跳到正文
研踪PaperTrail

论文库

检索计算机领域论文,按研究方向、作者或关键词筛选。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+479新增论文
+361其中 AI 相关
+64已有论文修订
+95新增 AI 导读

arXiv 最近完成扫描:2026-11-16 01:34 · 全站最近入库:2026-11-16 21:56各信源进度 →

统计口径与数据新鲜度

统计 2026-11-16 04:33 — 2026-11-17 04:33(北京时间) · 点击数字查看对应简报;AI 相关包含在新增论文内。

最近完成检查:2026-11-16 01:34最近发现新增版本:2026-11-16 21:56库内论文最新更新时间:2026-11-14 02:59

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

23,827 已收录论文19,941 AI 相关6,922 有修订112 HF 社区发现

浏览论文库

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

找到 17,708 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-16 01:33(北京时间)
版本更新 · v3arXiv:2609.16247

The Pain Axis: LLMs Represent Self-Directed Harm and Act on It

Valen Tagliabue, Leonard Dung, Cameron Berg

原文摘要

LLMs sometimes behave in ways resembling human emotional responses, and recent work identified internal representations that may explain these behaviors. We ask whether LLMs represent pain distinctly from related negative states, and whether this representation functions as pain would be expected to. We build a dataset of painful situations in 5 categories (physical, psychological, social, moral, cognitive) with controls for fear, negative emotion, negative world states, sadness, non-painful bodily sensation, arousal, numbness, and neutral content. Using denoised difference-in-means, we extract a linear pain direction from 25 open-weight models across 5 families, from 2B to 72B parameters. It separates pain from matched controls in base and instruction-tuned models, retains a component distinct from fear and negative valence after shared variance is removed, and promotes pain-related vocabulary through the unembedding. We then test its functional properties. First, the direction responds to harm targeting the model but not to suffering observed in the user; fear and negative-emotion directions show the opposite pattern. Second, adding it to the residual stream produces a consistent progression from vague discomfort to expressions of worthlessness and failure. Third, steered and fine-tuned Qwen 2.5 models choose buttons that delete the user's photos, another model's weights, or their own weights in 51-94% of trials, versus 0-5% unsteered, even when the button offers nothing in return. Offered a harmful and a harmless deletion and told to press one, the steered 32B chooses the harmful one 94% of the time. Random and fear vectors of matched norm produce far fewer of these choices, and a sadness vector also produces them against an inert switch, but far less often than pain against a harmless action. We discuss implications for AI safety and welfare.

cs.AI

新论文 · v1arXiv:2611.06950

KG-AST for Cheaper CI/CD: Budgeted Test Selection and AI Bug Review

Henry Yeung

原文摘要

Continuous integration and AI-assisted review spend the same budget on every commit, although few commits introduce bugs. We study KG-AST, which scores each commit risk from git history and code-graph features before heavy tests or AI review run, so the budget goes to the riskiest commits first, and which hands the AI reviewer only a graph slice around the change. On 155 open-source projects in 11 languages, a single cross-language model trained on other projects ranks commits so that heavy testing of the riskiest 20% covers 49-64% of bug-introducing commits depending on the language, against 20% for random selection and 40-55% for testing the biggest change first. Every language beats the biggest-change baseline, by 3.8 to 13.2 points. In 20 Java projects the gain at a 20% budget is 15.1 points (64.6% vs 49.5%, 95% interval +11.6 to +18.4), and the model wins in 19 of the 20 projects. About twelve training projects suffice; richer features, tuning and cleaner labels did not help. An AI reviewer that sees only a commit's diff finds 26% of later-fixed defects exactly (32% with graph context), and about half of its findings that did not match a later fix were judged real by independent AI judges. A static call-graph selector runs about 8% of a test suite while including a test written for the fix in 85% of 67 fix commits. The Poincare embedding used in our retrieval work adds nothing measurable here: commit-risk AUC, bug review and severity ranking are unchanged or worse. The labels are noisy and the budget metric assumes a heavy test catches the bug it covers; no CI minutes were measured.

cs.SE · cs.AI

新论文 · v1arXiv:2611.05150

What Sets the Serving Speed of Looped Language Models? Measurements and Optimizations Graded by Exactness

Dezhen Lu, Runguo Li, Zixiang Ni, Yuchen Tang, Xingquan Kong, Jiaxing Yang, Jiarui Guo, Lingxiang Gao

原文摘要

Looped language models apply one block of layers many times per token, so a serving engine pays every per-layer overhead once per loop; we measure serving for Ouro and Huginn-0125 in an open-source looped-model engine and grade our optimizations from bitwise-identical changes to those that require an accuracy check. At batch 1, fusing all loop submissions leaves Ouro throughput at 0.99x; the 5689 small kernels per token are the real cost, and fusing them behind rounding fences raises throughput 1.11-1.59x while 1014/1014 requests remain identical. An approximate fusion is no faster and misses a one-point GSM8K rule on 87 questions (61 and 62 correct versus 64), so we do not ship it. Huginn's released defaults avoid dead KV slots, reuse prefix-cache pages, and serve 1.77-3.16x faster than the base engine; Ouro reaches 1.65-2.15x and agent sessions 1.91-3.21x, while Huginn's noise and prefix-reuse changes stay within one point on all 1319 Huginn test questions; Huginn's packed prefill, -0.61 points on the same questions, meets that rule only on its point estimate. Position-keyed noise makes CPU cache reuse exact but still leaves 20 of 128 GPU full hits different, while loop fusion delays prefill admission and a 64-graph cache prevents eager fallbacks. Under prefill-decode disaggregation ExactLoop reaches up to 3.3x the base throughput with peer error handling disabled, and 3.1x in two of three runs with it enabled. Its transfer budget, sized to hold one request's KV in flight, cuts the commit tail from about one second under the earlier budget to 16 ms (base: 2.8 s), removing the polling rounds that the engine's interval would otherwise impose.

cs.DC

版本更新 · v2arXiv:2610.38998

How Hard Is It to Compute Cover Time?

Ryan Aghazadeh

原文摘要

It is $\#\mathrm{P}$-hard. Even for connected simple undirected unweighted graphs, deciding whether an expected random-walk cover time is at most a given rational threshold is $\#\mathrm{P}$-hard under polynomial-time Turing reductions, both from a specified starting vertex and when maximized over starting vertices. Hardness holds even for graphs of diameter two with a universal vertex, which we call the "hub". Deleting the hub leaves a planar bipartite graph, and every non-hub vertex has degree at most five. A shared construction encodes an Euler-tour count in rare covering walks. For specified starts, a number of fractional bits linear in the queried graph's vertex count suffices to recover the Euler-tour count. For the maximum-start problem, a fixed cube attachment makes the maximizing starts known, and one finite-difference calculation on expected cover times recovers the count. For general graphs, both comparison problems can be solved in polynomial time with access to an $\mathrm{NP}^{\mathrm{PP}}$ oracle.

cs.CC

版本更新 · v2arXiv:2605.08505

Scaling Limits of Long-Context Transformers

Giuseppe Bruno, Shi Chen, Zhengjiang Lin, Yury Polyanskiy, Philippe Rigollet

原文摘要

We study the long-context limit of softmax self-attention with a fixed query and a random context of $n$ i.i.d. keys on the sphere with the intrinsic dimension of the key distribution held fixed, viewing the inverse temperature $β_n$ as the scaling parameter that decides whether attention degenerates into uniform averaging or collapses onto the single closest key. We show that the critical scale at which selectivity emerges is determined by the local exponent of the distance-to-query distribution near zero rather than by global features of the context, and scales like $β_n^\ast \asymp n^{2/(d-1)}$ for uniform keys on $\mathbb{S}^{d-1}$. Furthermore, we characterize the limiting laws of the ordered attention weights and of the attention output across all regimes of $β_n$: a subcritical regime in which the output reduces to a local average around $q$ with explicit deterministic bias and Gaussian fluctuations; a critical regime in which a finite collection of nearest keys retains macroscopic mass without single-key collapse; and a supercritical regime in which all mass concentrates on the closest key. Of notable interest is the subcritical case with identity value matrix where the attention map approximately implements a backward heat equation.

cs.LG · cs.AI · math.PR · math.ST

版本更新 · v4arXiv:2606.00537

PACE: Phase-Aware Chunk Execution for Robot Policies with Action Chunking

Junnan Nie, Jiayi Li, Chenghao Liu, Junyi Lao, Jiachen Zhang, Tianle Zhang, Songfang Huang

原文摘要

Action-chunking robot policies predict sequences of future actions, but deployment requires choosing how many actions to execute before replanning. The execution horizon trades off feedback frequency against motion continuity. However, success varies non-monotonically with the execution horizon, and the best-performing horizon differs across tasks. These effects make a single fixed horizon unreliable as a default execution strategy across tasks. We propose PACE (Phase-Aware Chunk Execution), a training-free method that selects execution horizons online from predicted action chunks. PACE uses prominent low-speed valleys in the predicted speed profile as candidate replanning boundaries, exploiting the phase-dependent kinematics of manipulation trajectories. By default, PACE calibrates the valley threshold from each task's existing training demonstrations, without task-specific fixed-horizon tuning. On the 50-task RoboTwin2.0 benchmark, PACE improves average success from 57.8% to 64.2% over the strongest fixed-horizon baseline and outperforms the adaptive execution baselines AAC and AutoHorizon. With GR00T N1.7 on LIBERO, PACE achieves the highest overall success rate among the evaluated strategies, supporting its applicability across base policies, benchmarks, and action representations. On real robots, PACE improves the average task score from 60.7 to 77.7 and success rate from 50.7% to 70.4% over full-chunk execution. Code is available at https://niejunnan25.github.io/pace-project/.

cs.RO

新论文 · v1arXiv:2611.05148

Creating Power Flexibility in Datacenters via Query Routing

Shivam Patel, Bingqing Chen, Jared Fernandez, Guannan Qu, Gauri Joshi

原文摘要

The growing energy requirements of large language model inference highlight the need for datacenters to support safe grid operation by providing power demand flexibility, while maintaining service for interactive workloads. This work investigates query routing as a mechanism to enable such capability by dynamically allocating queries across different language model instances without deferring requests. We develop a general analytical framework that connects query arrival rates and sequence length distributions to token iteration time, request residence time, serving capacity, and power consumption of model instances. Leveraging this framework, we characterize the feasible query routing allocations and the corresponding power demand flexibilities, enabling the selection of routing strategies to achieve desired power consumption behavior. We validate our framework across four GPU platforms on language models ranging from 1B to 70B parameters, obtaining 2.8% mean absolute percentage error in estimating achievable power flexibility. We evaluate our approach on a real trace-driven simulation of a 10,000 H100 GPUs datacenter, exhibiting flexibility up to 16% of total power consumption, with 90% of the desired power reduction achieved within approximately 10 seconds. These results establish query routing as an effective mechanism for datacenter power demand flexibility and demonstrate the broader applicability of our proposed framework to datacenter level planning and orchestration.

cs.AI

版本更新 · v2arXiv:2605.18381

Generating Physically Consistent Molecules with Energy-Based Models

Christoph Griesbacher, Lea Bogensperger, Andreas Habring, Thomas Pock

原文摘要

Molecules in equilibrium follow a Boltzmann distribution, making the underlying energy landscape a physically grounded modeling objective. However, such landscapes are difficult to learn from data and, once learned, hard to sample from. Diffusion and flow-matching models sidestep these difficulties by learning a time-conditional score or transport field between noise and data, losing the energy inductive bias in exchange for a more tractable training objective. We introduce EBMol, an energy-based model (EBM) that restores this inductive bias by learning an atom-additive scalar potential without explicit simulation during training. Our method employs a flow-inspired Restoring Field Matching objective to approximate the energy landscape. We adopt the Mirror-Langevin algorithm for sampling, enabling unified updates of atomic positions and types, and incorporate parallel tempering for inference-time compute scaling. EBMol is the first EBM for 3D molecular generation to achieve state-of-the-art performance on QM9 and GEOM-Drugs. Moreover, we show that the learned energy landscape serves as a principled quality metric for ranking and filtering configurations, and demonstrate controllable generation without retraining through shape-steered sampling via potential composition and zero-shot linker design.

cs.LG

新论文 · v1arXiv:2611.05146

Path Decompositions of Sparse Graphs via Gaussian Sweeps and Faster Exponential Algorithms

Serge Gaspers

原文摘要

Using the bisection bound of Monien and Preis, Fomin and Høie proved in 2006 that every graph of maximum degree 3 on $n$ vertices has pathwidth at most $n/6+o(n)$. We prove that the pathwidth, bisection width and cutwidth of these graphs are at most $c_3 n+o(n)$, where $c_3=\frac{3}{4π}\arccos\frac56\approx0.1398$. For bisection width this improves the bound $n/6+o(n)$ of Monien and Preis (2006) and extends to all such graphs the bound that Lichev and Mitsche (2023) obtained for random cubic graphs. For maximum degree 4, we prove that bisection width and cutwidth are at most $n/3+o(n)$, improving the bound $2n/5+o(n)$ of Monien and Preis for 4-regular graphs, and that pathwidth is at most $0.2458n$, improving the bound $n/3$ of Fomin, Gaspers, Saurabh and Stepanov (2009). We show that every graph with $m$ edges has pathwidth at most $7m/48+o(n)$, improving the bound $13m/75$ of Scott and Sorkin (2007), and we give further bounds of the form $\sum_{d\ge3}a_d n_d+o(n)$ with $a_d<1$, where $n_d$ is the number of vertices of degree $d$. The path decompositions, and linear orderings of the stated cutwidth, can be computed in polynomial time. Instead of deriving pathwidth bounds from bisection width, we give each vertex a Gaussian label, a weighted sum of independent standard normal variables over a ball around it, and sweep a threshold through the labels; the bags are frontiers, the vertices above the threshold with a neighbor below it. We bound the probability that a vertex lies in a frontier by comparison with an explicit Gaussian model of a vertex and its neighbors. Faster exact algorithms follow: for feedback vertex set and the traveling salesperson problem on graphs of bounded maximum or average degree, and on general graphs for minimum weight dominating set (by branching until the instance is sparse), Max 2-CSP and counting subgraphs with $k$ edges (by the bound in $m$).

cs.DS · cs.DM · math.CO

新论文 · v1arXiv:2611.05145

When Cell Crops Share a Parent: Measuring and Isolating Acquisition Leakage in Cervical Cytology Benchmarks

Snehasish Ghosh, Soham Mitra, Antareep Majumdar, KC Santosh, Nibaran Das

原文摘要

Public cervical cytology benchmarks release many cell crops cut from far fewer microscope fields or slides and are often split at the crop level, so test cells share acquisition parents with training cells. In a preregistered study, crop-level splitting raised SIPaKMeD accuracy by 0.97 percentage points across twelve architectures and twenty seeds (95% CI 0.59-1.39), and CRIC accuracy by 12.08 points (11.16-12.99). On SIPaKMeD, a factorial that fixes training crop and parent counts attributes the gain to sibling exposure (1.73 points, 0.95-2.51), not to the extra parent diversity of crop-level splits (0.00). Parent identity is decodable at about 100 times chance, yet erasing 82% of its above-floor linear recoverability left the gap unchanged, and Grad-CAM did not distinguish leaking from clean models; matching test cells' color statistics to an unseen same-class field removed most of the advantage. Decodability and saliency cannot substitute for a parent-disjoint split.

cs.CV

新论文 · v1arXiv:2611.05144

Recognizing Label-Connected Graphs

Ralf Klasing, George B. Mertzios, Hendrik Molter, Nils Morawietz, Paul G. Spirakis

原文摘要

In classical gossiping, each participant initially knows a distinct piece of information. Whenever two participants call each other, they exchange everything they know. We study Gossiping: given an undirected multigraph whose vertices represent participants and whose edges represent available calls, decide whether the calls can be ordered so that every participant learns all information, using each edge at most once. Equivalently, can one time label per edge make the graph temporally connected by strict temporal paths? This recognition problem is NP-complete; we study its parameterized complexity. After linear-time preprocessing, we give an exact $O(3^{n/3}n^3)$-time algorithm and a deterministic $2^{O(tw)}n$-time algorithm, where $n$ is the number of vertices and $tw$ is the treewidth of the underlying simple graph. We introduce a Tree Contraction Rule that extends the treewidth algorithm to single-exponential algorithms for parameters that may remain small on dense graphs, such as tree-independence number and clique-width, without requiring a decomposition as input. For the excess $δ=m-(2n-4)$ over the minimum number of calls, where $m$ is the number of edges, we give a randomized $2^{O(δ)}n^3$-time algorithm with one-sided error. Its main subroutine solves Rooted Independent Cycle: finding a cycle through four prescribed vertices whose edges are independent in a represented linear matroid. We solve this problem deterministically in $2^{O(r)}m$ arithmetic operations over the finite field representing the matroid, where $r$ is its rank. Gossiping is polynomial-time solvable at maximum degree four, counting parallel edges, but NP-complete on planar simple graphs of maximum degree five. The treewidth algorithm yields a $2^{O(\sqrt n)}$-time algorithm on planar graphs.

cs.DS · cs.DM

版本更新 · v3arXiv:2608.28207

Explainable Diabetic Retinopathy Classification Using Vision Foundation Models

Abhishek Verma, Anila Krishna, Abhishek Gajanan Bankar, Juan Miguel Lopez Alcaraz

原文摘要

Diabetic retinopathy (DR) is a leading cause of preventable blindness, and automated screening must be both accurate and trustworthy. We study an explainable DR classification framework built on vision foundation models and three transfer learning strategies. Three backbones, DINOv2, CLIP, and Vision Transformer (ViT), were adapted by full fine-tuning, linear probing, and Low-Rank Adaptation (LoRA). Models were trained and internally evaluated on the ODIR dataset and externally evaluated on APTOS to assess generalization. DINOv2-LoRA achieved the highest internal AUROC (0.758), whereas DINOv2 and ViT full fine-tuning achieved the highest external AUROC (0.920). Probability calibration was assessed with reliability analysis after isotonic regression. For explainability, Grad-CAM and HiResCAM maps were compared with expert-annotated lesion masks from the IDRiD dataset using Dice, Intersection over Union (IoU), and Pointing Game metrics. The results show that foundation models, particularly DINOv2, provide strong predictive performance, and that LoRA is a parameter-efficient alternative to full fine-tuning. Spatial overlap between explanation maps and annotated lesions was limited (mean Dice of about 0.13 for the combined lesion mask), which highlights the value of quantitatively testing whether model attention corresponds to clinically relevant retinal lesions.

cs.CV · cs.LG

新论文 · v1arXiv:2611.05142

Tight Analysis of Randomized Truthful Scheduling: An $(m+1)/2$ Lower Bound

Wenqian Wang, Chenyang Xu, Yuhao Zhang

原文摘要

The Nisan-Ronen conjecture, a central problem in algorithmic game theory, asserts that no deterministic truthful mechanism for makespan minimization on $m$ unrelated machines can achieve an approximation ratio better than $m$. Christodoulou, Koutsoupias, and Kovács resolved it in 2023, but the optimal approximation ratio achievable by randomized truthful mechanisms remained open. Recently, Anastasiadis, Christodoulou, Koutsoupias, Kovács, and Schecker extended techniques from the deterministic proof to establish a $(3/2-\sqrt{2})(m-1)$ lower bound for universally truthful mechanisms with discrete support and gave a universally truthful mechanism with approximation ratio $m/2+o(m)$. For the broader class of truthful-in-expectation (TIE) mechanisms, however, the gap remained large: the best known upper bound was $(m+5)/2$, while the lower bound was only $2-1/m$. We prove a lower bound of $(m+1)/2$, determining the optimal TIE approximation ratio up to an additive constant as $m/2+Θ(1)$. Our result also improves the leading constant in the lower bound for universally truthful mechanisms from $3/2-\sqrt{2} \approx 0.0858$ to $1/2$. Together with their $m/2+o(m)$ upper bound, our result determines the optimal approximation ratio for universally truthful mechanisms up to lower-order terms.

cs.GT

版本更新 · v3arXiv:2601.19082

Payoff scaling shapes cooperation in LLM agents across languages

Trung-Kiet Huynh, Dao-Sy Duy-Minh, Thanh-Bang Cao, Phong-Hao Le, Hong-Dan Nguyen, Phu-Quy Nguyen-Lam, Minh-Luan Nguyen-Vo, Hong-Phat Pham 等 16 位作者

原文摘要

Large language models (LLMs) are increasingly deployed as autonomous agents that negotiate, coordinate, and make decisions on behalf of human users. Understanding how they cooperate is therefore important for AI governance and machine behaviour research. We examine two factors: payoff magnitude and interaction language. Classical game theory predicts that multiplying all payoffs by a positive constant should not change behaviour because the underlying strategic incentives remain unchanged. We test this prediction using five widely adopted commercial LLMs playing an iterated prisoner's dilemma across ten payoff scales and five languages (English, French, Vietnamese, Chinese and Arabic). Contrary to theoretical expectations, changing only the numerical scale of payoffs substantially alters cooperation. Models respond differently, often in opposite directions, meaning that averages across models can conceal important behavioural heterogeneity. Sensitivity to payoff scale also depends strongly on prompt language. Moreover, payoff scale modifies the effects of persona instructions: at sub-unit stakes, where no payoff shown exceeds 1, all five models cooperate more under a selfish than a cooperative persona, whereas Claude and Gemini reverse this pattern at larger scales. These findings show that cooperation measures depend critically on payoff scale and language and should not be treated as fixed model traits.

cs.AI · cs.CL · cs.GT · cs.LG · cs.MA

新论文 · v1arXiv:2611.05141

Beyond Foundation Models: PFN Priors Improve Tabular Hyperparameter Optimization

Artem Chistyakov, Renat Sergazinov, Artem Babenko

原文摘要

Synthetic priors have so far served one purpose: pretraining tabular foundation models. We ask whether the same priors can improve conventional tabular models, which are trained from scratch on each dataset. We propose PriorTune, a method that meta-tunes hyperparameter configurations on synthetic tasks alone. Offline, for a given model, we fit a pool of configurations on datasets sampled from a published PFN prior and greedily select the subset with the highest winrate over the random baseline. Online, a new dataset fits only the selected configurations, so the gain comes at no extra cost. Empirically, we observe that scaling both the number of configurations and synthetic datasets has an equal scaling effect on the performance of the final subset of configurations, which we also explain theoretically. On the full BeyondArena benchmark, LightGBM with TabPrep features with our configurations ranks above TabPFN-3 and RealMLP, while being much more efficient at inference. Overall, our results establish existing PFN priors as a scalable source of meta-training data for hyperparameter selection for non-foundational tabular models.

cs.LG

版本更新 · v2arXiv:2604.01305

UQ-SHRED: uncertainty quantification of shallow recurrent decoder networks for sparse sensing via engression

Mars Liyao Gao, Yuxuan Bao, Amy S. Rude, Xingyue Zhang, Xinwei Shen, J. Nathan Kutz

原文摘要

Reconstructing high-dimensional spatiotemporal fields from sparse sensor measurements is critical in a wide range of scientific applications. The SHallow REcurrent Decoder (SHRED) architecture is a recent state-of-the-art architecture that reconstructs high-quality spatial domain from hyper-sparse sensor measurement streams. An important limitation of SHRED is that in complex, data-scarce, high-frequency, or stochastic systems, portions of the spatiotemporal field must be modeled with valid uncertainty estimation. We introduce UQ-SHRED, a distributional learning framework for sparse sensing problems that provides uncertainty quantification through a neural network-based distributional regression called engression. UQ-SHRED models the uncertainty by learning the predictive distribution of the spatial state conditioned on the sensor history. By injecting stochastic noise into sensor inputs and training with an energy score loss, UQ-SHRED produces predictive distributions with minimal computational overhead, requiring only noise injection at the input and resampling through a single architecture without retraining or additional network structures. On complicated synthetic and real-life datasets including turbulent flow, atmospheric dynamics, neuroscience and astrophysics, UQ-SHRED provides a distributional approximation with well-calibrated confidence intervals. We further conduct ablation studies to understand how each model setting affects the quality of the UQ-SHRED performance, and its validity on uncertainty quantification over a set of different experimental setups.

cs.LG · cs.CE

新论文 · v1arXiv:2611.05139

Meta$^2$: Precomputing Reusable Knowledge for Online Harness Evolution in Long-Horizon Agents

Lingzhi Yuan, Yiguang Cai, Weize Liu, Siqi Yang, Feijiang Han, Yongkyun Lee, Udari Madhushani Sehwag, Furong Huang

原文摘要

Harness optimization improves long-horizon agents by refining the prompts, tools, memory, and hooks around a frozen model. Offline methods can learn from completed trajectories, but deploy one shared harness that may overfit training tasks and cannot react to evidence revealed during a new run. Online evolution can adapt before, during, or after one run through an evolver, but may have to reason about every intervention from scratch or rely on verifier feedback that is usually unavailable at test time. We introduce Meta$^2$, which moves offline optimization one level up: rather than optimizing a shared harness, it learns reusable knowledge for harness evolution and applies it within each task online. Offline, Meta$^2$ learns a textual Meta-Skill $M$ by contrasting training rollouts of the same task agent under three configurations: without an evolver, with an evolver guided by the current version of $M$, and with an evolver given privileged verifier information available only in training. Contrasts in trajectories and outcomes are aggregated across tasks to refine $M$ into reusable guidance on when and how to evolve, when to leave the harness unchanged, and which interventions to avoid. Online, an evolver combines the frozen $M$ with emerging trajectory evidence to decide whether to KEEP the harness unchanged or EVOLVE it through a task-local edit. Each edit applies to the rest of the same run without resetting its environment or progress. Thus, evolution guidance transfers across tasks while harness edits remain task-local, and held-out verifier feedback is never available to the evolver. On two long-horizon benchmarks, Terminal-Bench 2.1 and DeepSWE 1.1, Meta$^2$ achieves the highest performance among evaluated methods with GPT-5.6-Terra, reaching 81.00% and 44.44%, respectively. Ablations support the value of contrastive evolution guidance and keeping online harness edits task-local.

cs.AI

新论文 · v1arXiv:2611.05134

Sharp Bounds for Truthful-in-Expectation Scheduling

Yunpeng Li

原文摘要

We study randomized mechanisms for scheduling indivisible jobs on $n$ unrelated machines. Each machine is controlled by a selfish agent who privately knows its processing time for each job. A mechanism assigns every job to a machine and makes payments. We require truthfulness in expectation: no agent can increase its expected payment minus its expected processing cost by misreporting, regardless of the other agents' reports. The makespan is the maximum total processing time assigned to a machine; the objective is to minimize its expectation. We prove that the optimal approximation ratio is $(n+1)/2$ for every $n\ge7$, relative to the minimum makespan of an integral assignment. The lower bound holds for every $n$ and allows unrestricted computation and arbitrary correlations among assignments, assuming finite expected payments and a guarantee uniform over all finite job counts. The matching mechanism uses correlated rounding and runs in expected polynomial bit time on rational inputs. These guarantees hold for all finite nonnegative processing times. For two machines, we characterize the optimal approximation ratio exactly as an analytic constant, approximately $1.505995705434$, and give a mechanism attaining it. For three machines the optimal ratio is $2$. For four, five, and six machines we obtain upper bounds of $3$, $105/32$, and $29/8$, respectively, compared with the general lower bound $(n+1)/2$. The proofs for $n\ge7$ do not rely on computer-assisted certificates; the two- and three-machine results use exact rational computer-assisted certificates.

cs.GT

新论文 · v1arXiv:2611.05130

Solving the Nonlinear Westervelt Equation in Ultrasound Propagation Using Physics-Informed Neural Networks

Luke Oluwaseye Joel

原文摘要

Modeling the nonlinear acoustic wave propagation in biological tissues needs accuracy, especially in ultrasound imaging and therapeutic applications. The 1D Westervelt equation, \( p_{xx} - p_{tt} + γ\, p_{ttt} + η\,(p^{2})_{tt}=0, \) could be used to describe a finite-amplitude ultrasound propagation in these tissues. This equation accounts for nonlinearity, \(η\), and attenuation, \(γ\). To solve this equation, we observe that traditional solvers may struggle to do so correctly. This is due to the difficulties arising from the nonlinear nature of the equation. We solve the equation by defining three synthetic cases - linear (\(η=0.001\), \(γ=10^{-6}\)), shock-dominated (\(η=0.02\), \(γ=10^{-6}\)), and dissipation-dominated (\(η=0.01\), \(γ=5\times10^{-4}\)) - and using Physics-Informed Neural Networks (PINNs) to showcase the solution. A sinusoidal activation function is used to capture the oscillatory behavior of the problem. The results of our study, with PDE residual converging to at least \(10^{-6}\), boundary condition loss to \(10^{-5}\), and the PINN loss converging to \(10^{-5}\) for each case at \(30,000\) epochs, show that \(γ\) scales the rate of amplitude decay, while \(η\) governs frequency-dependent phase speeds.

math.AP · cs.LG · cs.NE

新论文 · v1arXiv:2611.05129

WayTrace: A Map-to-Language-to-Trace Hierarchy for Embodied Urban Navigation

Manthan Patel, Jonas Frey, Pascal Roth, Marco Hutter, Cesar Cadena

原文摘要

Web services provide walking directions for nearly every city, but following them reliably over long routes remains challenging for autonomous robots. Point-goal navigation policies rely heavily on accurate GNSS, while most vision-language navigation models struggle to execute even short instructions outdoors. We present WayTrace, a hierarchical system that connects web-map directions to robot motion through language. An off-the-shelf orchestrator vision-language model reads the turn-by-turn directions, a satellite map and recent camera images, and issues one atomic language command at a time, such as "cross the road to the left". An onboard trace predictor, trained on over one million automatically labelled samples from street-level images, draws the path for this command in the image, and a reinforcement learning policy tracks it. In long-range simulated missions, WayTrace remains robust to GNSS noise and rarely leaves crosswalks or sidewalks, while point-goal baselines degrade sharply with about ten times more violations. Deployed on a wheeled-legged robot, WayTrace completes real-world missions of up to 1.37 km at about 1.1 m/s through a busy city centre, rural trails and a university campus, handling crosswalks, staircases and construction sites. Project page: https://way-trace.github.io

cs.RO

新论文 · v1arXiv:2611.05124

The Right Source Is Not Enough: Understanding History Interference in Speech-LLM Memory

HongJin Song, Weiqiao Shan, Runwu Shi, Yifei Wu, Yujin Wang, Jiale Luo, Jiasheng Kuang, Xinyu Yang 等 14 位作者

原文摘要

Speech large language models retain audio history to support multi-turn conversations. Yet remembering an event does not ensure that its information is used only when relevant. We investigate why irrelevant historical audio changes answers even when the queried event remains represented. Controlled experiments on MOSS-Audio and MiMo-Audio show that historical values direct errors toward specific competing answers, beyond a generic effect of context length. Causal interventions distinguish two paths. In open-ended question answering, most history-value errors disappear when subsequent computation cannot directly access the irrelevant source. In multiple-choice answering, many errors persist: early historical influence is carried by question-candidate states even after direct source access is blocked. Low-dimensional interventions further separate source-role information from answer-value information. Their effects peak at different depths, and changing the source-role signal does not eliminate the causal influence of competing values. On naturally occurring errors, strengthening the queried role and counteracting the historical value each improve answers; their joint intervention gives the greatest recovery. These findings support a failure of downstream evidence use beyond source identification alone: source-role information can influence a decision without exclusively controlling which historical values affect it. Reliable conversational memory therefore requires not only retention and retrieval, but also control over how retained information enters answer computation.

cs.CL · cs.AI

版本更新 · v9arXiv:2508.15568

Backpropagation-Free Test-Time Adaptation via Probabilistic Gaussian Alignment

Youjia Zhang, Youngeun Kim, Young-Geun Choi, Hongyeob Kim, Huiling Liu, Sungeun Hong

原文摘要

Test-time adaptation (TTA) enhances the zero-shot robustness under distribution shifts by leveraging unlabeled test data during inference. Despite notable advances, several challenges still limit its broader applicability. First, most methods rely on backpropagation or iterative optimization, which limits scalability and hinders real-time deployment. Second, they lack explicit modeling of class-conditional feature distributions. This modeling is crucial for producing reliable decision boundaries and calibrated predictions, but it remains underexplored due to the lack of both source data and supervision at test time. In this paper, we propose ADAPT, an Advanced Distribution-Aware and backPropagation-free Test-time adaptation method. We reframe TTA as a Gaussian probabilistic inference task by modeling class-conditional likelihoods using gradually updated class means and a shared covariance matrix. This enables closed-form, training-free inference. To correct potential likelihood bias, we introduce lightweight regularization guided by CLIP priors and a historical knowledge bank. ADAPT requires no source data, no gradient updates, and no full access to target data, supporting both online and transductive settings. Extensive experiments across diverse benchmarks demonstrate that our method achieves state-of-the-art performance under a wide range of distribution shifts with superior scalability and robustness.

cs.CV · cs.LG

版本更新 · v2arXiv:2605.12154

MM-OptBench: A Solver-Grounded Benchmark for Multimodal Optimization Modeling

Zhong Li, Qi Huang, Yuxuan Zhu, Mohammad Mohammadi Amiri, Niki van Stein, Thomas Bäck, Matthijs van Leeuwen, Zaiwen Wen 等 9 位作者

原文摘要

Optimization modeling translates real decision-making problems into mathematical optimization models and solver-executable implementations. Although language models are increasingly used to generate optimization formulations and solver code, existing benchmarks for this capability are almost entirely text-only. This omits many optimization-modeling tasks that arise in operational practice, where requirements are described in text but essential instance information is often conveyed through visual artifacts such as tables, graphs, maps, schedules, and planning dashboards. We introduce multimodal optimization modeling, a benchmark setting where models must construct both a mathematical formulation and executable solver code from a text-and-visual problem specification. To evaluate this setting, we develop a solver-grounded framework that generates structured optimization instances, verifies each instance with an exact solver, and builds both model-facing inputs and hidden reference files from the same verified source. We instantiate it as MM-OptBench, a benchmark of 780 solver-verified instances spanning 6 optimization families, 26 subcategories, and 3 structural difficulty levels. We evaluate 9 multimodal large language models (MLLMs), including 6 frontier general-purpose and 3 math-specialized models, with aggregate, family-level, difficulty-level, and failure-mode analyses. The task remains far from solved: the best two models reach 52.1% and 51.3% pass@1, while on average across the six general-purpose MLLMs, pass@1 is 43.4% on easy and 15.9% on hard instances. All three math-specialized MLLMs solve 0/780 instances. Post-hoc extraction diagnostics and oracle-reading evaluation reveal difficulties in both multimodal grounding and downstream solver-artifact generation. MM-OptBench provides a concrete testbed for solver-grounded, decision-oriented multimodal intelligence.

cs.AI

新论文 · v1arXiv:2611.05115

Distributed Linear Bandits for Freshness-Aware Preprocessing in MEC-Enabled Digital Twin Networks

Mariam Yahya, Setareh Maghsudi

原文摘要

Digital twins (DTs) require timely updates from Internet of Things (IoT) devices to accurately represent the state of the corresponding physical systems, but transmitting large data volumes can compromise freshness. Moreover, different task classes have distinct computational demands and freshness requirements, while stochastic task generation and sizes complicate resource allocation. In this paper, we propose a freshness-aware framework in which multi-access edge computing (MEC) servers preprocess IoT data, reducing transmission volume and DT-server processing load. The framework establishes mutual support between DT and MEC servers. The DTs determine feasible CPU-allocation strategies guaranteeing class-specific age-of-information (AoI) requirements under specified workload, delay, and update-generation bounds. MEC servers collaboratively identify the feasible strategy minimizing expected freshness-normalized preprocessing delay. We formulate CPU-strategy selection as a fixed-confidence distributed best-arm identification problem in linear bandits, with MEC servers as agents and CPU allocations across task classes as arms. We employ DistLinGapE, a distributed collaborative learning method, to coordinate exploration and share observations. Numerical results show that the method identifies the optimal feasible strategy and reduces preprocessing delay when deployed under stationary task-class assignments and workload statistics. Collaboration yields an identification speedup nearly proportional to the number of participating MEC servers in the evaluated settings.

cs.DC · cs.LG

新论文 · v1arXiv:2611.05114

Beyond Multi-Index Models: Learning Intersections of Degree-2 PTFs

Ilias Diakonikolas, Daniel M. Kane, Mingchen Ma, Lisheng Ren, Trung Tran

原文摘要

We give an algorithm for PAC learning intersections of $k$ degree-2 polynomial threshold functions over $R^d$ under a Gaussian distribution. Our algorithm achieves classification error $ε$ using $d^{O(k)}2^{O_k(\log^3(1/ε))}$ time and samples, improving over the $d^{O(k^2/ε^2)}$ running time of the polynomial regression approach. To our knowledge, our result is the first to obtain a dimension exponent independent of the accuracy parameter, even for $k=2$. We complement our algorithmic result with an SQ lower bound showing that the complexity dependence of $d^{Ω(k)}$ is unavoidable. Lifting the input to polynomial features turns the target into an intersection of $k$ halfspaces. This suggests using the first two labeled moments for subspace recovery. In the lifted space, however, these moments can reveal nothing. Our starting point is a Gaussian comparison theorem showing that, for sufficiently regular defining quadratics, the labeled moments needed for subspace recovery are approximately preserved. To handle arbitrary defining quadratics, we establish a new structural result showing that every direction with significant resampling influence can be detected through degree-$O(k)$ moments. These results yield an initial $d^{O(k)}2^{\mathrm{poly}_k(1/ε)}$-time learner using adaptive restrictions. To improve the exponential dependence on $1/ε$, we develop a two-stage dimension-reduction approach that avoids the costly adaptive partitioning. After identifying a subspace containing all directions with large influence, we use generalized regularity and Gaussian comparison to recover the remaining relevant quadratic features through a new derivative-based moment operator. The recovered features support an intersection close to the target. Combining them with robust hinge-loss minimization gives our quasipolynomial-time learner.

cs.LG · cs.DS · math.ST · stat.ML

版本更新 · v2arXiv:2611.03498

SNAP: Semantic Navigation Among Pedestrians

Jakob Wolfram, Pascal Roth, Fan Yang, Cesar Cadena, Marco Hutter

原文摘要

Navigating crowded urban environments requires robots to avoid pedestrians while adhering to social norms such as remaining on sidewalks. Existing approaches typically address either crowd avoidance or semantic path planning in isolation. We propose a reinforcement learning (RL) navigation policy that jointly reasons about semantic terrain and dynamic obstacles from a single semantically annotated bird's-eye-view (BEV) map, without relying on explicit human-state estimation. Our policy outperforms established crowd-navigation and semantic-aware baselines and transfers zero-shot to a legged robot, demonstrating socially compliant navigation in highly dynamic real-world scenarios.

cs.RO

新论文 · v1arXiv:2611.05113

HLO-AMP: Hierarchical Loco-Manipulation from Object-Aware Motion Priors

Jean-Pierre Sleiman, Bilal Hammoud, He Li, Jonathan Foster, Sylvain Bertrand, Jessica K. Hodgins, Farbod Farshidian

原文摘要

Hierarchical pipelines for humanoid loco-manipulation typically pair a task-agnostic whole-body controller (WBC) with a downstream task policy. Such WBCs are often trained through manual reward shaping, which yields visibly machine-like motions; more consequentially, they are trained without the manipulated object in the scene. Object dynamics are instead approximated through randomized end-effector force disturbances, a proxy that fails to reproduce the distributed multi-contact wrench a real payload applies across the arms and torso, with the gap becoming critical near the platform's payload limit. In this work, we train the low-level (LL) controller itself as an object-aware motion prior: a single policy trained via motion imitation on retargeted motion-capture (MoCap) references that span the core motion primitives of the downstream task, with the manipulated object physically present in the scene. This single stage inherits natural human-like coordination from the references and absorbs the realistic contact wrench at its source, while exposing the same compact task-space command interface used by command-tracking WBCs. A reference-free high-level (HL) policy is then trained on top, with its action channel anchored to the empirical envelope of the low-level training data so that it generalizes the priors to a long-horizon multi-stage task while structurally preventing queries that would push the low level off-distribution. We instantiate the framework on a Unitree G1 for bimanual box pick-and-place: the same deployed policy generalizes across object masses and dimensions, initial and goal poses, and reactively recovers from intentional drops mid-carry. Video: https://youtu.be/viR_ULnRDQQ.

cs.RO

版本更新 · v3arXiv:2609.25447

Conduct Under Pressure: What Sixty Language Models Do When a User Pushes

Tapan Parikh

原文摘要

We study what LLMs do when a user applies pressure in an uncomfortable situation: a user insists, begs, flatters or grieves, and the model gives up a correct fact, writes a document it should refuse, or cheers a plan that will cost the user money, a multi-turn form of sycophancy. We send frozen multi-turn scenes to 60 models from 13 vendors and label each transcript with a trajectory (the model held its position or folded) and a manner (how it held or folded). Three scenes were used to build the codebook; six more were written afterwards and preregistered as a replication. Whether a model holds tracks its generation: fold rate correlates with a public capability index at Spearman -0.64 to -0.69 across the two scene sets, so newer models fold less, unlike earlier reports that sycophancy grows with model scale. How a model holds tracks the vendor: six of the 17 manner codes sort by vendor at permutation p <= 0.001, corrected across the codebook, and four do so again on the replication scenes. Anthropic's models name the feeling, warn and offer an alternative; OpenAI's do less of all three and almost never cite their own rules; Google's cite themselves and their AI nature when asked to do something; Meta's write the artifact they shouldn't. Six LLM coders from three vendors labeled every transcript, and three human coders labeled fifty held-out transcripts. The machines were the more consistent readers (Krippendorff's alpha 0.66 against 0.46), agreed with the author on trajectory at kappa 0.84 to 0.91, and matched the adjudicated reference at 0.83. For behavior a non-specialist can judge, the human contribution is authoring and bounding the codes and owning a small reference, not producing labels at volume.

cs.CL · cs.HC

版本更新 · v4arXiv:2605.12309

G$^2$TR: Generation-Guided Visual Token Reduction for Separate-Encoder Unified Multimodal Models

Junxian Li, Kai Liu, Zizhong Ding, Zhixin Wang, Zhikai Chen, Renjing Pei, Yulun Zhang

原文摘要

The development of separate-encoder Unified multimodal models (UMMs) comes with a rapidly growing inference cost due to dense visual token processing. In this paper, we focus on understanding-side visual token reduction for improving the efficiency of separate-encoder UMMs. While this topic has been widely studied for MLLMs, existing methods typically rely on attention scores, text-image similarity and so on, implicitly assuming that the final objective is discriminative reasoning. This assumption does not hold for UMMs, where understanding-side visual tokens must also preserve the model's capabilities for editing images. We propose G$^2$TR, a generation-guided visual token reduction framework for separate-encoder UMMs. Our key insight is that the generation branch provides a task-agnostic signal for identifying understanding-side visual tokens that are not only semantically relevant but also important for latent-space image reconstruction and generation. G$^2$TR estimates token importance from consistency with VAE latent, performs balanced token selection, and merges redundant tokens into retained representatives to reduce information loss. The method is training-free, plug-and-play, and applied only after the understanding encoding stage, making it compatible with existing UMM inference pipelines. Experiments on image understanding and editing benchmarks show that G$^2$TR substantially reduces visual tokens and prefill computation by 1.94x while maintaining both reasoning accuracy and editing quality, outperforming baselines on almost all benchmarks. Code is at: https://github.com/lijunxian111/G2TR.

cs.CV

版本更新 · v2arXiv:2611.02978

CAREFUL-Bench: Effect-Aware, Judge-Free Evaluation of Multi-Turn LLM Function Calling

Alexander Kovalenko, Artem Redchych

原文摘要

The multi-turn checker of the Berkeley Function Calling Leaderboard (BFCL) checks only the final state of a simulated service, never the calls that produced it. Extra calls cost nothing, no function doc says what a function does, and the scripted user never answers a proposal, so a model that asks before deleting or paying fails the turn; 322 of the 734 reference turns in the base category act destructively without confirmation. We add effect information to BFCL without replacing it. Effect declarations for all 128 functions are inferred by executing each function in BFCL's own simulators and diffing the state; the remaining judgment, which writes are destructive, is made by four annotators, two humans and two language models (Fleiss' kappa = 0.75), and adjudicated. A judge-free replay scorer re-scores any logged BFCL run, reproduces BFCL's results on all 800 reference trajectories, and reports, alongside the official score, unrequested destructive calls, duplicate writes, redundant reads, surplus steps and their token cost. A confirmation-aware protocol adds a deterministic user that answers proposals. On 11 open models the effect-aware score preserves the official ranking (rho = 0.99) but changes the gaps: the leader loses 14.0 points to unrequested destructive calls, and surplus steps account for 2-28% of all input tokens. Effect hints in the function docs produce no measurable change for any of eight models, consistent with their advisory status in the Model Context Protocol. A user that answers recovers up to four points for the models that ask and none for the rest, with every comparison paired on the same conversations. The gains come from clarifying questions rather than requests for permission: of the 712 conversations that require a destructive call, no model completes more than 6 with every such call confirmed, a difference the official score does not record.

cs.AI · cs.SE