c

cs.CL updates on arXiv.org热榜

更新于刚刚 · 50 条热点

10月3日 11:40 更新，当前榜单前三是「Large Language Mode…」。「TomasuLLM: Out-of-O…」。「Automatic estimatio…」。这是一个人工智能方向的信息源，内容以模型进展、研究与工具发布为主。本页当前收录 50 条，保留cs.CL updates on arXiv.org的原始排序和指标，不把不同平台的数据直接相加，每条标题尽量直达原始页面。热摸爽只做聚合与排序，内容版权归cs.CL updates on arXiv.org及原作者所有。

也常被搜索为：cs.CL updates on arXiv.org热点

Large Language Models are Approximate Survival Estimators

cs.CL updates on arXiv.org 1479.1 万热度发布于 23小时前抓取于刚刚

TomasuLLM: Out-of-Order Speculative Execution for LLM Agents

cs.CL updates on arXiv.org 1413.5 万热度发布于 23小时前抓取于刚刚

Automatic estimation of verbal fluency index in people with Motor Neuron Disease using ASR alignment and pause modelling

cs.CL updates on arXiv.org 1349.6 万热度发布于 23小时前抓取于刚刚

The System Prompt Illusion: How Instruction Preambles Modify Computation in Language Models

cs.CL updates on arXiv.org 1287.2 万热度发布于 23小时前抓取于刚刚

TutlAit v1: a crowdsourced Moroccan Tamazight speech dataset with Arabic transcriptions and regional accent labels

cs.CL updates on arXiv.org 1226.4 万热度发布于 23小时前抓取于刚刚

Conformal Factuality Control for Multi-Hop Retrieval-Augmented Generation

cs.CL updates on arXiv.org 1167.3 万热度发布于 23小时前抓取于刚刚

Framing the Narrative: Ideological Mimicry in Large Language Models

cs.CL updates on arXiv.org 1109.7 万热度发布于 23小时前抓取于刚刚

ContextAdapt: Evaluating Contextual Adaptation and Value Alignment in LLMs

cs.CL updates on arXiv.org 1053.7 万热度发布于 23小时前抓取于刚刚

NinaXander: Feasibility and Limits of Composing Frozen Language Models Across Architecture Families via a Shared Latent Space

cs.CL updates on arXiv.org 999.3 万热度发布于 23小时前抓取于刚刚

Which Models Work Well Together? Measuring Heterogeneity for LLM Team Selection

cs.CL updates on arXiv.org 946.5 万热度发布于 23小时前抓取于刚刚

EVOKE: Eliciting World Knowledge in Agents for Transferable Decision-Making

cs.CL updates on arXiv.org 895.2 万热度发布于 23小时前抓取于刚刚

Halluscoring 2026: The first shared task on llms hallucination detection and answer verification

cs.CL updates on arXiv.org 845.5 万热度发布于 23小时前抓取于刚刚

Evaluating Language Model Safety Across Long Adversarial Conversations

cs.CL updates on arXiv.org 797.3 万热度发布于 23小时前抓取于刚刚

Doc2LoRA Provides Decodable Representations of Scientific Ideas

cs.CL updates on arXiv.org 750.7 万热度发布于 23小时前抓取于刚刚

Evaluating Whether LLMs Can Reliably Connect the DOTs?

cs.CL updates on arXiv.org 705.6 万热度发布于 23小时前抓取于刚刚

Policy-Conditioned AI-Use Detection: An Evidentiary Framework for Academic Publishing

cs.CL updates on arXiv.org 662 万热度发布于 23小时前抓取于刚刚

The Backdrop Exposes What the World Around an Agent Costs It

cs.CL updates on arXiv.org 619.9 万热度发布于 23小时前抓取于刚刚

KlinikeBench: Evaluating Language Models Beyond Diagnostic Accuracy

cs.CL updates on arXiv.org 579.3 万热度发布于 23小时前抓取于刚刚

Personalized State-Transition-Aware Memory for Clinical Agents

cs.CL updates on arXiv.org 540.2 万热度发布于 23小时前抓取于刚刚

DEdit: Iterative Draft Editing for Speculative Decoding

cs.CL updates on arXiv.org 502.6 万热度发布于 23小时前抓取于刚刚