Reading group
The Snorkel AI Reading Group (SARG) is a recurring forum for researchers and practitioners to explore the latest AI research papers and frontier developments while building meaningful connections within the community.
We’ll dive into the most talked-about research in benchmarking and evaluation, pressure-test the ideas, share our POV, and bring in the authors themselves for open discussion. Every talk gets a written recap, linked below.
Paper reading //
November 18, 2026, San Francisco, CA
Reimagining Research Papers As Interactive and Reliable AI Agents
Jiacheng Miao
Postdoctoral Fellow, Stanford
↳ Register now
December 02, 2026
Steering “AI Scientists” Toward Valid & Trustworthy Research
with Harit Vishwakarma, Postdoc, University of Oxford
↳ Register now
Past events
Past
How Efficiently does AI Turn Energy into Useful Work?
with Jon Saad-Falcon, Ph.D. and MBA student
with Avanika Narayan, Ph.D. student
Past
Measuring Physicians and AI Relevance Alignment with MedPAIR
with Yuexing Hao, Postdoctoral Researcher, MIT
↳ MedPAIR recap
Past
OS World 2.0 (Virtual)
with Mengqi Yuan, PhD Student,
Hong Kong University (XLangNLP)
↳ OSWorld 2.0 recap
Past
Continual Learning Bench
with Parth Asawa, PhD Student, UC Berkeley
↳ Continual Learning Bench recap
Past
Train-to-Test (T²) Scaling Laws
with Nicholas Roberts, Postdoctoral Research Fellow,
UW-Madison
↳ T² scaling laws recap
Past
Senior SWE-Bench
with Henry Ehrenberg, Lead Researcher, Senior SWE-Bench
↳ Senior SWE-Bench recap
Past
Agents’ Last Exam
with Yiyou Sun, Postdoctoral Researcher, UC Berkeley
with Xinyang Han, PhD Student, UC Berkeley
↳ Agents’ Last Exam recap
Past
JUDGMENTBENCH: Comparing Rubric and Preference Evaluation for Quality Assessment
with Russell Yang, AI Engineering Fellow, Stanford Law School
↳ JudgmentBench recap
Past
Collaborative Gym: A Framework for Enabling and Evaluating Human-Agent Collaboration
with Yijia Shao, PhD candidate, Stanford NLP
↳ Collaborative Gym recap
Past
Improving LLM Agents with Code World Models & AutoHarness
with Carter Wendelken, Google Deepmind
↳ Code World Models & AutoHarness recap
Past
Olmix: A Framework for Data Mixing Throughout LM Development
with Mayee Chen, PhD Candidate, Stanford AI Lab; Founding Scientist, Stealth Startup
↳ Olmix recap