Kenneth Li

17 papers A* 6Journal 11
YearRankTypeTitle / Venue / Authors
2025 A* conf
Communicating Activations Between Language Model Agents.
ICML
Vignav Ramesh, Kenneth Li
2025 J jnl
Communicating Activations Between Language Model Agents.
CoRR
Vignav Ramesh, Kenneth Li
2025 J jnl
What Does it Mean for a Neural Network to Learn a "World Model"?
CoRR
Kenneth Li, Fernanda B. Viégas, Martin Wattenberg
2025 A* conf
When Bad Data Leads to Good Models.
ICML
Kenneth Li, Yida Chen, Fernanda B. Viégas, Martin Wattenberg
2025 J jnl
When Bad Data Leads to Good Models.
CoRR
Kenneth Li, Yida Chen, Fernanda B. Viégas, Martin Wattenberg
2024 A* conf
An AI-Resilient Text Rendering Technique for Reading and Skimming Documents.
CHI
Ziwei Gu, Ian Arawjo, Kenneth Li, Jonathan K. Kummerfeld, Elena L. Glassman
2024 J jnl
An AI-Resilient Text Rendering Technique for Reading and Skimming Documents.
CoRR
Ziwei Gu, Ian Arawjo, Kenneth Li, Jonathan K. Kummerfeld, Elena L. Glassman
2024 J jnl
Designing a Dashboard for Transparency and Control of Conversational AI.
CoRR
Yida Chen, Aoyu Wu, Trevor DePodesta, Catherine Yeh, Kenneth Li, Nicholas Castillo Marin, Oam Patel, Jan Riecke, Shivam Raval, Olivia Seow, Martin Wattenberg, Fernanda B. Viégas
2024 J jnl
Dialogue Action Tokens: Steering Language Models in Goal-Directed Dialogue with a Multi-Turn Planner.
CoRR
Kenneth Li, Yiming Wang, Fernanda B. Viégas, Martin Wattenberg
2024 J jnl
Measuring and Controlling Persona Drift in Language Model Dialogs.
CoRR
Kenneth Li, Tianle Liu, Naomi Bashkansky, David Bau, Fernanda B. Viégas, Hanspeter Pfister, Martin Wattenberg
2024 A* conf
Q-Probe: A Lightweight Approach to Reward Maximization for Language Models.
ICML
Kenneth Li, Samy Jelassi, Hugh Zhang, Sham M. Kakade, Martin Wattenberg, David Brandfonbrener
2024 J jnl
Q-Probe: A Lightweight Approach to Reward Maximization for Language Models.
CoRR
Kenneth Li, Samy Jelassi, Hugh Zhang, Sham M. Kakade, Martin Wattenberg, David Brandfonbrener
2023 A* conf
Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Task.
ICLR
Kenneth Li, Aspen K. Hopkins, David Bau, Fernanda B. Viégas, Hanspeter Pfister, Martin Wattenberg
2023 A* conf
Inference-Time Intervention: Eliciting Truthful Answers from a Language Model.
NeurIPS
Kenneth Li, Oam Patel, Fernanda B. Viégas, Hanspeter Pfister, Martin Wattenberg
2023 J jnl
Inference-Time Intervention: Eliciting Truthful Answers from a Language Model.
CoRR
Kenneth Li, Oam Patel, Fernanda B. Viégas, Hanspeter Pfister, Martin Wattenberg
2022 J jnl
Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Task.
CoRR
Kenneth Li, Aspen K. Hopkins, David Bau, Fernanda B. Viégas, Hanspeter Pfister, Martin Wattenberg
2021 J jnl
Towards Tokenized Human Dynamics Representation.
CoRR
Kenneth Li, Xiao Sun, Zhirong Wu, Fangyun Wei, Stephen Lin