Large language models

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

32,584 stories · RSS feed

arXiv AI

Sep 16

Safe Error Correction for Language Models: Frozen-Base Adjustment with Capability Preservation

arXiv:2609.16145v1 Announce Type: new Abstract: We study a practical question: can a small correction module fix errors in a frozen language model's outputs without degrading its base capabilities? W...

By Gautam Kishore

llms reinforcement-learning robotics fine-tuning benchmarks

arXiv AI

Sep 16

Fine-Tuning Fixes Mode Collapse and Over-Dispersion in LLMs

arXiv:2609.16454v1 Announce Type: new Abstract: Recent work by Doshi and Hauser (2024), Bisbee et al. (2024), and Xie et al. (2026) raises concerns that outputs from large language models (LLMs) tend...

By Kirill Skobelev, Eric Fithian, X. Y. Han

llms fine-tuning safety

arXiv AI

Sep 16

Interactive Memory Learning for Long-Term Conversations

arXiv:2609.17088v1 Announce Type: new Abstract: Recent advancements in large language models have significantly enhanced the capabilities of agents in modeling long-term conversations. Despite these...

By Cai Ke, Jiangyue Yan, Han Zhang, Xin Liu, Zike Yuan, Yue Yu, Hui Wang, Ruifeng Xu

llms agents reinforcement-learning

arXiv AI

Sep 16

Extracting ontology-compliant knowledge from scientific text describing irradiated materials using large language models

arXiv:2609.17291v1 Announce Type: new Abstract: The quest for new materials increasingly relies on predictive models and comprehensive simulations that span scales from atomic to macroscopic levels....

By Marco Luca Sbodio, Marcos Mart\'inez Galindo, Vanessa Lopez, Blanca Biel, Pablo Canca, Pedro Delgado, Jes\'us I. Mendieta-Moreno, Raphael Tack, Maria J. Caturla

llms benchmarks

arXiv AI

Sep 16

Verifiable Social Reasoning for LLM Assistants

arXiv:2609.17496v1 Announce Type: new Abstract: LLM assistants are widely used for daily social advice, yet evaluating their social reasoning in such consultation settings remains challenging since (...

By Amir Taubenfeld, Zorik Gekhman, Avigail Grinstein-Dabush, Itay Laish, Ariel Goldstein, Marian Croak, Avinatan Hassidim, Yossi Matias, Amir Feder

llms agents

arXiv AI

Sep 16

Competence-Preserving Resume Perturbations Expose Presentation Sensitivity in LLM Screening

arXiv:2609.16517v1 Announce Type: cross Abstract: Resume screeners must infer job-relevant competence from resumes whose presentation can vary substantially in wording, structure, stylistic polish, a...

By Qiangju Chen, Yang Xiao

llms

arXiv AI

Sep 16

RepoAtlas: Guiding Coding Agents via Evolving Multimodal Repository Views

arXiv:2609.16936v1 Announce Type: cross Abstract: Large language model (LLM)-powered coding agents have made rapid progress in automating software engineering tasks, yet repository-level issue resolu...

By Yunxiang Zhang, Haiquan Wang, JiaWei Guo, Hanyang Xia, Yan Chen, Tong Chen, Zhang Zhiwei, Junchen Ye

llms agents multimodal

arXiv AI

Sep 16

Skill-based Agentic Evaluation for Real-time Data Science Tasks

arXiv:2609.16487v1 Announce Type: new Abstract: We present a framework for evaluating data-science agents on live, continuously updated data using executable ground truth and format-agnostic factoid...

By Aniruddha Tamhane, Raghavendra Addanki, Ayushi Aggarwal, Aditya Bansal, Rui Wang, Charles Menguy, Swati Jain

llms agents

arXiv Machine Learning

Sep 16

Divergence Timing and Cumulative Disagreement under KV-Cache Eviction

The paper studies how evicting entries from a KV‑cache affects the token distributions used in autoregressive language models. It derives an exact decomposition showing that the expected fraction of mismatched tokens equals a first‑mismatch contribution plus a post‑divergence exposure term weighted by its mismatch rate. Experiments on Meta‑Llama‑3.1‑8B‑Instruct and Qwen2.5‑7B‑Instruct demonstrate that a 50% retention strategy (SnapKV) delays divergence and reduces its frequency compared to other retention schemes, while post‑divergence total variation remains high and dominates the overall mismatch gap.

By Xinyue Luo, Fei Yu

llms

arXiv AI

Sep 16

VideoMM: Adaptive Macro-Micro Inference for Efficient Video MLLMs

arXiv:2609.16722v1 Announce Type: new Abstract: Scaling Multimodal Large Language Models (MLLMs) to long-form video understanding is bottlenecked by the explosion of visual tokens, which saturates co...

By Haoyu Guo, Yuan Feng, Junlin Lv, Mingjun Xiao, S Kevin Zhou, Xike Xie

llms multimodal

arXiv AI

Sep 16

Layers, Sinks, and Scaling: Adaptive Evidence Selection for Multimodal Large Language Models

arXiv:2609.16795v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) can answer knowledge-intensive visual questions by combining visual evidence from images with facts retrieved...

By Zhenbin Wang, Lei Zhang, Lituan Wang, Wei Huang, Yan Wang, Zhenwei Zhang

llms multimodal benchmarks

arXiv AI

Sep 16

SKIP: a Self-knowledge-guided Step-wise Preference Learning Framework for Concise Reasoning

arXiv:2609.17019v1 Announce Type: new Abstract: While Chain-of-Thought (CoT) reasoning has been proven to be effective, it often leads to overthinking, resulting in computational overhead, inference...

By Qinhong Lin, Yuhao Zhang, Yinglun Feng, Zhongliang Yang, Linna Zhou

llms reinforcement-learning fine-tuning efficiency benchmarks

arXiv AI

Sep 16

Efficient Reasoning Distillation: Small Video-Language Models via Synthetic CoT and Difficulty-Aware Fine-Tuning

arXiv:2609.16255v1 Announce Type: cross Abstract: We present an efficient method to distill reasoning capabilities into compact video-language models (VLMs) for video question answering (VideoQA). Ou...

By Mantek Singh, Jeshwanth Challagundla, Siddharth Raina, Jasmin Jarsania

llms nlp fine-tuning efficiency safety

arXiv AI

Sep 16

Early-Bird Decoding: Accelerating Diffusion LLMs with Learnable Block Sizes and Parallel Sampling

arXiv:2609.16450v1 Announce Type: cross Abstract: Diffusion large language models (dLLMs) offer a promising parallel decoding paradigm as an alternative to autoregressive generation through iterative...

By Lixuan Wei, Wei Zhou, Jianwen Wu, Yipeng Shen, Meiling Wang, Haoran You

llms diffusion efficiency benchmarks

arXiv AI

Sep 16

StackTok: Accelerating VLMs Inference with Budget-Adaptive Visual Token Selection

arXiv:2609.16841v1 Announce Type: cross Abstract: Increasing image resolution produces ever-longer visual-token sequences in vision-language models (VLMs), substantially raising their inference cost....

By Zhenbin Wang, Lei Zhang, Lituan Wang, Wei Huang, Yan Wang, Zhenwei Zhang

llms multimodal benchmarks

arXiv AI

Sep 16

Same Answer, Different Representations: Hidden instability in VLMs

arXiv:2602.06652v2 Announce Type: replace Abstract: The robustness of Vision Language Models (VLMs) is commonly assessed through output-level invariance, implicitly assuming that stable predictions r...

By Farooq Ahmad Wani, Alessandro Suglia, Rohit Saxena, Aryo Pradipta Gema, Wai-Chung Kwan, Fazl Barez, Maria Sofia Bucarelli, Fabrizio Silvestri, Pasquale Minervini

llms rag multimodal benchmarks safety

arXiv AI

Sep 16

Thinking Deeper, Not Longer: Memory-Efficient Test-Time Reasoning with Depth-Recurrent Transformers for Compositional Generalization

arXiv:2603.21676v2 Announce Type: replace-cross Abstract: Standard Transformers have a fixed computational depth, limiting their ability to generalize to tasks that require variable-depth reasoning....

By Hung-Hsuan Chen

llms safety

arXiv Machine Learning

Sep 16

Agentic Search Spaces for Tabular Machine Learning

arXiv:2609.16309v1 Announce Type: new Abstract: Despite the rapid progress of LLM-based agents for planning, code generation, and debugging, their practical value for tabular machine learning remains...

By Renat Sergazinov, Artem Chistyakov, Sergey Pankevich, Artem Babenko

llms rag agents benchmarks

arXiv Machine Learning

Sep 16

FlowATC: Aircraft Trajectory Prediction via Flow Matching

arXiv:2609.16528v1 Announce Type: new Abstract: Building accurate decision-support tools for next-generation air traffic control requires robust trajectory prediction models. We present a flow-matchi...

By Mathurin Petit, Emir Torun, Louis Brusset, Jordan Kam, Alexandre M. Bayen

llms diffusion

arXiv Machine Learning

Sep 16

Large Language Models Develop Belief State Geometry In-Context

arXiv:2609.17376v1 Announce Type: new Abstract: Large language models (LLMs) trained on next-token prediction exhibit remarkable in-context learning (ICL) abilities, yet the representations that supp...

By Daniel Balcells, Andrew Jun Lee, Chirag Rastogi, Paul M. Riechers, Adam Shai, Xavier Poncini

llms

Get the daily briefing