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 16:40 — 2026-11-16 16:40(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)
版本更新 · v4arXiv:2602.02711

Dynamic Mixed-Precision Routing for Efficient Multi-step LLM Interaction

Yuanzhe Li, Jianing Deng, Jingtong Hu, Tianlong Chen, Song Wang, Huanrui Yang

原文摘要

Large language models (LLMs) achieve strong performance in long-horizon agentic tasks through multi-step interaction and reasoning at test time. However, repeatedly invoking these models incurs substantial inference cost over an episode. Post-training quantization (PTQ) reduces this cost, but using a fixed low precision throughout an episode can sharply lower task success. We observe that interaction steps differ sharply in their sensitivity to quantization: given the same context, decoding a step at low precision turns a success into a failure at only about 1% of the steps, which we call critical steps. Based on this observation, we propose Dynamic Mixed-Precision Routing (DMR), a framework that adaptively selects between high-precision and low-precision versions of the same LLM at each decision step. Leveraging an entire-space factorization to jointly predict a step's divergence and its effect on the outcome, DMR learns to recognize these critical steps by training only a lightweight router on the model's hidden state. By reusing a shared high-precision KV cache and the hidden states, DMR avoids redundant context prefill and a separate text encoder, enabling precision switching with less than 1 ms of overhead per step and adding less than 0.11% to step latency for routing. This mixed-precision design also enables nested quantization, reducing the memory footprint by 39% on average compared with FP16. Across three long-horizon agentic benchmarks with nested quantization, DMR achieves an average 2.44x decode speedup over FP16 at comparable task success rates, without fine-tuning the underlying model. On ALFWorld with GPTQ, DMR achieves a 1.71x decode speedup with 86.9% success, compared with 87.2% for FP16.

cs.AI

版本更新 · v4arXiv:2512.08991

Action-Guided Reachability for Deterministic World Models of Vision-Based Systems

Yuang Geng, Zhongzheng Zhang, Chengzhen Jiang, Yanru Li, Xinyang Wang, Zhuoyang Zhou, Ivan Ruchkin

原文摘要

End-to-end vision-based controllers map raw camera images directly to control actions, yet verifying their closed-loop behavior remains challenging: no analytical model maps a physical state to a camera image, and propagating a set of states through high-dimensional images yields loose reachable sets. We address both challenges with two techniques. First, we propose a Deterministic World Model (DWM), a latent-free decoder that generates camera images only from the physical state, trained with a saliency-weighted reconstruction loss and a control-consistency loss. The DWM replaces the camera in the closed loop, so a sound and tight reachable tube of the surrogate closed loop can be computed without bounding any latent variables. Second, we propose an action-guided partition-and-verify scheme that certifies as many initial sets as possible within a fixed computation budget. A Gaussian process learned from cheap simulations orders the cells and splits those that lie across the property boundary, while every other unknown cell is split along the state dimension that most changes the controller's actions along the simulated closed loop, i.e., the one that most reduces the pixel uncertainty reaching the controller. On a CARLA braking system and three Gym benchmarks, the DWM is more accurate to the real camera than a cGAN surrogate and yields 1.9x to 69x narrower robustness intervals. The vision part causes most of the remaining overapproximation, and our splitting reduces the unknown fraction of the four-dimensional CartPole initial set from 21.88% to 8.18% within the budget.

cs.CV · cs.LG

版本更新 · v2arXiv:2609.37431

Query Complexity of Testing Structured Parenthesis Languages

Tim Jackman, Diptaksho Palit, Sofya Raskhodnikova

原文摘要

We study the query complexity of testing membership in structured string languages, focusing on Dyck languages and natural generalizations. A tester receives query access to a word and must distinguish valid inputs from words that are far in Hamming distance, while inspecting only a sublinear number of positions. Our results sharpen the boundary between constant-query testability and polynomial query complexity. First, we prove an $Ω(n^{2/5})$ lower bound for testing Dyck languages $D_m$ with any fixed number $m\ge2$ of parenthesis types, improving the previous $Ω(n^{1/5})$ lower bound of Fischer, Magniez, and Starikovskaya (SODA `18) and nearly matching their upper bound of $O(n^{2/5+o(1)})$. Furthermore, we show that all nonadaptive algorithms for these problems require $Ω(n^{1/2})$ queries. Our lower bounds use a Pólya-urn process to construct the hard distribution; Second, we identify a broad class of weighted-parenthesis languages, which we call {\em excursion languages,} that remain constant-query testable. These languages encode bounded-step walks that stay nonnegative and return to zero. For every fixed excursion language, we give a nonadaptive tester with query complexity $O(1/\varepsilon^2)$, and we prove this dependence on $\varepsilon$ is optimal, even for adaptive algorithms. As a special case, we obtain the tight $Θ(1/\varepsilon^2)$ query complexity of testing $D_1$, improving the previous $O(\log(1/\varepsilon)/\varepsilon^2)$ upper bound and giving the first matching two-sided-error lower bound. Third, we construct a simple hard language, Hidden String, that is generated by a deterministic linear grammar but nevertheless requires $Ω(n^{2/5})$ adaptive queries and $Ω(n^{1/2})$ nonadaptive queries to test. This shows that polynomial query complexity appears even for highly restricted string languages.

cs.DS · cs.CC

版本更新 · v2arXiv:2609.06044

Art2Song: Enhancing Visual Art Appreciation with Contextual Music Generation

Sungeun Jo, Myung Jin Kim, Chi Yoon Jeong

原文摘要

Art2Song is a conceptual framework that expresses artworks as sound by separating Non-Visual Context, which is difficult to perceive from the image alone, from Visual Evidence. Visual Evidence, such as objects, colors, and spatial composition, is transformed into Lyrics, while the Contextual Mood derived from the historical and art-historical context in the museum's artwork description is reflected in the background soundtrack. Rather than describing artworks textually, Art2Song aims to explore the possibility of a new mode of art appreciation in which viewers experience hidden stories and emotional context through music. As future interaction directions, we plan an Emotional Layer Blending Slider interface and a structured, traceable song-generation scenario, presenting the possibility that users can explore the relationship between Visual Evidence and Contextual Mood.

cs.HC

版本更新 · v3arXiv:2604.19019

Smiling Regulates Emotion During Traumatic Recollection

Marcus Ma, Emily Zhou, Leonard Ludwig, Julia Hörath, Christina Winkler, Kleanthis Avramidis, Tiantian Feng, Gabor Mihaly Toth 等 10 位作者

原文摘要

We study when, where, and why 978 Holocaust survivors smile in video testimonies. We create an automatic smile detection model from facial features with an F1 of 85% and annotate detected smiles under two established taxonomies of smiling. We produce narrative features on 1,083,417 transcript sentences as well as emotional valence from three different modalities: audio, eye gaze, and text transcript. Smiling rates are associated with specific semantic topics, narrative structures, and temporal syntaxes across the corpus. Smiles often occur during periods of intense negative affect and we find negative-affect smiles are associated with more positive subsequent valence trajectories across all three modalities. Smiling reduces eye dynamics and blink rates, with both of these effects modulated by narrative valence. Taken together, we conclude that smiling plays a critical role in regulating emotion and social interaction during traumatic recollection.

cs.MM

新论文 · v1arXiv:2611.13457

Deep Personalization: Continuous Context Observation for Local Proactive Agents

Yijia Shao, Song Bian, Junxian Guo, Jindong Jiang, Zhen Dong, Huizi Mao, Diyi Yang, Song Han

原文摘要

As AI agents become increasingly capable, users must still recognize when and how to ask for help, leaving many opportunities for useful support unrealized. Proactive agents hold the promise to close this gap, but effective proactivity requires understanding users' real-time activities and personalizing intervention timing so that assistance is offered when it is likely to be helpful rather than disruptive. We introduce deep personalization, a closed-loop process that connects continuous context observation with continual agent adaptation: an agent observes user activity, selectively retains long-term context, and adapts intervention decisions from implicit feedback. We initiate deep personalization in Coco, a proactive desktop agent, by having a separate Observer Agent that monitors desktop activity, maintains compact long-term user modeling context, and learns when assistance is desirable from implicit signals such as engagement, dismissal, explicit requests, and repeated work patterns. We further develop efficient on-device strategies to enable the Observer Agent to run on local AI compute platforms without streaming raw activity data off-device. For evaluation at scale without collecting sensitive real-user activity, we introduce a data-synthesis pipeline that transforms existing computer-use trajectories into OSWorld-Persona, a benchmark with persona-conditioned intervention labels. On OSWorld-Persona, deep personalization improves macro F1 across three model backbones from 0.12-0.19 under direct prompting to 0.35-0.48 after adaptation, a 2.4-3.0x relative gain. In a week-long user study, we observe that Coco improves its macro-averaged F1 score from 0.199 to 0.515. These results suggest that on-device deep personalization can make proactive agents more timely and increasingly aligned with each user's everyday work.

cs.LG

新论文 · v1arXiv:2611.13456

When Reasoning looks like Hidden Understanding: How Chain-of-Thought Fosters Anthropomorphism, Perceptions of Epistemic Superiority, and Cognitive Deference

Oliver Lack, Lana Tikhomirov, Humphrey Thompson, Carolyn Semmler, Melissa McCradden

原文摘要

As AI-based products permeate society, specifically systems using generative models, a belief system is emerging about what these products are and what they can and cannot do. One common belief concerns the perception of the form of language as evidence of 'understanding' language. As these generative models reproduce linguistic patterns rather than understand meaning, interpreting these patterns as evidence of understanding is a misattribution. This misconception is extended by the notion that generative models can 'explain their reasoning.' Anecdotally, these misconceptions which mistake the form of language for its meaning are driving cognitive deference to AI systems in a way that displaces human agency. Such mistakes involve attributions of cognitive properties to the system, allowing them to be treated like intentional 'minds'. Importantly, perceived epistemic authority and superiority make that 'mind' worth deferring to. Anthropomorphism in tandem with epistemic authority and superiority, we argue, increases the risk that users generalise from narrow task performance to broader competencies, committing the "fallacy of composition". This paper explores misconceptions of Chain of Thought (CoT) reasoning as sufficient transparent 'explanation'. We explore how anthropomorphism acts as a vehicle for cognitive deference of decision-making and perceptions of epistemic superiority in AI systems. Finally, we explore CoT's ramifications using medicine as an example. Taken together, we conclude that explainability operationalised through CoT is highly problematic and undercuts the very ethical principles it is intended to serve. While CoT may be valuable for improving output accuracy and moderation, we suggest that the practice of asking models to provide user facing reasons for their outputs should be actively discouraged, and AI literacy training should explicitly caution against its use.

cs.AI · cs.HC

版本更新 · v2arXiv:2603.06639

RECAP: Local Hebbian Prototype Learning as a Self-Organizing Readout for Reservoir Dynamics

Heng Zhang

原文摘要

Robust perception in brains is often attributed to high-dimensional population activity together with local plasticity mechanisms that reinforce recurring structure. In contrast, most modern image recognition systems are trained by error backpropagation and end-to-end gradient optimization, which are not naturally aligned with local computation and local plasticity. We introduce RECAP (Reservoir Computing with Hebbian Co-Activation Prototypes), a bio-inspired learning strategy for robust image classification that couples untrained reservoir dynamics with a self-organizing Hebbian prototype readout. RECAP discretizes time-averaged reservoir responses into activation levels, constructs a co-activation mask over reservoir unit pairs, and incrementally updates class-wise prototype matrices via a Hebbian-like potentiation-decay rule. Inference is performed by overlap-based prototype matching. The method avoids error backpropagation and is naturally compatible with online prototype updates. We illustrate the resulting robustness behavior on MNIST-C, where RECAP remains robust under diverse corruptions without exposure to corrupted training samples.

cs.NE · cs.AI · cs.CV · cs.LG · q-bio.NC

版本更新 · v2arXiv:2610.14804

AgentWeave: Efficient Distributed Agent Serving via Flow Decomposition

Kaibin Guo, Pengtu Li, Zicong Hong, Zhiyuan Fang, Wuhui Chen, Rachid Guerraoui, Anne-Marie Kermarrec

原文摘要

Agent applications are evolving into workflows composed of large language model (LLM) calls and tool invocations, termed modules. Existing distributed agent systems typically adopt a disaggregated architecture, deploying different modules on separate devices. While this enables pipeline parallelism, imbalanced module latencies introduce severe pipeline bubbles that limit throughput. A natural alternative is an aggregated architecture that co-deploys all modules across all devices, eliminating bubbles by keeping every device continuously active. However, this introduces request blocking: short requests are forced to progress in lockstep with longer ones in the same batch, and newly arriving requests must wait until the entire ongoing batch completes the full workflow. To address this, we propose AgentWeave, an efficient distributed serving system for agent applications built on the aggregated architecture. Its core mechanism, flow decomposition, decomposes long agent requests into fine-grained execution units while preserving agent semantics, allowing short and newly arriving requests to proceed without being stalled by long-running ones. Complementing this, a flow-adaptive scheduler dynamically balances the throughput gains of decomposition against its scheduling overhead. Experiments across four representative agent applications demonstrate that AgentWeave improves throughput by 1.28-1.81x, reduces average latency by 2.08-2.45x, and reduces P90 latency by 1.56-2.04x over state-of-the-art baselines.

cs.DC

新论文 · v1arXiv:2611.13454

SafeRoute-MAS: Coverage-Aware and Risk-Calibrated Message Routing for Multi-Agent Scientific Reasoning

Zhiming Xue, Jianan Liu, Weiran Yan, Aicheng Xie, Jing Yang

原文摘要

We introduce SafeRoute-MAS, a coverage-aware, risk-calibrated procedure that keeps all specialist messages or removes one using only pre-integration information. On a sealed cache of 5,328 committees, its source-compatible operating point gains +1.80 percentage points (95% CI [+1.13, +2.52]) while removing 12.76% of message tokens. We also run 4,580 controlled integrations over a stratified 458-committee sample on a public gpt-oss-120b endpoint. The sample estimate is -0.09 points; design weighting to the original workload gives +0.94 points (95% CI [-1.31, +3.50]) and 18.72% acted-input reduction. A weighted token-matched comparison likewise does not establish a selector advantage (+1.68 points, CI [-1.27, +5.10]). Only 13 of 88 cached non-neutral labels (14.8%) retain direction by repeated-call majority. Expected-corruption UCL is 4.60% after weighting, but only 2 of 11 prespecified marginals pass the support/risk rule. We therefore propose a numerical qualification with Keep-All fallback, applied here retrospectively rather than validated prospectively.

cs.AI

新论文 · v1arXiv:2611.13453

Camera3R: Feed-Forward 3D Reconstruction from Heterogeneous Camera Models

Ruxiao Duan, Yunwen Zhou, Erin Hong, Dongxu Zhao, Eric Turner, Federico Tombari, Alex Wong

原文摘要

Feed-forward 3D reconstruction (3R) from unconstrained multi-view images has recently emerged as a prominent research direction in 3D computer vision. Various foundational reconstruction models have been proposed with different architectures and reconstruction paradigms. Despite being trained on large-scale perspective datasets with substantial computational resources, these models generally struggle to generalize to non-perspective images. This limitation hinders their deployment in applications involving arbitrary, unknown, and potentially heterogeneous camera models. To build on the capabilities of existing reconstruction models with minimal retraining, we propose Camera3R, a low-rank adaptation framework that efficiently extends perspective-pretrained 3R networks to any other camera models. Extensive experiments demonstrate that, with only 0.5% additional parameters and less than a day of training, Camera3R adapts a foundational 3D reconstruction model to non-perspective cameras while fully preserving its original performance on perspective inputs. Code is available at https://github.com/android-xr/camera3r.

cs.CV

新论文 · v1arXiv:2611.13452

SLIC: Reinforcement Fine-Tuning Small LMs for Multi-Turn Analog Circuit Optimization

Yan Xu, Zhengqi Gao, Ching-Yun Ko, Zhang-Wei Hong, Wenjie Lu, Tao Yu, Xin Zhang, Duane S. Boning 等 9 位作者

原文摘要

Analog integrated circuit (IC) design is a multi-turn, simulator-grounded engineering task that demands iterative reasoning over coupled performance tradeoffs, yet proprietary process design kits (PDKs) make hosted large language model workflows difficult to deploy in practical design settings. We present SLIC (Small LM Iterating on Circuits), a reinforcement fine-tuning framework that trains a 4B-scale, locally deployable language model for closed-loop analog circuit optimization. SLIC pairs structured circuit representations with discretized relative-update actions, domain- and success-gated multiobjective rewards, and Reward-Guided Rollout Steering (RGRS), which turns per-step simulator feedback into signals for both policy learning and online rollout construction in this sparse, highly constrained domain. We further release an open, goal-conditioned benchmark of 41 amplifier topologies with standardized protocols for in-distribution validation, held-out topology evaluation, and cross-technology-node transfer. Trained solely with open source 65 nm CMOS predictive technology models (PTM), SLIC-Random achieves 87.1% in-distribution four-specification Pass@1 within eight turns, surpassing the 763B DeepSeek-V4.1-Flash Think by 48.5 percentage points. Beyond the training distribution, SLIC-Softmax reaches 72.6% Pass@1 on held-out topologies, and the trained policies consistently transfer across five unseen technology nodes without additional fine-tuning. This transfer extends to a commercial TSMC 65 nm CMOS PDK, where SLIC-Random achieves 49.9% held-out-topology Pass@1, demonstrating practical deployment through local inference without transmitting proprietary data to external LLM services. Performance further scales with test-time simulator interactions, reaching up to 83.2% held-out-topology Pass@1 at a 32-turn budget.

cs.LG

版本更新 · v2arXiv:2607.16319

Deep Learning-based Filtering for Video Coding: A Survey on Architectures, Algorithms, and Complexity Analysis

Young-Woon Lee, Byung-Gyu Kim

原文摘要

As Ultra-High-Definition (UHD) displays and immersive media services become ubiquitous in the Internet of Things (IoT) and Consumer Electronics (CE) sectors, including 8K display and mobile devices, the demand for high-efficiency video coding is unprecedented. While Deep Learning-based Filtering (DLF) has emerged as a promising solution to mitigate compression artifacts inherent in standards like High Efficiency Video Coding (HEVC/H.265) and Versatile Video Coding (VVC/H.266), its deployment in CE devices is severely constrained by computational complexity, memory bandwidth, and power consumption. To bridge the gap between academic research and practical deployment, this paper presents a comprehensive, hardware-oriented survey of DLF techniques. We propose a systematic three-dimensional taxonomy classifying methods into (1) Integration Scheme within the Video Coding, (2) Coding Information Utilization, and (3) Network Design Strategy. Unlike prior reviews, this work critically analyzes the trade-offs between Rate-Distortion (RD) performance and hardware feasibility, highlighting the evolution from heavy, performance-oriented models to lightweight, hardware-friendly architectures targeting Neural Processing Units (NPUs). Furthermore, we incorporate the latest standardization activities from the Joint Video Experts Team (JVET) on Neural Network-based Video Coding (NNVC) to provide realistic guidelines. We also identify open challenges such as real-time inference latency and error propagation, providing a roadmap toward robust, low-power intelligent video coding in next-generation CE vision endpoints.

cs.CV

版本更新 · v2arXiv:2610.02418

Mitigating Private Data Leakage in LLMs with Whiteout

Anna Yoo Jeong Ha, Ronik Bhaskar, Haitao Zheng, Ben Y. Zhao

原文摘要

Modern large language models (LLMs) are trained on massive, largely unfiltered datasets, including content scraped from nearly every accessible website and user inputs. As a result, LLMs often memorize and reproduce personally sensitive information (PSI) such as birth dates, phone numbers, and home addresses. This leads to significant privacy risks, particularly for high-profile individuals such as executives, politicians, and judges. Existing mitigations largely rely on machine unlearning. However, these methods often remove more information than needed, degrade model utility and safety, and are highly vulnerable to attacks. This paper presents Whiteout, a practical tool that, upon requests by individuals, prevents LLMs from regurgitating their genuine PSIs, by overwriting them using precise and carefully designed obfuscation samples. We evaluate Whiteout on modern LLMs of varying sizes and makers, including a widely-used OpenAI model. Results show that Whiteout effectively prevents disclosure of the targeted PSIs, has negligible impact on model utility and safety, and outperforms existing alternatives. We also test Whiteout against a wide range of countermeasures, from black-box attacks like jailbreaking to white-box adaptive attacks like relearning and quantization. Finally, we conclude with a discussion on the security and ethical implications of Whiteout.

cs.CR · cs.AI

新论文 · v1arXiv:2611.13448

Fermat ProbCover: Coverage-Based Active Learning on Fermat Distance Graphs

Gokul Bhusal, Delanna Do, Kevin Miller, Ekaterina Merkurjev, Stephan Wojtowytsch

原文摘要

Active learning aims to reduce annotation cost by selecting the most informative examples for labeling. In the low-budget regime, where uncertainty-based methods often fail, coverage-based active learning methods such as ProbCover have been shown to be effective. However, these methods typically rely on Euclidean distance, which may not adequately capture the underlying geometry of data. In this article, we propose PC-Fermat, a coverage-based active learning method that replaces Euclidean coverage neighborhoods with neighborhoods defined by a Fermat distance, i.e., a data-adaptive distance that reflects the underlying distribution of the data points. To facilitate computation, we construct a weighted Euclidean neighborhood graph and use a truncated shortest-path algorithm to compute the Fermat coverage graph. We also establish sufficient conditions under which this construction exactly preserves the Fermat coverage computed on the complete graph. Finally, we evaluate PC-Fermat on eight classification datasets across varying labeling budgets, with particular emphasis on the very low-budget regime. Our experimental results demonstrate that PC-Fermat outperforms several existing active learning methods on the majority of the datasets.

cs.LG

版本更新 · v2arXiv:2606.27385

RANSAC Scoring Done Right

James Pritts, Felix Seegräber, Kevin Köser

原文摘要

The most widely used RANSAC variants score a candidate model by summing per-point scores that fall to zero beyond a threshold. These scores derive from the negative log-likelihood of a Gaussian-uniform mixture. The threshold separates inliers, which follow the noise model, from outliers, and it is a function of the observation noise level, also known as the inlier scale. We remove this dependence on the threshold by reversing the usual order of RANSAC inference: rather than estimating the scale and then scoring against it, we marginalize the scale in closed form for a fixed inlier partition and then optimize over partitions. The resulting score is maximized by a single sort followed by a linear sweep. A single closed-form expression spans informative and non-informative priors, so the same score adapts to datasets with extensive or limited ground-truth annotation. The proposed score has a single hyperparameter, the half-width of the uniform outlier distribution. We show that this half-width is far less sensitive to miscalibration than the inlier threshold and is easier to cross-validate. The proposed score is the first likelihood-based RANSAC score in which the inlier scale is absent from the formula. Project page: https://ransacscoringdoneright.github.io

cs.LG · cs.CV

版本更新 · v3arXiv:2408.16852

The Star Geometry of Critic-Based Regularizer Learning

Oscar Leong, Eliza O'Reilly, Yong Sheng Soh

原文摘要

Variational regularization is a classical technique to solve statistical inference tasks and inverse problems, with modern data-driven approaches parameterizing regularizers via deep neural networks showcasing impressive empirical performance. Recent works along these lines learn task-dependent regularizers. This is done by integrating information about the measurements and ground-truth data in an unsupervised, critic-based loss function, where the regularizer attributes low values to likely data and high values to unlikely data. However, there is little theory about the structure of regularizers learned via this process and how it relates to the two data distributions. To make progress on this challenge, we initiate a study of optimizing critic-based loss functions to learn regularizers over a particular family of regularizers: gauges (or Minkowski functionals) of star-shaped bodies. This family contains regularizers that are commonly employed in practice and shares properties with regularizers parameterized by deep neural networks. We specifically investigate critic-based losses derived from variational representations of statistical distances between probability measures. By leveraging tools from star geometry and dual Brunn-Minkowski theory, we illustrate how these losses can be interpreted as dual mixed volumes that depend on the data distribution. This allows us to derive exact expressions for the optimal regularizer in certain cases. Finally, we identify which neural network architectures give rise to such star body gauges and when do such regularizers have favorable properties for optimization. More broadly, this work highlights how the tools of star geometry can aid in understanding the geometry of unsupervised regularizer learning.

cs.LG · math.MG · math.OC · math.ST · stat.ML

新论文 · v1arXiv:2611.13442

ForeSpark: Joint Draft and Verification Budget Scheduling for Parallel Speculative Decoding

Jaeyong Song, Daniel Wang, Pratyush Patel, Jonny Kong, Preyas Shah, Shobhit Kanaujia

原文摘要

Parallel drafters accelerate LLM decoding by proposing a whole block of draft tokens in a single forward pass, making them attractive for LLM serving. Recent systems optimize goodput (i.e., accepted tokens per second) because rejected drafts steal compute from other requests. However, these systems control only how many tokens are verified, so every request still drafts a full block. At high serving load, we find that drafting consumes up to 40% of step time, of which a majority is discarded because it is never verified. Yet, draft-length control from autoregressive drafters does not carry over, because a parallel drafter commits to its length before its single pass. Moreover, draft and verification lengths are coupled across the batch, so they must be selected jointly for every request while keeping the decision computationally cheap at every step. In this paper, we propose work-conserving speculation (WCS), a design that jointly allocates draft and verification budgets per request in LLM serving, and present ForeSpark as its prototype. ForeSpark realizes WCS with (i) a context-aware pre-draft estimator that predicts acceptance for every candidate length from the target model's context features, (ii) a lightweight joint draft-verify scheduler that selects both lengths through a global sort, and (iii) a post-draft budget distribution that spends verification on the most promising tokens. Implemented on vLLM, ForeSpark improves throughput by up to 1.77x over the state-of-the-art DSpark and by 2.49x over autoregressive decoding at a matched per-request generation rate, without changing the output distribution. It also tolerates higher load, sustaining up to 1.51x the request rate of DSpark under latency service level objectives (SLOs).

cs.DC · cs.AI · cs.LG

新论文 · v1arXiv:2611.13441

Inherited Weights: Measuring the Tokenization Premium and the Sovereignty Cost of Language Models for the Global South

Paulo Camara

原文摘要

Most language models deployed outside the United States and China are fine-tuned descendants of a small number of pretrained families. We argue that fine-tuning leaves the decisive properties of a model inherited from its base: vocabulary, knowledge, cultural alignment, memorized data and license. We contribute (i) an updated, reproducible measurement of the tokenization premium across 31 languages and 7 tokenizers released in 2023-2024, on 27 articles of the Universal Declaration of Human Rights, with explicit correction for unknown-token collapse. Premiums reach 10x for Amharic in older tokenizers and remain at 6.8x in the newest commercial one; Indigenous South American languages pay between 1.6x and 2.7x under global tokenizers. A vocabulary built from Portuguese lowers the premium for 11 of 13 Latin-script Global South languages relative to every global tokenizer tested, by about 24% for Guarani and Quechua. (ii) An eight-test black-box sovereignty audit that a buyer can run without access to weights. (iii) A regime argument: under energy and liability constraints, accuracy per joule, calibration and selective risk should replace raw accuracy as design objectives, with metrics that abstention cannot inflate. We close with preliminary internal measurements from a proprietary architecture built around these objectives, reported with their limitations. Data and measurement code are included.

cs.CL · cs.AI · cs.CY

版本更新 · v2arXiv:2611.05538

Recasting Visual Language Models as EEG Foundation Models for Clinical Neurodiagnostics

Xin Zhou, Elena Sergeeva, Nan Bi, Wenyu Song, Lijun Yin, Hao Deng

原文摘要

Historically automated electroencephalography (EEG) systems was done using task-specific supervised models trained on purposely annotated data. A more recent modeling paradigm relies on creating a unified representation, or a 'foundation model', for multiple modality-specific tasks through large-scale domain-specific pretraining on the unannotated EEG signal. The quality of this unified representation directly depends on the availability of the modality-specific datasets. At the same time, generalist Vision Language Models (VLMs) exhibit broad few-shot learning capacities as well as good fine-tuning adaptability to different visual and language tasks. Can VLMs learn how to read and interpret EEG visually, emulating to a degree the way human clinicians inspect the EEG patterns? In this work we make a case that they can: we show that generalist VLMs without large-scale EEG pretraining perform competitively across a collection of clinical neuro-diagnostic tasks, can adapt to different tasks in a few-shot learning setting better than specialist foundation models, and exhibit better performance with model scaling when modality-specific models fail to exhibit.

cs.LG

版本更新 · v2arXiv:2611.08685

Fibered Separation Logic: Weakest Preconditions, Universal Constructions, and Local Independence

John M. Li, Ryan Doenges, Pedro H. Azevedo de Amorim

原文摘要

This paper introduces a categorical framework for models of higher-order separation logic including the weakest precondition modality. Weakest preconditions are interpreted by strong monads on a category fibered over a model of Moggi's computational lambda calculus. The fibered categories in question are termed bunched fibrations and shown to be equivalent to BI hyperdoctrines. Well-known heuristic techniques for obtaining weakest preconditions are shown to be expressible in categorical terms and are proved sound. The ``baking in the frame rule'' trick is shown to satisfy a universal property. Constructions are given for extending a program logic with cost credits. Our largest case study is parametrized by a category having local independent product (LIP) structure, a notion due to Simpson that axiomatizes name generation, randomness, and local state. From an LIP category, we construct an allocation monad, a bunched fibration, and a monad lifting. We instantiate our construction to obtain a novel probabilistic separation logic for measure-theoretic probability that both simplifies and clarifies prior work.

cs.PL · cs.LO

版本更新 · v2arXiv:2611.09183

On derandomizing the line-point low-degree test

Prahladh Harsha, Swastik Kopparty, Mrinal Kumar, Ramprasad Saptharishi, Madhu Sudan

原文摘要

We investigate the derandomized version of the classical line-point low-degree test in the high-error regime in the bivariate setting and reveal some unexpected behavior. The classical line-point low-degree test takes an $m$-variate function $f$ over a finite field $\mathbb{F}_q$, queries it along a random line and accepts if the restriction of the function to this line is a low-degree polynomial. The high-error version of (the analysis of) this test asserts that if $f$ has at least $ε$ expected local agreement with a degree-$k$ univariate polynomial on a random line, then in fact $f$ is non-trivially close to some degree-$k$ $m$-variate polynomial even as $ε$ approaches $0$ (provided $q$ is sufficiently large compared to $k$ and $1/ε$). The derandomized version of the problem asks if the same holds when the line is not chosen uniformly over all possible lines but rather from a small set. In this paper we explore the bivariate case ($m=2$). We show, somewhat surprisingly, that there exists a function $f\colon \mathbb{F}_q^2 \to \mathbb{F}_q$ and a set of lines $\mathcal{L}$ in $\mathbb{F}_q^2$ of size nearly $q^{3/2}$ such that on every line in $\mathcal{L}$, $f$ has agreement $\approx 1/2$ on average with some univariate polynomial of degree $k=\sqrt{q}/2$ and yet $f$ is $1 - O(q^{-1/2})$ far from every polynomial of degree $O(k)$. On the positive side, we show that once the field size $q$ is at least $k^2/(ε)^{O(1)}$ such examples disappear and we can get sets of lines of size $O(q/ε^{O(1)})$ on which the low-degree test has a high-error analysis. On the way to our positive result, we show a new high-degree version of the classical two-direction Polishchuk--Spielman divisibility lemma.

cs.CC · cs.IT

新论文 · v1arXiv:2611.13437

5G CATS: 5G-Based Cryptographic Authorization Against Trafficking of Semiconductor Devices

Sreevikram Rajaraman, Derek Ladd, Nicholas Heinrich-Barna, Srinivas Shakkottai, Jeyavijayan Rajendran

原文摘要

Enforcing export controls on AI accelerators requires verifying where a chip operates after deployment. Round-trip time measurements can bound a chip's distance from a trusted landmark, but any jitter in honest replies forces the verifier to loosen its timeout, giving an attacker slack to relay traffic over fiber from a restricted region. Prior approaches either cross the untrusted host stack or rely on noisy power-line channels that inflate this slack. We present 5G CATS, a location certification system that pairs a co-packaged 5G radio, cryptographic engine, and hardware interlock on the GPU package with 5G Ultra-Reliable Low-Latency Communication (URLLC) base stations. Terminating the radio link inside the package removes the host software, PCIe fabric, and baseboard management controller from the verification path. We map an authenticated challenge-response protocol onto 5G's slot structure, replace permanent key erasure with short-lived execution leases, and specify a cross-check of the chip's reported serving cell against the carrier's own attachment records. In ns-3 and 5G-LENA simulations of an unloaded cell, our URLLC-based radio profile narrows the 99.9th percentile round-trip spread to 2.60 ms, more than ten times narrower than default 5G. A tracer rejects every round of a fiber relay from 300 km, against 2,000 km for default 5G. Against a 3,000 km attacker, a tracer stays safe at up to 1,410 km from the chip, against at most 619 km for PowerCATS's published model, while default 5G is never safe.

cs.CR · cs.NI

新论文 · v1arXiv:2611.13434

Minimum Synthesis Time for Single-Deletion Correcting Codes of Prescribed Size

Yuling Li, Wenjun Yu, Gennian Ge

原文摘要

DNA is a promising medium for data storage, but the high cost of DNA synthesis remains a major obstacle to its widespread deployment. In the reference-based synthesis model, synthesis time is measured by the length of the reference sequence. Previous studies have analyzed code sizes under synthesis-time constraints and constructed cost-constrained codes, including codes that correct deletions. In this paper, we study the minimum synthesis time for codes of prescribed size, considering both unrestricted codes and single-deletion correcting codes. We establish finite-length lower bounds valid for arbitrary reference sequences and upper bounds using existing $q$-ary differential VT codes. For fixed $q\geq3$, $A>0$, and $R\in(0,1)$, we determine the minimum synthesis time up to an $O(1)$ term when the code size is $M_n=\lfloor Aq^{Rn}\rfloor$. The corresponding redundancy is $(1-R)n\log_2q-\log_2A+o(1)$ bits. In particular, for these code sizes, the minimum synthesis time for single-deletion correcting codes exceeds that for unrestricted codes by a term proportional to $\ln n$, and we determine its coefficient. We also derive upper and lower bounds for code sizes $\lfloor q^n/(nq^{s_n})\rfloor$, where $s_n\to\infty$ and $s_n=o(n)$. These sizes correspond to redundancy $\log_2 n+s_n\log_2q+o(1)$ bits. For code sizes $\lfloor q^n/(nq^c)\rfloor$ with fixed constant $c$, the redundancy is $\log_2 n+c\log_2q+o(1)$ bits. We give conditions under which single-deletion correcting codes exist or do not exist for all sufficiently large $n$, together with bounds on the minimum synthesis time. These results show how code size and the single-deletion correcting constraint affect synthesis time and provide theoretical guidance for the construction of cost-constrained codes.

cs.IT

新论文 · v1arXiv:2611.13431

You Don't Need Feature Learning to Do In-Context Learning

Joseph Turnbull, Jamie Simon

原文摘要

How should we understand memorization vs. generalization in in-context learning (ICL) tasks, and when does one happen vs. the other? To investigate, we study how rotation-invariant kernel machines learn the ICL linear regression task proposed by Raventos et al. (2023). By decomposing the task in the kernel eigenbasis, we obtain a spectral picture of learning in this ICL setting. This picture reveals that as sample size grows, kernel machines learn successively-higher-degree components of the memorizing solution, which depend upon successively-higher-degree statistics of the training task set. The generalizing solution is approximately a low-degree truncation of the memorizing solution, and so the model first generalizes, then degrades into memorization as training proceeds. The takeaways are threefold. First, if this case is indicative, learning a "generalizing" solution on an ICL task is really just underfitting the memorizing solution! Second, the fact that kernel machines can generalize on this task means that feature learning is inessential to ICL. Third, transformers empirically trained to do ICL are shown to memorize tasks through the higher-order statistics of the training tasks that the kernel-based theory predicts, suggesting it applies even to models capable of feature learning.

cs.LG · cs.AI

新论文 · v1arXiv:2611.13430

Off-target posterior mass: A training-free diagnostic for fine-tuning spoken language diarization probes

Zhuoyan Tao, Aditya Kommineni, Shrikanth Narayanan

原文摘要

Spoken Language Diarization (SLD) models provide frame-level language information of a speech recording. SLD modeling pipeline includes feature extraction from pre-trained speech models followed by clustering. Fine-tuning pre-trained speech models has shown performance gains for SLD with its magnitude being dataset dependent. Off-target Posterior Mass (OTPM), computed based on posterior language identification probabilities is proposed as a diagnostic metric to forecast performance gains from SLD fine-tuning. OTPM obtains positive correlation with probe gains, 0.56 on unseen synthetic bilingual data and 0.23 on real-world corpora. These results indicate promising potential of OTPM as a diagnostic metric for SLD.

eess.AS · cs.CL · cs.SD

版本更新 · v2arXiv:2610.40658

Explain, Then Forget: Recommendation Unlearning under Asymmetric Access Constraint

Haichao Zhang, Chong Zhang, Wen Wang, Shi Qiu, Peiyu Hu, Jia Wang

原文摘要

Recommendation unlearning aims to remove the influence of specific historical interactions that should no longer affect future recommendations, such as noisy clicks or outdated user preferences. However, performing recommendation unlearning becomes much harder in modern Machine-Learning-as-a-Service (MLaaS) architectures, where the recommendation model is remotely hosted by a model provider (MP) and accessed by service operators (SO) only through APIs. In this setting, a service operator serves end users but cannot access or modify internal model parameters. As a result, existing unlearning methods are difficult to apply because they require access to model internals for retraining, fine-tuning, or other parameter updates. To address these challenges, we propose Causal Explanation-based Unlearning (CEU), a novel unlearning framework tailored for MLaaS environments. Instead of modifying internal model weights, CEU operates through a two-step output-level score intervention. First, it estimates how the forgotten interaction causally contributes to the current recommendation. Second, guided by this causal signal, the system removes the influence of the forgotten interaction by adjusting the final ranking scores of affected items. Because CEU leaves model parameters untouched, it naturally supports continuous unlearning requests while avoiding the error accumulation that can result from repeated parameter updates. Extensive experiments on three public datasets show that CEU achieves unlearning effectiveness comparable to full retraining while preserving recommendation utility and significantly reducing computational cost. The source code is available at: https://github.com/zhang-haichao/Explain-and-Forget.

cs.IR

新论文 · v1arXiv:2611.13426

BioMem: Agentic Memory for Biomedical LLM Agents

Mingchen Li, Yongkang Xiao, Rumeng Li, Rui Zhang, Hong Yu

原文摘要

Large language models (LLMs) have demonstrated considerable potential in biomedical applications, while agentic memory systems have emerged as a promising approach to preserving and utilizing information across tasks and interactions. However, a fundamental question remains: how effectively and reliably do these systems retain, retrieve, update, and utilize biomedical information? In this study, we systematically investigate agentic memory in the biomedical domain, focusing on task effectiveness, retrieval quality, information fidelity, and long-term robustness. We evaluate five memory approaches across four LLMs in three representative settings: biomedical knowledge reasoning, clinical prediction, and long-term conversational memory. Our analyses reveal how memory construction, evidence relevance, and conversational history shape downstream biomedical performance, highlighting the distinct challenges of retrieving relevant information, preserving its fidelity, and effectively utilizing it for reasoning. Building on these findings, we introduce Hybrid Memory Retrieval and Utilization (HMRU), a training-free framework that jointly enhances evidence accessibility and task-aware utilization to improve biomedical reasoning and long-term conversational memory. HMRU achieves an average relative improvement of 8.6% across biomedical question-answering and conversational memory benchmarks, with additional gains in selected clinical prediction tasks. It further reduces conversational answer errors by 24.9% and closes 48.0% of the performance gap to full-context inference. Our findings provide a systematic characterization of biomedical agentic memory and practical insights into developing more effective and reliable memory-augmented LLM systems.

cs.CL

新论文 · v1arXiv:2611.13425

Prompt2Echo (P2Echo): Text-Prompted Multi-View Echocardiography Segmentation with One Model

Moein Heidari, Alvin Kimbowa, Mahdi Tabatabaei, Afshin Bozorgpour, Sina Ghorbani Kolahi, River Jiang, Wenjin Chen, David J. Foran 等 11 位作者

原文摘要

Automated segmentation in multi-view echocardiography is essential for comprehensive cardiac assessment, yet remains challenging due to substantial anatomical and appearance variability across standard acoustic windows. In this paper, we present P2Echo, a unified framework capable of segmenting diverse cardiac structures across multiple standard echocardiographic views using a single model controlled by structured view-aware text prompts. Unlike recent foundation model adaptations that rely on implicit latent priors, P2Echo explicitly conditions the segmentation process via a novel Dynamic Image-Text Alignment (DyITA) module. DyITA integrates a differentiable routing mechanism and token differential operators to dynamically modulate decoder features according to the requested view and target structure. This allows the network to adaptively switch its processing pathways for different views without architectural changes. Extensive experiments on a large-scale internal dataset (seen during training) and held-out evaluation on three external datasets demonstrate that P2Echo matches or exceeds both view-specific specialists and state-of-the-art universal models at a fraction of their computational cost, offering an efficient, controllable solution for multi-plane cardiac analysis. Code is available at: https://github.com/moeinheidari7829/P2Echo.

cs.CV

版本更新 · v2arXiv:2611.01399

Deterministic $n^{O(\log n)}$ Graph Isomorphism Algorithm via Inherited Decompositions

MohammadTaghi Hajiaghayi, Mahdi JafariRaviz

原文摘要

We give a deterministic $n^{O(\log d)}$-time algorithm for isomorphism of graphs with $n$ vertices and maximum degree $d$. Since $d<n$, this is an $n^{O(\log n)}$-time algorithm for graph isomorphism. To our knowledge, this is the first proof of such a worst-case bound for general graphs. It improves the explicit cubic logarithmic exponent in the analysis of the corrected Luks--Babai algorithm. For graphs of maximum degree $d$, the best previous bound known to us, due to Grohe, Neuen, and Schweitzer, is $n^{O((\log d)^c)}$ for an absolute constant $c$. The algorithm uses the local-certificate and color-refinement steps of Babai's quasipolynomial algorithm, but retains the decomposition of each recursive instance when it solves an auxiliary coloring problem. There are two reasons for retaining this information. First, a local test can make progress without removing many vertices: it can eliminate a layer of permutation symmetry. A logarithmic bound on the total weight of these layers allows tests of constant size. Second, the auxiliary bipartite graphs arising in the symmetry-breaking step can be represented inside the current instance, after retaining a row decomposition when necessary. These recursive calls inherit the progress already made. Two further observations give the dependence on the degree. The recursive calls made by a local test check disjoint parts of the instance, so together they cost about as much as one call on the whole instance. And for graphs of maximum degree $d$, each large symmetric action that the algorithm must break acts on at most $d$ points. We present the algorithm through colorings, partitions, and bipartite graphs, and collect the permutation-group ingredients in one appendix. The result extends to string isomorphism and returns one isomorphism and generators for the full automorphism group.

cs.DS