The International Conference on Machine Learning (ICML) 2026 is being hosted July 6th - 11th in Seoul, South Korea. We’re excited to share all the work from SAIL that’s being presented, and you’ll find links to papers, videos and blogs below. Feel free to reach out to the contact authors directly to learn more about the work that’s happening at Stanford!

List of Accepted Papers

RoboMME: Benchmarking and Understanding Memory for Robotic Generalist Policies

Authors: Yinpei Dai, Hongze Fu, Jayjun Lee, Yuejiang Liu, Haoran Zhang, Jianing Yang, Chelsea Finn, Nima Fazeli, Joyce Chai
Contact: yuejiang.liu@stanford.edu
Award nominations: Oral
Links: Paper | Website
Keywords: robot learning, memory

Stop Automating Peer Review Without Rigorous Evaluation

Authors: Joachim Baumann, Jiaxin Pei, Sanmi Koyejo, Dirk Hovy
Contact: joabau@stanford.edu
Award nominations: Oral
Links: Paper | Blog Post
Keywords: peer review, large language models, paper laundering, artificial hivemind effect

A Unified Definition of Hallucination: It’s The World Model, Stupid!

Authors: Emmy Liu, Varun Gangal, Chelsea Zou, Michael Yu, Xiaoqi Huang, Alex Chang, Zhuofu Tao, Karan Singh, Sachin Kumar, Steven Y. Feng
Contact: syfeng@stanford.edu
Links: Paper | Blog Post | Website
Keywords: large language models, hallucination detection, hallucination mitigation, world models

AI Cartography: Mapping the Latent Landscape of AI Benchmark Ecosystems

Authors: Michael Hardy, Anka Reuel, Lijin Zhang, Jodi M. Casabianca, Sang Truong, Yash Dave, Hansol Lee, Benjamin Domingue, Sanmi Koyejo
Contact: hardym@stanford.edu
Links: Paper
Keywords: evaluation, benchmark, leaderboard, ecosystem, ai, statistical methods, statistical applications, latent variable, psychometrics,

Anchoring Self-Play for Code Repair

Authors: Caroline Choi, Zeyneb Kaya, Shirley Wu, Tengyu Ma, Tatsunori Hashimoto, Ludwig Schmidt
Contact: cchoi1@stanford.edu
Links: Paper
Keywords: self-play, code repair, synthetic data, reinforcement learning

Behavioral Mode Discovery for Fine-tuning Multimodal Generative Policies

Authors: Alberta Longhini, David Emukpere, Jean-Michel Renders, Seungsu Kim
Contact: alberta@stanford.edu
Links: Paper | Website
Keywords: rlft, generative policies

Belief Dynamics Reveal the Dual Nature of In-Context Learning and Activation Steering

Authors: Eric Bigelow*, Daniel Wurgaft*, YingQiao Wang, Noah Goodman, Tomer Ullman, Hidenori Tanaka, Ekdeep Singh Lubana
Contact: Eric Bigelow ebigelow@g.harvard.edu, Daniel Wurgaft wurgaft@stanford.edu
Links: Paper
Keywords: belief dynamics, in-context learning, activation steering, interpretability

BroRL: Scaling Reinforcement Learning via Broadened Exploration

Authors: Jian Hu, Mingjie Liu, Ximing Lu, Fang Wu, Zaid Harchaoui, Shizhe Diao, Yejin Choi, Pavlo Molchanov, Jun Yang, Jan Kautz, Yi Dong
Contact: fangwu97@stanford.edu
Links: Paper
Keywords: rlvr, llm reasoning,

Can LLMs Reason Structurally? Benchmarking via the Lens of Data Structures

Authors: Yu He, Yingxi Li, Colin White, Ellen Vitercik
Contact: heyu@stanford.edu
Links: Paper | Website
Keywords: large language models, algorithmic reasoning, benchmark

CaP-X: A Framework for Benchmarking and Improving Coding Agents for Robot Manipulation

Authors: Letian Fu, Justin Yu, Karim El-Refai, Ethan Kou, Haoru Xue, Huang Huang, Wenli Xiao, Li Fei-Fei, Guanya Shi, Jiajun Wu, S. Shankar Sastry, Yuke Zhu, Ken Goldberg, Linxi Fan
Contact: ravenh@stanford.edu
Links: Paper | Website
Keywords: gym, benchmark, agent, rl post-training, large language models

CodeClash: Benchmarking Goal-Oriented Software Engineering

Authors: John Yang, Kilian Lieret, Joyce Yang, Carlos E. Jimenez, Muhtasham Oblokulov, Aryan Siddiqui, Ofir Press, Ludwig Schmidt, Diyi Yang
Contact: johnby@stanford.edu
Links: Paper | Blog Post | Website
Keywords: language models, software engineering, benchmark

Consensus is Not Verification: Why Crowd Wisdom Strategies Fail for LLM Truthfulness

Authors: Yegor Denisov-Blanch*, Joshua Kazdan*, Jessica Chudnovsky, Rylan Schaeffer, Sheng Guan, Soji Adeshina, Sanmi Koyejo
Contact: ydebl@stanford.edu, jchud@cs.stanford.edu
Links: Paper
Keywords: inference-time scaling, wisdom of crowds, llm truthfulness, ensemble aggregation, surprisingly popular algorithm, verification

Cross-Embodiment Robot Foundation World Models with Latent Actions

Authors: Huang Huang, Sriram Yenamandra, Arjun Majumdar, Elie Aljalbout, Tushar Nagarajan, Tsung-Yen Yang, Akshara Rai, Michael Rabbat, Li Fei-Fei, Jiajun Wu, Tingfan Wu, Franziska Meier
Contact: ravenh@stanford.edu
Links: Paper
Keywords: robot world model; robot foundation model; latent action model;

Curriculum-Guided Layer Scaling for Language Model Pretraining

Authors: Karanpartap Singh, Neil Band, Ehsan Adeli
Contact: karanps@stanford.edu
Links: Paper | Website
Keywords: large language models, pretraining, curriculum learning, scaling

DexMachina: Functional Retargeting for Bimanual Dexterous Manipulation

Authors: Zhao Mandi, Yifan Hou, Dieter Fox, Yashraj Narang, Ajay Mandlekar, Shuran Song
Contact: mandi@stanford.edu
Links: Paper | Website
Keywords: dexterous manipulation; reinforcement learning; learning in simulation

Don’t Walk the Line: Boundary Guidance for Filtered Generation

Authors: Sarah Ball, Andreas Haupt
Contact: h4upt@stanford.edu
Links: Paper | Website
Keywords: safety, classifier, compound system, finetuning

Fast Byte Latent Transformer

Authors: Julie Kallini, Artidoro Pagnoni, Tomasz Limisiewicz, Gargi Ghosh, Luke Zettlemoyer, Christopher Potts, Xiaochuang Han, Srinivasan Iyer
Contact: kallini@stanford.edu
Links: Paper | Blog Post
Keywords: byte-level models, tokenizers, tokenizer-free models, large language models, byte latent transformer, blt, text diffusion, speculative decoding, inference

Formalizing Learning from Language Feedback with Provable Guarantees

Authors: Wanqiao Xu, Allen Nie, Ruijie Zheng, Aditya Modi, Adith Swaminathan, Ching-An Cheng
Contact: wanqiaoxu@stanford.edu
Links: Paper | Blog Post
Keywords: language feedback, no-regret learning, hypothesis testing, large language models

From Prior to Pro: Efficient Skill Mastery via Distribution Contractive RL Finetuning

Authors: Zhanyi Sun, Shuran Song
Contact: zhanyis@stanford.edu
Links: Paper | Website
Keywords: robotic manipulation, reinforcement learning finetuning

Gaming Consensus: Coordinated Manipulation in Crowdsourced Fact-Checking

Authors: Nikil Roashan Selvam, Jay Baxter, Sophie Hilgard, Brad Miller, Keith Coleman, Ellen Vitercik, Sanmi Koyejo
Contact: nrs@cs.stanford.edu
Links: Paper
Keywords: community notes, adversarial attacks, matrix factorization, fact-checking

Maximizing mutual information between prompt and response improves LLM performance with no additional data

Authors: Hyunji (Alex) Nam, Haoran Li, Natasha Jaques
Contact: hjnam@stanford.edu
Links: Paper
Keywords: contrastive learning for llms, llm personalization, data augmentation, learning with no human labels/verifiers

Mode Seeking meets Mean Seeking for Fast Long Video Generation

Authors: Shengqu Cai, Weili Nie, Chao Liu, Julius Berner, Lvmin Zhang, Nanye Ma, Hansheng Chen, Maneesh Agrawala, Leonidas Guibas, Gordon Wetzstein, Arash Vahdat
Contact: shengqu@stanford.edu
Links: Paper | Website
Keywords: long context, diffusion models, diffusion distillation, video generation, world model

One Bias After Another: Mechanistic Reward Shaping and Persistent Biases in Language Reward Models

Authors: Daniel Fein, Max Lamparth, Violet Xiang, Mykel Kochenderfer, Nick Haber
Contact: lamparth@stanford.edu
Links: Paper | Blog Post
Keywords: reward models, interpretability, biases, robustness, language models, rlhf

PluRel: Synthetic Data unlocks Scaling Laws for Relational Foundation Models

Authors: Vignesh Kothapalli, Rishabh Ranjan, Valter Hudovernik, Vijay Prakash Dwivedi, Johannes Hoffart, Carlos Guestrin, Jure Leskovec
Contact: vigneshk@cs.stanford.edu
Links: Paper | Video | Website
Keywords: foundation models, scaling law, synthetic data, relational data

Proteo-R1: Reasoning Foundation Models for De Novo Protein Design

Authors: Fang Wu, Weihao Xuan, Heli Qi, Hanqun Cao, Heng-Jui Chang, Zeqi Zhou, Haokai Zhao, Ma Jian, Carl Ma, Yu-Chi Cheng, Kuan Pang, Xiangru Tang, Zehong Wang, Guanlue Li, Hanchen Wang, Kejun Ying, Pan Lu, Chiho Im, Seungju Han, Peng Xia, Tinson Xu, Yinxi Li, Deyao Zhu, Pheng-Ann Heng, Naoto Yokoya, Masashi Sugiyama, Li Erran Li, Jure Leskovec, Yejin Choi
Contact: fangwu97@stanford.edu
Links: Paper | Website
Keywords: protein design, llm reasoning, multimodal, ai4science

SP-Mind: An Autonomous Reasoning Agent for Spatial Proteomics Analysis

Authors: Yucheng Yuan, Yuanfeng Ji, Zhongxiao Li, Ruijiang Li
Contact: tomyyc@stanford.edu
Links: Paper | Website
Keywords: agents, computational biology, llms, healthcare, benchmarking

The Easy, the Hard, and the Learnable: Confidence and Difficulty-Adaptive Policy Optimization for LLM Reasoning

Authors: Zhanke Zhou, Xiangyu Lu, Chentao Cao, Brando Miranda, Tongliang Liu, Bo Han, Sanmi Koyejo
Contact: zhanke@cs.stanford.edu
Links: Paper | Website
Keywords: llm reasoning, reinforcement learning, grpo, rlvr, policy optimization, post-training

Thoughtbubbles: an Unsupervised Method for Parallel Thinking in Latent Space

Authors: Houjun Liu, Shikhar Murty, Christopher Manning, Róbert Csordás
Contact: houjun@stanford.edu
Links: Paper
Keywords: neural networks, language modeling, adaptive computation, parallel computation

Towards execution-grounded automated AI research

Authors: Chenglei Si, Zitong Yang, Yejin Choi, Emmanuel Candès, Diyi Yang, Tatsunori Hashimoto
Contact: zitong@berkeley.edu
Links: Paper
Keywords: auto research, self improving ai

Transitive Representation Learning Enhances Histopathology Annotation

Authors: Moritz Schaefer
Contact: moritzs@stanford.edu
Links: Paper
Keywords: multimodal learning, contrastive learning, computational biology, computational histopathology, representation learning, zero-shot learning, spatial transcriptomics, ai for science, medical imaging, cross-modal transfer

VLAW: Iterative Co-Improvement of Vision-Language-Action Policy and World Model

Authors: Yanjiang Guo*, Tony Lee* ,Lucy Xiaoyang Shi*, Jianyu Chen, Percy Liang, Chelsea Finn
Contact: tonyhlee@stanford.edu
Links: Paper
Keywords: iterative self-improvement, world models, robotics

List of Accepted Workshop Papers

Bruno: An AI Product Manager for Scientists

Authors: Andreas Haupt, Prashaant Ranganathan
Contact: h4upt@stanford.edu
Workshop: AI4Science
Workshop Award nominations: Oral
Links: Workshop Paper | Blog Post | Website
Keywords: human coordination, context management, organizational redesign

Internal Data Repetition Destroys Language Models

Authors: Jessica Chudnovsky, Joshua Kazdan, Noam Levi, Rylan Schaeffer, Yegor Denisov-Blanch, Bo He, Mehmet Donmez, Sanmi Koyejo, David Donoho
Contact: jchud@cs.stanford.edu
Workshop: The Second Workshop on the Impact of Memorization on Trustworthy Foundation Models at ICML; Structured Probabilistic Inference & Generative Modeling; Foundations of Deep Generative Models: Understanding Memorization, Generalization, and Reasoning; Combining Theory and Benchmarks: Towards a Virtuous Cycle to Understand and Guarantee Foundation Model Performance; High-dimensional Learning Dynamics.
Workshop Award nominations: Oral Presentation at ICML 2026 Workshop on Foundations of Deep Generative Models
Links: Workshop Paper
Keywords: data repetition, compute-equivalent loss, scaling laws, pretraining, memorization

ProgramBench: Can Language Models Rebuild Programs From Scratch?

Authors: John Yang, Kilian Lieret, Jeffrey Ma, Parth Thakkar, Dmitrii Pedchenko, Sten Sootla, Emily McMilin, Pengcheng Yin, Rui Hou, Gabriel Synnaeve, Diyi Yang, Ofir Press
Contact: johnby@stanford.edu
Workshop: The 5th Deep Learning for Code Workshop
Workshop Award nominations: Oral
Links: Workshop Paper | Blog Post | Website
Keywords: language models, benchmark, software engineering

Manifold Steering Reveals the Shared Geometry of Neural Network Representation and Behavior

Authors: Daniel Wurgaft*, Can Rager*, Matthew Kowal*, Vasudev Shyam, Sheridan Feucht, Usha Bhalla, Tal Haklay, Eric Bigelow, Raphael Sarfati, Thomas McGrath, Owen Lewis, Jack Merullo, Noah D. Goodman, Thomas Fel, Atticus Geiger, Ekdeep Singh Lubana
Contact: wurgaft@stanford.edu
Workshop: Mechanistic Interpretability Workshop
Workshop Award nominations: Spotlight
Links: Workshop Paper | Blog Post
Keywords: representation geometry, activation steering, interpretability

TRACE: Capability-Targeted Agentic Training

Authors: Hangoo Kang, Tarun Suresh, Jon Saad-Falcon, Azalia Mirhoseini
Contact: hangook@stanford.edu, tsuresh@stanford.edu
Workshop: Second Workshop on Agents in the Wild: Safety, Security, and Beyond
Workshop Award nominations: Spotlight
Links: Workshop Paper | Blog Post | Website
Keywords: llm agents, synthetic environment generation, agent reinforcement learning, failure-driven training

SWE-chat: Coding Agent Interactions From Real Users in the Wild

Authors: Joachim Baumann, Vishakh Padmakumar, Xiang Li, John Yang, Diyi Yang, Sanmi Koyejo
Contact: joabau@stanford.edu
Workshop: Deep Learning for Code: Towards Human-Centered Coding Agents
Links: Workshop Paper | Blog Post | Website
Keywords: in-the-wild data, vibe coding, coding agents, human-ai interaction

AI Coding Benchmarks Need Proofs, Not Just Tests

Authors: Daneshvar Amrollahi, Mahyar Karimi, Brando Miranda, Leni Aniva, Chuyue Sun, Clark Barrett, Sanmi Koyejo
Contact: daneshvar@cs.stanford.edu
Workshop: The 5th Deep Learning for Code (DL4C) Workshop
Links: Workshop Paper
Keywords: code generation, lean, verification, evaluation and benchmarks

Calibrating Generative Models to Feature Distributions with MMD Finetuning

Authors: Nathaniel L. Diamant, Brian L. Trippe
Contact: diamant@stanford.edu
Workshop: SPIGM
Links: Workshop Paper | Website
Keywords: generative modeling, finetuning, ai for science

CertJudge: Evaluating Lean Formal-Code With Falsifiable Properties

Authors: Ethan S Hersch, Brando Miranda, Elyas Obbad, Srivatsava Daruru, Zhanke Zhou, Kirill Acharya, Sanmi Koyejo
Contact: ehersch@stanford.edu
Workshop: ICML 2026 AI4Math Workshop
Links: Workshop Paper
Keywords: llm-as-a-judge, code evaluation, human-centered coding agents, formal methods, benchmarking and evaluation

ContinuityBench: A Framework and Taxonomy for Evaluating Agent Recovery from Interrupted State

Authors: Aryan Gulati
Contact: aryangul@cs.stanford.edu
Workshop: The Combining Theory and Benchmarks (CTB) Workshop at the 43 rd International Conference on Machine Learning
Links: Workshop Paper
Keywords: agent benchmarks, interrupted-state recovery, task resumption, checkpoint sufficiency, handoff quality, agent reliability, benchmark construction, multi-domain evaluation, partial-state reasoning, recovery robustness

Distill to Detect: Exposing Stealth Biases in LLMs through Cartridge Distillation

Authors: Shayan Talaei, Abhinav Chinta, Devvrit Khatri, Amin Karbasi, Azalia Mirhoseini, Amin Saberi
Contact: achinta@stanford.edu
Workshop: TAIGR, AI4GOOD, Mechanistic Interpretability, and CoLoRAI.
Links: Workshop Paper | Website
Keywords: language model safety, bias detection, distillation, model auditing

Mixture of Graders: Adaptive Strategy Routing for LLM-Based Mathematical Evaluation

Authors: Eric Chen, Aryan Gulati, Brando Miranda, Zeyu Tang, Sanmi Koyejo
Contact: ericc27@cs.stanford.edu
Workshop: The 3rd AI for Math (AI4Math) Workshop at the 43rd International Conference on Machine Learning
Links: Workshop Paper
Keywords: llm judge, llm evaluations, chain-of-thought, math reasoning evaluation, prompting, grading, judge reliability, score stability, routing, mixture of models

Neural Algorithmic Reasoning Must Explain When Neuralization Adds Value

Authors: Yu He, Robert R. Nerem, Timo Stoll, Semih Cantürk, Dobrik Georgiev, Chendi Qian, Solveig Wittig, Floris Geerts, Stefanie Jegelka, Ellen Vitercik, Yusu Wang, Nikolaos Karalias, Christopher Morris
Contact: heyu@stanford.edu
Workshop: AI4Math
Links: Workshop Paper
Keywords: neural algorithmic reasoning

Proximal State Nudging: Reducing Skill Atrophy from AI Assistance

Authors: Megha Srivastava, Jonathan Ouyang, Eric Zhou, Andrew Silva, Emily Sumner, Dorsa Sadigh, Yuchen Cui, Deepak Gopinath, Guy Rosman
Contact: megha@cs.stanford.edu
Workshop: Trustworthy AI4GOOD
Links: Workshop Paper
Keywords: robotics, human-ai interaction, deskilling, control, autonomous driving

Reward Bias Substitution: Single-Axis Bias Mitigations Redirect Optimization Pressure

Authors: Max Lamparth, Daniel Fein, Andreas Haupt, Marcel Hussing, Mykel Kochenderfer
Contact: lamparth@stanford.edu
Workshop: Second Workshop on Agents in the Wild: Safety, Security, and Beyond (ICML 2016 AIWILD)
Links: Workshop Paper | Blog Post
Keywords: reward hacking, rlhf, preference learning, evaluations, robustness, rl, nlp, theory

Scale Dependent Data Duplication

Authors: Joshua Kazdan*, Noam Levi*, Rylan Schaeffer, Jessica Chudnovsky, Abhay Puri, Bo He, Mehmet Donmez, Sanmi Koyejo, David Donoho
Contact: jkazdan@stanford.edu
Workshop: The Impact of Memorization on Trustworthy Foundation Models (MemFM) Structured Probabilistic Inference & Generative Modeling (SPIGM) Foundations of Deep Generative Models: Understanding Memorization, Generalization, and Reasoning (FoGen) Combining Theory and Benchmarks: Towards a Virtuous Cycle to Understand and Guarantee Foundation Model Performance (CTB)
Links: Workshop Paper
Keywords: semantic duplicates, scaling laws, data deduplication, pretraining data diversity, semantic collisions

Steerable Cultural Preference Optimization of Reward Models

Authors: Minsik Oh, Advit Deepak, Sophie Wu, Douwe Kiela, Ekaterina Shutova
Contact: minsik@stanford.edu
Workshop: Pluralistic Alignment @ ICML 2026
Links: Workshop Paper
Keywords: pluralistic alignment, reward models

Subliminal Learning is Steering Vector Distillation

Authors: Camila Blank, Agam Bhatia
Contact: agam2026@stanford.edu
Workshop: Mechanistic Interpretability Workshop
Links: Workshop Paper
Keywords: interpretability for ai safety

We look forward to seeing you at ICML 2026!