Uploaded May 2025 | Updated September 2026, 2 weeks ago
Allen School Colloquia Series
Title: Controlling Language Models
Speaker: Lisa Li (Stanford)
Date: March 4, 2025
Abstract: Controlling language models is key to unlocking their full potential and making them useful for downstream tasks. Successfully deploying these models often requires both task-specific customization and rigorous auditing of their behavior. In this talk, I will begin by introducing a customization method called Prefix-Tuning, which adapts language models by updating only 0.1% of their parameters. Next, I will address the need for robust auditing by presenting a Frank-Wolfe-inspired algorithm for red-teaming language models, which provides a principled framework for discovering diverse failure modes. Finally, I will rethink the root cause of these control challenges, and propose a new generative model for text, called Diffusion-LM, which is controllable by design.
Bio: Lisa Li is a PhD candidate at Stanford University, where she is advised by Percy Liang and Tatsunori Hashimoto. Her research focuses on developing methods to make language models more capable and controllable. Lisa is supported by the Two Sigma PhD fellowship and Stanford Graduate Fellowship and is the recipient of an EMNLP Best Paper award.
This video is closed captioned
Allen School Colloquia Series
Title: Controlling Language Models
Speaker: Lisa Li (Stanford)
Date: March 4, 2025
Abstract: Controlling language models is key to unlocking their full potential and making them useful for downstream tasks. Successfully deploying these models often requires both task-specific customization and rigorous auditing of their behavior. In this talk, I will begin by introducing a customization method called Prefix-Tuning, which adapts language models by updating only 0.1% of their parameters. Next, I will address the need for robust auditing by presenting a Frank-Wolfe-inspired algorithm for red-teaming language models, which provides a principled framework for discovering diverse failure modes. Finally, I will rethink the root cause of these control challenges, and propose a new generative model for text, called Diffusion-LM, which is controllable by design.
Bio: Lisa Li is a PhD candidate at Stanford University, where she is advised by Percy Liang and Tatsunori Hashimoto. Her research focuses on developing methods to make language models more capable and controllable. Lisa is supported by the Two Sigma PhD fellowship and Stanford Graduate Fellowship and is the recipient of an EMNLP Best Paper award.
This video is closed captioned
![[Audio Descriptions] Faculty In Focus: Nirvan Tyagi](https://i.ytimg.com/vi/v8NzVoue-Ac/mqdefault.jpg "[Audio Descriptions] Faculty In Focus: Nirvan Tyagi
In this episode of the Allen School’s “Faculty in Focus” series, Assistant Professor Nirvan Tyagi discusses his work on cryptography techniques such as zero knowledge proofs, which would enable systems used in banking and other services to provide strong privacy for users while protecting them against bad actors. His research aims to give technology users more control over their data while maintaining efficiency and performance.
For a version without audio descriptions, visit https://youtu.be/PMaUVhjEJlA.")
