AI Research Papers
New papers daily from arXiv and Hugging Face Daily Papers, plus posts from top AI lab research blogs.
New AI research papers from arXiv and Hugging Face Daily Papers, updated every two hours.
- arXiv · 27 Oct 2026When All Rollouts Fail: Policy-Compatible Joint Decoding for Hard-Problem Reasoning
Zehao Wang, Hanzhu Chen, Runquan Gui, Zhe Wang, Lin Yang
- arXiv · 27 Oct 2026Evolving Agent Teams
Shiyi Cao, Ziming Mao, Dacheng Li, Joseph E. Gonzalez, Ion Stoica
- arXiv · 27 Oct 2026Conservative Policy Iteration For Zero-Sum POSGs
Erwan C. Escudie, Matthia Sabatelli, Jilles Steeve Dibangoye
- arXiv · 27 Oct 2026When the Critic Falls Behind: Co-Evolving Rubrics and Harnesses for Continual Agent Self-Improvement
Hao Chen, Yukun Yan, Ziyu Han, Ye He, Chuanzhe Guo, Shangzhan Li, Qi Shi, Qingfu Zhu
- arXiv · 27 Oct 2026Who Knows What - and What Does It Support? Perspective-Bounded Evidence Calibration for Multimodal Theory of Mind
Ziqi Liao, Jiaqi Shen, Yiming Na, Kedi Lyu, Shao-Yuan Lo, Hongxia Xie
- arXiv · 27 Oct 2026Words or Deeds? How LLMs Predict Users' Decisions in Human-Agent Dialogue
Baiqiao Zhang, Shuai Ma, Zhifeng Liao, Chao Zhou, Jianshu Wang, Yuhan Zhao, Chaoyue Zhang, Tongjun Si
- arXiv · 27 Oct 2026Weakness Localization Is Not Enough: State-Aware Recursive Knowledge Improvement for Large Language Models
Runze Jiang, Yan Wang, Boran Jin, Bingyu Zhu, Longtao Huang
- arXiv · 27 Oct 2026Latent Visual Contrastive Reasoning with Multi-Branch Exploration
Kegeng Tang, Jingbo Wang, Yangming Zhang, Huanshu Zhang, Shaogang Ren, Zihao Wang
- arXiv · 27 Oct 2026SetGRPO: Calibrating Grader-Induced Policy Optimization
Jiale Huang, Zixu Li, Zhiwei Chen, Zhiheng Fu, Jinhe Bi, Yupeng Hu
- arXiv · 27 Oct 2026LLM knowledge for improved Out-of-Distribution Tabular Prediction - A Cross-Validated Regularisation Approach
Sølve Langseth, Kristoffer Herland Hellton, Hugo L. Hammer
- arXiv · 27 Oct 2026Personality Latents: Conditioning Frozen Language Models on Individuals from Their Own Writing
Gunhee Cho, Yun-Gyung Cheong
- arXiv · 27 Oct 2026API-TAIL: Detecting Model Substitution in LLM APIs through Temperature-Aligned Likelihood Tails
Haocheng Mei, Yuchuan Tian, Ye Yuan, Chao Xu, Allen Chen, Kai Han, Yunhe Wang
- arXiv · 27 Oct 2026Language Models Keep Abandoned Analyses: Residue and Adaptation after Garden-Path Sentences
Xiang Ding
- arXiv · 27 Oct 2026Select-TTT: Query-Conditioned Selective Fast Weight Updates for In-Place Test-Time Training
Yue Wu, Yang Zhang, Weinan Gan, Chongming Gao, Xinsen Guo, Xiaoyuan Li, Fuli Feng
- arXiv · 27 Oct 2026Does LLM Pre-Training Typically Occur at the Edge of Stability?
Yuhang Cai, Haodong Wen, Haofeng Huang, Deyi Liu, Yiyuan Ma, Kaifeng Lyu
- arXiv · 27 Oct 2026Dependency Graphs of Mathematical Arguments: Representation, Structural Measurement, and Proof Review
Eunji Park
- arXiv · 27 Oct 2026Can LLMs Reliably Grade Olympiad Proofs? A Study of Mathematical Verification with LLMs
Azim Ospanov, Zijin Feng, Ding Ding, Chengwu Liu, Haoli Bai, Jiacheng Sun, Lifeng Shang, Farzan Farnia
- arXiv · 27 Oct 2026Beyond the Evidence Coherence Trap: Agent-Driven Evidence Selection for Temporally Disordered Fake News Detection
Chi Wang, Min Gao, Yifei Zhang, Bincheng Gu, Zongwei Wang, Wentao Li
- arXiv · 27 Oct 2026The Vocabulary Is Already Here: Adapting Pretrained Interpretable Features for Generative Recommendation
Ziluowen Luo, Ruochen Liu, Jun Yin, Hao Yan, Zhengxuan Hu, Xinze Shi, Chaozhuo Li, Chengqi Zhang
- arXiv · 27 Oct 2026Frequency-Uniform Log-Regret Reinforcement Learning for Diffusions
Jose Blanchet, Lutong Hao, Renyuan Xu
- arXiv · 27 Oct 2026Robust LLM Data Selection via Content-Invariant Gradient Attribution
Jia Gu, Liang Pang, Huawei Shen
- arXiv · 27 Oct 2026CAMR: Compatibility-Aware Memory Replay for Continual Learning in Large Language Models
Jingpu Yang, Tianhao Li, Fengxian Ji, Yufeng Wang, Shuxin Zheng, Xiaoqing Zhang
- arXiv · 27 Oct 2026From Voiceprints to User Entities: Online Speaker-Memory Construction for Multi-User Spoken Agents
Yucong Zhang, Cancan Li, Bo Xiao, Ke Zeng, Ming Li
- arXiv · 27 Oct 2026Is Token Confidence Just Chain Length? A Signal-Processing Audit of Reasoning Traces
Pragna Mallappa
- arXiv · 27 Oct 2026Conservative Dueling Bandits: Preference Learning under Baseline Constraints
Rohan Deb, Mohammad Ghavamzadeh, Arindam Banerjee
- arXiv · 27 Oct 2026Specialization, Diversity, and Scaling in Child-Scale Language Models
Steven Y. Feng, Michael C. Frank
- arXiv · 27 Oct 2026A Quantitative Approach to Asch Conformity in Frontier Large Language Models
Senthil Rajasekaran, Stylianos Loukas Vasileiou
- arXiv · 27 Oct 2026Japanese Discourse Structure Analysis Based on Segmented Discourse Representation Theory Using LLMs
Satomu Tomisaki, Yosuke Mikami, Amane Watahiki, Hitomi Yanaka
- arXiv · 27 Oct 2026Remembering Is Not Enough: Refining Memories and Learning When to Use Them in Multi-Agent Systems
Yipeng Hao, Yuxuan Cai, Wenkai Ji, Jie Zhou, Liyang Yu, Qin Chen, Zhiguo Yan, Shulei Zheng
- arXiv · 27 Oct 2026God Does Not Play Dice: Enforcing Batch Invariance in Agentic Kernel Generation
Youwei Bao, Chengpeng Wang, Simin Chen
From AI lab blogs
- Microsoft Research · 30 Oct 2026 · labMicrosoft Research Focus – Edition #97
Host Sowbhagya Suresh shares some of the latest news and cutting-edge research from across Microsoft—from understanding the challenges facing coding agents to uncovering how people use AI at work and improving extended-range weather forecasting. The post Microsoft Research Focus
- Hugging Face Blog · 29 Oct 2026 · labGranite Switch 4.1.1 — Update models like you update software
- OpenAI News · 29 Oct 2026 · labHow we detect frontier model misalignment
OpenAI uses layered monitoring to detect frontier model misalignment and unauthorized agent behavior so responders can intervene before incidents escalate.
- OpenAI News · 29 Oct 2026 · labHow Ropes & Gray is developing AI for transactional diligence
Ropes & Gray is developing AI for transactional diligence with OpenAI. Lawyers review grouped documents, summaries, and reports before advising clients.
- Hugging Face Blog · 29 Oct 2026 · labAnnouncing Project ProtoStar: Making Publicly Funded Science AI-Ready
- Hugging Face Blog · 29 Oct 2026 · labWelcome Le Chonk: Mistral Large 4 on Hugging Face
Top AI blogs
Latest posts from OpenAI, Google, DeepMind, Hugging Face, BAIR, Lil'Log, Simon Willison, Import AI, Latent Space and more, via RSS.
- Latent Space · 30 Oct 2026 · blog[AINews] Building AIE CODE: The top Coding Agent Conference
Mistral’s open-weight release, frontier monitoring, and the day’s AI Twitter and Reddit recaps.
- Simon Willison · 30 Oct 2026 · blogAPI credits benefit for eligible ChatGPT subscriptions
API credits benefit for eligible ChatGPT subscriptions Earlier this month Anthropic announced a generous new benefit for their Max and Team subscriptions: $100, $200, and $500 per month subscribers would get the equivalent quantity of (expiring) credits to use with the Anthropic
- AWS Machine Learning Blog · 30 Oct 2026 · blogHow the UK Amazon Quick User Group grew in 2026
The UK Amazon Quick User Group's 2026 roundup: three quarterly meetups, 150+ attendees, and customer stories from Arriva, Carsa, and Propulse Lab on using agentic AI and Amazon Quick to turn data into faster decisions, higher satisfaction, and scalable growth.
- AWS Machine Learning Blog · 29 Oct 2026 · blogBuilding intent-driven UI testing with Amazon Nova Act
Epsilon's Agentic Platform team delivers UI testing as a managed, multi-tenant service built on Amazon Nova Act and Amazon Bedrock AgentCore Browser Tool. Tests are written in plain English and run on browsers managed by AWS, giving every product team resilient, intent-driven tes
- AWS Machine Learning Blog · 29 Oct 2026 · blogBuild a repeatable PHP-to-Java migration engine with Kiro
Turn a PHP-to-Java migration from a one-time leap of faith into a measurable, repeatable process with Kiro. This post walks through a reusable engine: a mechanical discovery scanner and an eight-phase, parity-first workflow that treats the original PHP application as the specific
- Simon Willison · 29 Oct 2026 · blogBuilding a multiplayer game world with autonomous Fable-class models
How I Built the Most AI-Pilled Company in the World Don't be put off by the clickbait headline, this piece by Jacob Sansbury describing how they are building Spawn is a fascinating exploration of the bleeding edge of what's possible with agentic engineering against frontier model
- AWS Machine Learning Blog · 29 Oct 2026 · blogLend and borrow resources across teams with AWS Batch’s quota management for SageMaker Training jobs
AWS Batch quota management is a new scheduling capability you can use to divide GPU capacity into quota shares per team or project, automatically lend idle capacity across shares, and use preemption to run your highest-priority SageMaker Training jobs first, all without manual co
- AWS Machine Learning Blog · 29 Oct 2026 · blogHow MoEngage turned over 400 APIs into agent tools with AgentCore Gateway
MoEngage turned roughly 400 private REST APIs across 7 AWS Regions into permission-scoped agent tools with Amazon Bedrock AgentCore Gateway. No per-service MCP servers, no public endpoints, and no 400-tool context bloat. See how OpenAPI specs, Cedar policies, and managed Amazon V
- AWS Machine Learning Blog · 29 Oct 2026 · blogAccelerate agentic AI with NVIDIA Nemotron models on Amazon EKS and SageMaker HyperPod EKS
NVIDIA Nemotron 3.5 Super VL is a 120B-parameter mixture-of-experts model that pairs reasoning and tool use with multimodal understanding. This post explores its architecture and use cases, then shows how to deploy it on Amazon EKS, EKS Auto Mode, and Amazon SageMaker HyperPod EK
- Latent Space · 29 Oct 2026 · blogCan Frontier Models See Past a Decoy? Elorian Tests Visual Reasoning
A new decoy counting benchmark exposes a gap between seeing and reasoning. Elorian’s Glimpse tries to bridge it by thinking with images.
- NVIDIA Technical Blog · 29 Oct 2026 · blogUnlocking Tensor Core Performance for Machine Learning Interatomic Potentials
Atomistic simulations help researchers understand how molecules and materials behave, from chemical reactions and phase transitions to mechanical failure. These...
- Simon Willison · 29 Oct 2026 · blogQuoting Ben Hylak
every product i use should just be a headless mcp, that feeds into an agent, where i never have to think about anything underneath. except for our product, of course. in our product, customers love the ui and love us very much. we're special that way. Ben Hylak Tags: model-contex
Frequently asked questions
Which arXiv categories are covered?
cs.AI, cs.CL and cs.LG, plus Hugging Face Daily Papers.