Large language models

Model releases, architecture work and prompting research on large language models — from frontier-lab announcements to the arXiv papers behind them.

28,678 stories · RSS feed

arXiv AI

Oct 2

Cog-VADU: A Training-Free Cognitive Reasoning Framework for Video Anomaly Detection and Understanding

Cog-VADU is a training‑free framework that transforms video anomaly detection into a sequential cognitive reasoning task. It uses Chain‑of‑Anomaly Detection Thought Prompting (CoADTP) to create a recurrent reasoning chain across video segments, preserving temporal memory and distinguishing complex anomalies from high‑motion normal activities. A cross‑modal re‑ranking stage aligns textual rationales with visual embeddings to enforce semantic consistency and temporal coherence, yielding competitive zero‑shot performance on multiple VAD benchmarks.

By Mohd Ubaid Wani, Sara Atito, Josef Kittler, Muhammad Awais

llms rag multimodal benchmarks

arXiv AI

Oct 2

VETO: Video Efficient Token Optimization for Vision Language Models

VETO (Video Efficient Token Optimization for Vision Language Models) is a plug‑in that reduces the quadratic cost of visual tokens in long‑video inference by applying dual‑axis compression: an intra‑frame compressor merges semantically similar tokens within each frame, and an inter‑frame compressor merges temporally redundant frames. By first compressing spatial dimensions, VETO lowers the cost of subsequent global temporal matching, surpassing single‑axis methods and achieving up to 45% faster inference on models such as LLaVA‑OneVision‑7B while maintaining or improving accuracy. The approach is universally applicable across LLaVA‑OneVision, InternVL‑2.5, and LongVA, preserving or enhancing zero‑shot accuracy even under extreme token budgets.

By Gueter Josmy Faure, Hao Ping Wang, Min-Hung Chen, Winston H. Hsu

llms multimodal

arXiv AI

Oct 2

A rubric landscape for evaluating clinical reasoning in large language models: what exists, what is missing, and what needs to be combined

The article reviews existing tools for assessing clinical reasoning in large language models, identifying six key dimensions—problem representation, temporal synthesis, differential and management reasoning, counterfactual reasoning, calibrated uncertainty, and reasoning faithfulness. It finds that no single instrument covers all dimensions, with some aspects reasonably addressed while others, especially faithfulness and counterfactual reasoning, remain weakly evaluated. The authors propose combining current tools through binary rubric items, separate completeness and correctness scores, importance weighting, safety caps, temporal consistency checks, and chance‑corrected reliability reporting, and call for further design work on uncertainty, counterfactual reasoning, and faithfulness in longitudinal free‑text contexts.

By Zhangshu Joshua Jiang, Zina Ibrahim, James T. Teo

llms benchmarks safety

arXiv AI

Oct 2

MMMG: a Comprehensive and Reliable Benchmark for Multitask Multimodal Generation

arXiv:2505.17613v2 Announce Type: replace Abstract: Automatically evaluating multimodal generation presents a significant challenge, as automated metrics often struggle to align with human evaluation...

By Jihan Yao, Yushi Hu, Wenyuan Wang, Bin Han, Shangbin Feng, Guang Yang, Yujie Yi, Bingbing Wen, Ranjay Krishna, Lucy Lu Wang, Yulia Tsvetkov, Noah A. Smith, Banghua Zhu

llms diffusion multimodal benchmarks

arXiv AI

Oct 2

Scalable Delphi: Large Language Models for Structured Risk Estimation

arXiv:2602.08889v2 Announce Type: replace Abstract: Quantitative risk assessment relies on structured expert elicitation to estimate unobservable properties. The Delphi method produces calibrated, au...

By Tobias Lorenz, Mario Fritz

llms benchmarks

arXiv AI

Oct 2

Skin-Deep: A Geometric Diagnostic for Alignment Fragility in Large Language Model Representations

arXiv:2606.22676v2 Announce Type: replace Abstract: Refusal on a safety benchmark does not reveal how stable that behavior will remain after model updates. Benign downstream fine-tuning can weaken re...

By Dongyub Jude Lee, Jungseob Lee, Seungyoon Lee, Seongtae Hong, Suhyune Son, Sugyeong Eo, Jaehyung Seo, Heuiseok Lim

llms fine-tuning benchmarks safety

arXiv AI

Oct 2

MINCE: Shrinking LLM Evaluation Datasets via Few-Model Monte Carlo Calibration

arXiv:2606.22826v2 Announce Type: replace Abstract: Evaluating LLMs across many model variants---quantized, fine-tuned, or deployment-specific---requires running large benchmarks repeatedly, a proces...

By Devleena Das, Rajeev Patwari, Vikram Kumar Bukka, Nithin Kumar Guggilla, Elliott Delaye, Ashish Sirasao

llms fine-tuning efficiency benchmarks

arXiv AI

Oct 2

Program Learning with Verifiable Rewards: Symbolic Backpropagation for Post-Training LLMs

arXiv:2608.28421v2 Announce Type: replace Abstract: Post training a language model to reason means updating its weights. Supervised finetuning and reinforcement learning both place the acquired capab...

By Vishvesh Bhat

llms reinforcement-learning fine-tuning benchmarks

arXiv AI

Oct 2

Not Too Hard, Not Too Easy: Learning from Intermediate States for LLM Structured Reasoning

arXiv:2609.33149v2 Announce Type: replace Abstract: A common principle of effective learning is to practice material that is neither already mastered nor too difficult to permit progress. We ask how...

By Hongbo Chen, Guohua Lu, Ting Dang, Hong Jia

llms fine-tuning

arXiv AI

Oct 2

In Dialogue with Intelligence: Toward Insightful Co-Augmentation

arXiv:2505.22767v4 Announce Type: replace-cross Abstract: Dialogue with a large language model can lead a person to insight: a sudden change in how they understand a problem. This perspective asks ho...

By Eleni Vasilaki

llms

arXiv AI

Oct 2

GenGait: A Transformer-Based Model for Human Gait Anomaly Detection and Normative Twin Generation

arXiv:2604.01997v2 Announce Type: replace Abstract: Gait analysis provides an objective characterization of locomotor function and is widely used to support diagnosis and rehabilitation monitoring ac...

By Elisa Motta, Marta Lorenzini, Clara Mouawad, Alberto Ranavolo, Mariano Serrao, Arash Ajoudani

llms

arXiv AI

Oct 2

ECHO: A Participatory Framework for Bias-Anchored AI Harm Anticipation

arXiv:2512.03068v2 Announce Type: replace-cross Abstract: Artificial Intelligence (AI) systems increasingly shape consequential decisions, creating value but also potential harms for individuals, soc...

By Nicoleta Tantalaki, Sophia Vei, Athena Vakali

llms safety

arXiv AI

Oct 2

Measuring Iterative Temporal Reasoning with Time Puzzles

arXiv:2601.07148v4 Announce Type: replace-cross Abstract: Tool use, such as web search, has become a standard capability even in freely available large language models (LLMs). However, existing bench...

By Zhengxiang Wang, Zeyu Dong

llms agents benchmarks

arXiv AI

Oct 2

MedVL-SAM2: A unified 3D medical vision-language model for multimodal reasoning and prompt-driven segmentation

arXiv:2601.09879v2 Announce Type: replace-cross Abstract: Recent progress in medical vision-language models (VLMs) has achieved strong performance on image-level text-centric tasks such as report gen...

By Yang Xing, Jiong Wu, Savas Ozdemir, Ying Zhang, Yang Yang, Wei Shao, Kuang Gong

llms rag computer-vision nlp multimodal benchmarks

arXiv AI

Oct 2

InterviewSim: A Scalable Framework for Interview-Grounded Personality Simulation

arXiv:2602.20294v2 Announce Type: replace-cross Abstract: Simulating real personalities with large language models requires grounding generation in authentic personal data. Existing evaluation approa...

By Yu Li, Pranav Narayanan Venkit, Yada Pruksachatkun, Chien-Sheng Wu

llms rag safety

arXiv AI

Oct 2

Large Reward Models: Generalizable Online Robot Reward Generation with Vision-Language Models

arXiv:2603.16065v3 Announce Type: replace-cross Abstract: Reinforcement Learning (RL) has shown strong potential for improving robotic manipulation policies, yet its practical use remains bottlenecke...

By Yanru Wu, Weiduo Yuan, Esteban Martinez Licon, Ang Qi, Vitor Guizilini, Jiageng Mao, Yue Wang

llms reinforcement-learning robotics fine-tuning multimodal benchmarks

arXiv AI

Oct 2

High Volatility and Action Bias Distinguish LLMs from Humans in Group Coordination

arXiv:2604.02578v2 Announce Type: replace-cross Abstract: Humans exhibit remarkable abilities to coordinate in groups. As large language models (LLMs) become more capable, it remains an open question...

By Sahaj Singh Maini, Robert L. Goldstone, Zoran Tiganj

llms safety

arXiv AI

Oct 2

Verbal tics in frontier language models: A critical review of current releases, research evidence, and public discussion

arXiv:2604.19139v4 Announce Type: replace-cross Abstract: Repeated praise, canned reassurance, familiar contrasts, and conspicuous vocabulary are recurring subjects in discussions of large language m...

By Shuai Wu, Xue Li, Zhijun Wang, Bolun Liu, Weilin Cai, Zihao Su, Ran Wang

llms benchmarks

arXiv AI

Oct 2

What Does a Skill Actually Do? Estimands and Evaluation Validity for Tool and Skill Use in LLM Agents: A Critical Review

arXiv:2609.33153v2 Announce Type: replace-cross Abstract: Reported improvements from tools and reusable skills in large language model agents refer to different comparisons. This critical narrative r...

By Shuyang Zhang

llms agents benchmarks

arXiv AI

Oct 2

Unifying Video Tasks via Spatiotemporal Analogy

arXiv:2609.33935v2 Announce Type: replace-cross Abstract: Adapting video models to new tasks typically requires dedicated data curation and fine-tuning. While visual analogy provides a training-free...

By Chia-Hsiang Kao, Belinda Zeng, Bharath Hariharan, Menglin Jia

llms fine-tuning

Get the daily briefing