virtual network. This results in a column generation approach whose pricing problems are also VNE problems. This method allows to compute better lower bounds than state-of-the-art methods. Finally, we devise an efficient Price-and-Branch heuristic for large instances.

cs.DM · cs.NI

版本更新 · v2arXiv:2512.23676

Web World Models

Jichen Feng, Yifan Zhang, Chenggong Zhang, Yifu Lu, Shilong Liu, Mengdi Wang

原文摘要

Language agents increasingly require persistent worlds in which they can act, remember, and learn. Existing approaches sit at two extremes: conventional web frameworks provide reliable but fixed contexts backed by databases, while fully generative world models aim for unlimited environments, but the world is constructed primarily through generation, making it harder to maintain a fixed, deterministic global framework and reducing controllability. In this work, we introduce the Web World Model (WWM), a middle ground where world state and ``physics'' are implemented in ordinary web code to ensure logical consistency, while large language models generate context, narratives, and high-level decisions on top of this structured latent state. We build a suite of WWMs on a realistic web stack, including an infinite travel atlas grounded in real geography, fictional galaxy explorers, web-scale encyclopedic and narrative worlds, and simulation- and game-like environments. Across these systems, we identify practical design principles for WWMs: separating code-defined rules from model-driven imagination, representing latent state as typed web interfaces, and utilizing deterministic generation to achieve unlimited but structured exploration. Our results suggest that web stacks themselves can serve as a scalable substrate for world models, enabling controllable yet open-ended environments.

cs.AI · cs.CL · cs.CV

版本更新 · v2arXiv:2611.07346

GreedTok Is a 0.3287-Approximation for Tokenization, Within 3% of the Best Possible Constant

Kundan Baliga, Aiden Xie, Aditya Byju

原文摘要

A tokenizer's vocabulary is chosen once and used on every input, so it matters how much compression a greedy vocabulary can lose. GreedTok (Lim et al., 2025) picks $k$ tokens one at a time, each time the string that saves the most tokens; its worst-case ratio was open. We prove that GreedTok saves at least 0.3287 of the optimal number of tokens saved, on every corpus and budget, while junction corpora show that no constant above 0.3387 is possible. The lower-bound proof is written out in full and uses no computer certificates. Its key idea is to track the order of greedy's choices: given that order, the pattern greedy leaves inside an optimal token depends on two numbers, and an induction peels off one chosen string at a time; without the order, a needed inequality fails for a 20-character token. Over 256 symbols, GreedTok's ratio exceeds 0.33861 for every $k\ge 104{,}910$, and over $k^c$ symbols with $c<1/2$ it is at least 0.4386 in the limit. We also show that the objective is APX-hard: it has no PTAS unless P = NP.

cs.DS

版本更新 · v2arXiv:2610.19730

Design and Analysis of Temporal Representations in Neural Networks

Alban Colas-Gattepaille, Alexandre Muzy

原文摘要

Neural networks, spiking or not, compute with amplitudes: activation values, spike counts, firing rates, never event times. No measure exists of how much a network uses time. We propose one, the degree of temporal representation: the fraction of a network's four functions (encoding, activation, decoding, learning) whose variables or parameters are temporal, and hypothesise that it predicts performance on temporal tasks. At fixed architecture, accuracy and energy efficiency increase with the degree: a delay network of maximal degree reaches 98.94\% on Google Speech Commands with 175k parameters, above transformers. To test whether networks of different degree really differ, we convert the delay network into artificial and spiking networks with the same inputs and outputs. From this common start, the artificial one learns differently and temporal gradients depend on event times. Hence we obtain a design rule: for temporal data, make event times variables of a network's functions. Spiking networks, converted after training, keep the delay network's accuracy, unreachable by spiking networks trained from scratch.

cs.NE

版本更新 · v3arXiv:2605.13644

Learning Equilibria in Coordination Games via Minorization-Maximization

Ashok Krishnan K. S., Helene Le Cadre, Ana Busic

原文摘要

This paper considers games where the utilities for agents are the sum of a term proportional to a social utility, and another term that is an individual cost or reward. The agents are assumed to be irrational in their perception of the individual cost or reward. The multi equilibrium game is regularized, and its strictly concave potential function is used to select a unique equilibrium. This selected equilibrium is shown to be an $ε-$equilibrium of the original game, where $ε$ is parametrized by the regularizing function. A minorization-maximization based iterative learning scheme is proposed to learn equilibria in this game. This scheme converges to the potential-optimal equilibrium, and has superior convergence behavior in comparison to gradient and best response methods.

cs.GT

新论文 · v1arXiv:2611.12545

Teamknitt: Configuring, Observing and Intervening in LLM-Agent Teams through the Lens of Dynamical-Systems and Human-Team Theory

Goran Mitevski

原文摘要

Most frameworks for teams of large language model (LLM) agents fix the team's shape inside the software: who talks to whom, how much is passed along, and what a human can do once a run has started. Human-team research has long asked those questions: where team phenomena live (in members or in their interaction), how team-level properties arise, and what separates what a team does from how a team is. I propose using that theory as the organizing lens for configuring, observing and intervening in LLM-agent teams, without assuming its findings transfer. I introduce teamknitt, a framework and open reference implementation in which a user configures a team of 1-16 AI agents for task completion, solution generation, achieving a goal or a research mission; selects one of fourteen deterministic routing topologies or a custom directed graph; chooses a sharing profile that limits what may travel along permitted routes; and binds sources, representations, work dependencies, handoffs and an expertise directory. The recorded process can be observed through non-system views (member reports and behavior) and system views (distributed problem-solving, differentiated transactive memory, interactive coordination, team dynamics). Interventions are explicit and human-led: reviewed reconfiguration between runs, planned disturbances, and private feedback, pause, resume or stop applied at dependency-wave boundaries, with "queued", "applied", "assembled", "understood" and "improved" kept as separate claims. Emergent states such as trust and shared understanding are declared unmeasured. This is an idea and system-description paper with falsifiable hypotheses on topology, information flow and intervention timing, including whether the fractal scaling of communication output between members, and between members on different models, relates to task and team performance. No comparative result is reported.

physics.soc-ph · cs.MA

版本更新 · v2arXiv:2508.03471

Learned Adaptive Indexing with Updatable Learned Index

Suvam Kumar Das, Suprio Ray

原文摘要

Indexes can significantly improve search performance in databases. However, if the query workload changes frequently or new data updates occur continuously, it may not be worthwhile to build a conventional index upfront for query processing. Building a conventional fine-grained index, like B+-tree on the entire dataset can be very expensive since it may involve processing terabytes to petabytes of data. Moreover, with a traditional indexing, data ingestion tends to slow down in case of heavy updates as the index also needs to be updated. In these cases, adaptive indexing is a viable solution instead of traditional indexing. Adaptive indexing is a technique in which an index gets built on the fly as a byproduct of query processing. In recent years, research has taken on a new direction where machine learning models are employed for the purpose of indexing. These indexes, known as learned indexes, are more efficient compared to traditional indexes in terms of memory footprints and query performance. To the best of our knowledge, no learned indexes exist yet for adaptive indexing. We propose a novel learned approach, called FLAIRS, for adaptive indexing. It is built on the fly as queries are submitted. We also introduce an updatable learned index TULSI that is utilized by FLAIRS for indexing data. FLAIRS can employ a query workload prediction technique that makes future workload projection based on past workload data. We have evaluated FLAIRS against existing adaptive indexes for various query workloads. Our results show that our approach performs better than others in most cases, offering significant improvement in query performance.

cs.DB

版本更新 · v2arXiv:2611.08671

Greedy Byte-Pair Encoding Achieves 0.391 of the Optimum, Past the One-Third Barrier for Packing-Based Analyses

Kundan Baliga, Aiden Xie, Aditya Byju

原文摘要

Byte-Pair Encoding (BPE) builds most language-model tokenisers by repeatedly merging a most frequent adjacent pair of symbols. How much shortening can this greedy rule lose against the best $k$ merges? Kozma and Voderholzer proved that greedy achieves at least one third of the optimal shortening; the best known upper bound is $(1-e^{-2})/2\approx 0.432$. We show that one third is a barrier for packing-based analyses, which see the optimum only through a pair packing, and we prove that greedy achieves at least $0.391$ of the optimum on every input, with any tie-breaking. The key idea is rule determinism: all copies of a rule of the optimum spell the same letters, so when one greedy replacement destroys three of the optimum's merge points, half a unit can be refunded on each side. A weaker bound, $0.374$, has a short proof that needs no computer; $0.391$ adds a certificate in exact arithmetic. We also show that the worst-case ratio is exactly $3/4$ for two merges and at most $7/12$ for three.

cs.DS · cs.CL

新论文 · v1arXiv:2611.12544

Learning inexact alternating minimization for convex problems with quadratic coupling

Paul Häusner, Jevgenija Rudzusika, Jens Sjölund, Ozan Öktem

原文摘要

Alternating minimization is a standard tool for optimization problems with two coupled variable blocks. Its cost is dominated by the inner subproblem solvers, which rarely have a closed-form solution. In this paper, we propose to learn an inexact alternating scheme in which neural networks approximate the subproblem solution at each iteration. For convex problems with quadratic couplings between the variable blocks that satisfy a quadratic growth condition, we derive a worst-case bound on the optimality gap that exposes the per-iteration error that each network is trained to minimize in a greedy fashion. Combining the bound with a PAC-Bayes argument, we obtain a fixed-horizon, high-probability bound on the expected optimality gap over the distribution of problem instances. We apply the learned scheme to robust matrix completion and low-dose CT reconstruction with dictionary regularization where we achieve a 5x-13x speedup over the inexact first-order baseline at matched accuracy. A brief end-to-end fine-tuning stage extends this up to 30x and outperforms other learned schemes. The code is available at https://github.com/paulhausner/learned-alternating-minimization.

math.OC · cs.LG

版本更新 · v2arXiv:2606.18634

EffiNav: Fusing Depth and Vision-Language for Efficient Object Goal Navigation

Zecheng Yin, Benedict Jun Ma

原文摘要

To locate a target object while exploring the unknown environment is a fundamental capability for autonomous agents, with applications ranging from search-and-rescue to field robots. A simplified version of such task is Object Goal Navigation (ObjNav). In ObjNav, successful arrival at the target object provides a basic measure of performance; however, the efficiency of the navigation trajectory is equally important, as it indicates how intelligently the agent explores and how much time remains for subsequent tasks. In unknown environments, the key to efficient navigation lies in deciding where to explore next. While many prior works aim to address this core challenge and achieved promising performance in certain settings, recent training-based models and non-training frameworks still suffer from generalization and efficiency issues respectively, which in the worst cases can lead to excessive exploration of already-visited areas or redundant back-and-forth motion. We evaluate EffiNav on two widely used simulation benchmarks Habitat Matterport 3D (HM3D) and Open-Vocabulary Object goal Navigation (OVON), and further validate its effectiveness on physical robots in real-world settings. We conduct failure analysis on massive simulation episodes. With minimal modification, we also extend EffiNav to a memory-augmented ObjNav task on the GOAT-BENCH dataset, demonstrating its adaptability beyond standard ObjNav settings. Across two standard metrics--Success Rate (SR) and Success weighted by Path Length (SPL), EffiNav matches or outperforms recent baselines, reflecting its efficiency, robustness, and practical applicability. Recognizing the different emphases of the two datasets, the performances reveals this framework is more balanced and generalizable for efficient ObjNav.

cs.RO · cs.AI

版本更新 · v2arXiv:2609.02302

Improving Evaluation Realism with Inference-Time Compute and Deployment Scaffolds

Axel Ahlqvist, Richard Guan, Juan-Pablo Rivera, Adeline Kassler, Dmitrii Troitskii, Alexandra Souly, Kai Fronsdal, Robert Kirk 等 9 位作者

原文摘要

A core obstacle to alignment evaluation is evaluation awareness: capable models can tell when they are being tested rather than deployed, weakening the conclusions a safety evaluation can support. We present two techniques that make simulated alignment evaluations harder to distinguish from real deployments. Our first technique, critique refinement, spends additional inference-time compute on each simulator action: the simulator generates multiple candidate actions, refines them using feedback from an instance of the target model on how to make them more realistic, and continues the evaluation with the most deployment-like candidate. Our second technique, DISH (Deployment-Imitating SWE-Agent Harness), wraps the target in an agent harness, reducing the gap between simulated and real deployment environments in coding settings. We test the techniques on multiple target models and find that applying both yields larger realism gains than either alone. We show that automated approaches can improve the realism of alignment evaluations, and that they compare favorably to spending the extra compute on more audit attempts.

cs.AI · cs.CL · cs.LG

新论文 · v1arXiv:2611.12539

Dynamic Graph Indexing for Label-Filtered Approximate Nearest Neighbor Search

Mengxu Jiang, Zhi Yang, Fangyuan Zhang, Yin Yang, Sibo Wang

原文摘要

Approximate nearest neighbor search (ANNS) over high-dimensional vectors is a core primitive in modern retrieval systems. In practice, objects are often also associated with labels (e.g., attributes or keywords), and retrieval must combine vector similarity with explicit label-based filtering, giving rise to label-filtered ANNS. This task is challenging because label filtering and ANNS favor very different index structures, selective filters fragment the admissible search space, and updates complicate index maintenance. To address these challenges, we propose DYLAN, a dynamic composite index for label-filtered ANNS supporting containment, overlap, and their conjunctions and disjunctions. DYLAN organizes label sets by shared prefixes in a frequency-ordered trie and selectively shares HNSW-style graph indices across trie-induced subsets, so that each trie region has a constant-factor covering index while each object participates in only logarithmically many indices. To mitigate fragmentation, DYLAN performs a cross-region multi-graph search that exploits ancestor-level indices as bridges between otherwise isolated trie regions. To support insertions, it adopts a threshold-based maintenance scheme that rebuilds shared indices only when necessary, yielding low amortized update cost. Extensive experiments with real data show that DYLAN consistently improves query throughput at comparable recall, by up to two orders of magnitude on challenging label-rich datasets, while retaining competitive memory, construction, and update costs.

cs.DB

版本更新 · v2arXiv:2511.00203

Diffusion LLMs are Natural Adversaries for any LLM

David Lüdke, Tom Wollschläger, Paul Ungermann, Stephan Günnemann, Leo Schwinn

原文摘要

We introduce a novel framework that transforms the resource-intensive (adversarial) prompt optimization problem into an \emph{efficient, amortized inference task}. Our core insight is that pretrained, non-autoregressive generative LLMs, such as Diffusion LLMs, which model the joint distribution over prompt-response pairs, can serve as powerful surrogates for prompt search. This approach enables direct conditional generation of prompts, effectively replacing costly discrete optimization with a small number of parallelizable samples. We provide a probabilistic analysis demonstrating that under mild fidelity assumptions, only a few conditional samples are required to recover high-reward (harmful) prompts. Empirically, our method substantially outperforms prior attacks in both success rate and compute cost, producing low-perplexity, diverse jailbreaks that transfer to a wide range of black-box target models, including robustly trained and proprietary LLMs. Beyond adversarial prompting, our framework opens new directions for red teaming, automated prompt optimization, and leveraging emerging Flow- and Diffusion-based LLMs.

cs.LG · stat.ML

版本更新 · v2arXiv:2610.19732

Controlling the latency in temporal coded spiking models for neuromorphic datasets

Alban Gattepaille, Alexandre Muzy

原文摘要

Spiking neural networks (SNNs) promise energy-efficient, event-driven computation on neuromorphic hardware, where response time, or latency, matters as much as energy for real-time applications. Time-to-first-spike (TTFS) coding with learnable synaptic delays is a natural way to exploit the timing of events. However, in state-of-the-art spiking neuron models, spike times emerge from nonlinear membrane dynamics and thresholds: no closed-form expression links network parameters to spike times, so latency can only be measured empirically, not specified or guaranteed at design time. Moreover, TTFS models are mostly evaluated on static datasets, which hides their latency on truly temporal data. Our hypothesis is that, if spike times are computed directly as closed-form functions of presynaptic spike times and synaptic delays, bypassing membrane potential and thresholding, latency becomes an explicit function of the learnable parameters that can be bounded, and therefore controlled, analytically. We build on this hypothesis with the Delay Neural Network (DeNN) in real time, which turns event streams into successive single-spike images, computes each spike time from standardised inter-spike intervals and synaptic delays, and links successive images through a long-term memory. We prove lower and upper bounds on the latency of every neuron of a layer, and show that bounding the synaptic parameters after training with a single hyperparameter places the latency anywhere between these bounds, without retraining. On a ten-class subset of the Google Speech Commands dataset, this post-training control exposes the accuracy--latency trade-off: a three-hidden-layer network reaches 94.25\% accuracy while deciding, on average, 477~ms after the end of the one-second input, and about 250~ms with confidence-based early decisions. Learning is also shown to concentrate the long-term memory on a few neurons. Latency thus becomes a design parameter of spiking models rather than a quantity measured after the fact, a step towards neuromorphic systems with real-time guarantees.

cs.NE

新论文 · v1arXiv:2611.12532

ML Models Are Not Ciphers: Power Analysis for Weight Extraction Is Harder Than You Think

Mohammed Qaiss, Peter Horvath, Michal Shagam, Lejla Batina, Zhuoran Liu, Eyal Ronen, Mahmood Sharif, Yuval Yarom

原文摘要

Can power-analysis techniques reveal secret weights of machine-learning models? Multiple recent works demonstrate proof-of-concept attacks that can partially recover such weights. To that end, these works build on techniques from cryptographic engineering, treating the model as a ``cipher'' and its secret weights as keys. We question the basic assumption behind such straightforward adaptations of cryptographic attacks and highlight the challenges arising when applying past techniques to fully recover all layers of deep neural networks. Unlike past works, which extract only a few weights, typically from shallow layers, we investigate the recovery of complete networks. Moreover, we target fully trained and functional models, rather than untrained, randomly initialized networks. To focus on fundamental challenges, we assume an unrealistically strong side-channel adversary that can directly observe the output of the leakage model without any added noise. With this adversary, we show that machine-learning models differ significantly from cryptographic implementations, giving rise to unique challenges that have not been observed in the prior art. We observe two main differences between machine-learning models and ciphers. First, in contrast to the high-diffusion nonlinear integer arithmetic prevalent in cryptographic schemes, machine-learning models tend to use limited-precision floating-point arithmetic and low-diffusion linear operations. We show that these operations introduce difficulties in weight recovery, often requiring an unexpectedly large number of traces, even under ideal channel conditions. Second, we observe that, unlike ciphers, which aim to increase the data entropy, machine-learning models aim to reduce it. This reduces data variability as the algorithm progresses, hampering weight recovery in deep layers of the model. (details can be found in the paper)

cs.CR

新论文 · v1arXiv:2611.12530

K-prop: Deep Online Learning by Backpropagating Temporal Kernels

Bingkun Liu, Paul Haider, Federico Benitez, Mihai A. Petrovici, Walter Senn

原文摘要

Self-recurrence represents an important component for temporal processing in neural networks, both artificial and biological, where it appears naturally in the form of leaky integration. However, due to the complex temporal dependencies of the gradient, an efficient and effective method of online training is still lacking. Here we propose K-prop, a novel online learning method for leaky neural networks. It dramatically reduces the computational and memory cost of real time recurrent learning (RTRL) by propagating kernel gains. K-prop also outperforms existing RTRL approximations and achieves performance comparable to backpropagation through time in sequence learning, classification, and reinforcement learning tasks. We further analyze how its sole approximation affects gradient quality and identify conditions under which K-prop remains accurate and effective.

cs.LG · q-bio.NC

新论文 · v1arXiv:2611.12529

Perceived Agency During Agent-Avatar Control Transitions in Virtual Reality

Jingyi Zhang, Anthony Steed

原文摘要

Prior beliefs about whether a virtual human (VH) is controlled by a human or a computer can shape how users interpret and respond to its behaviour. As emerging hybrid VH systems allow control to transfer between human operators and computer programs during an ongoing interaction, it remains unclear whether perceived agency updates with the actual controller or continues to reflect users' prior beliefs and first impressions. We investigated this question in a within-subject VR study with 24 participants across four conditions: always computer-controlled, computer-to-human, human-to-computer, and always human-controlled. We designed an interaction task in which participants acted as interviewers, asking the VHs a fixed set of questions. Before each interaction, participants were informed whether the VH was human- or computer-controlled, matching the actual controller at the beginning, but were unaware of possible control transitions. We found that perceived agency was strongly influenced by the controller presented at the beginning of the interaction, with no evidence of updating following the control transitions. Social evaluations including likeability and co-presence were generally higher in conditions where human control occurred later and covered a greater proportion of the interaction. The results suggest that prior beliefs and first impressions continued to shape perceived agency, whereas social evaluations were more sensitive to the VH's actual behaviour over the interaction.

cs.HC · cs.AI