The Stanford SALT & Talk Speaker Series takes place on the second Friday of every month, from 12:00–1:30pm PT. This interdisciplinary series brings together experts from a wide range of fields, including natural language processing, machine learning, AI, human-computer interaction, linguistics, and the social sciences, to explore cutting-edge research in socially aware NLP.
We would like to thank Amplify Partners for their support of the talk series from Summer 2026 – Spring 2027.
The series is organized by Joachim Baumann and Vishakh Padmakumar. Attendance is invite only. If you would like to attend, please reach out to the organizers.
| Date | Speaker | Talk Title |
|---|---|---|
| Aug 14, 2026 |
Eve Fleisig
|
Leaderboard Hacking: Preference-Based Model Evaluations are Vulnerable to Manipulation |
| Sep 11, 2026 | TBD | TBD |
| Oct 9, 2026 | TBD | TBD |
| Nov 13, 2026 | TBD | TBD |
| Dec 11, 2026 | TBD | TBD |
| Jan 8, 2027 | TBD | TBD |
| Feb 12, 2027 | TBD | TBD |
| Mar 12, 2027 | TBD | TBD |
| Apr 9, 2027 | TBD | TBD |
| May 14, 2027 | TBD | TBD |
Past Speakers
2025–2026
| Date | Speaker | Talk Title |
|---|---|---|
| July 30, 2026 | Jeremy Z Yang |
How AI Agents Reshape Knowledge Work: Autonomy, Efficiency, and Scope |
| July 17, 2026 | Ricardo Olmedo |
Scaling, specialization, and the $60 SWE-bench model |
| July 10, 2026 | Tiancheng Hu |
From Individuals to Groups to Interactions: What Is Missing in LLM Simulations? |
| June 12, 2026 | Yoonho Lee |
Meta-harness: End-to-end optimization of model harnesses |
| June 5, 2026 | Yuetai Li |
JobBench: Aligning Agent Work With Human Will |
| June 3, 2026 | Alec Radford |
- |
| May 29, 2026 | Jonas Hübotter |
Rethinking Post-Training with Self-Distillation |
| Apr 30, 2026 | Shannon Shen |
From Assistants to Collaborators: Building Agents for Long-Form, Open-ended Collaboration |
| Apr 10, 2026 | Bodhisattwa Majumdar |
Intriguing Speculations with AI on the Possibilities of Science |
| Jan 23, 2026 | Yu Sun |
Test-Time Training |
| Dec 11, 2025 | Ben Shi |
Towards Scalable Evaluation of Human-AI Interaction |