5 papers
Proximal State Nudging: Reducing Skill Atrophy from AI Assistance
Megha Srivastava, Jonathan Ouyang, Eric Zhou +6
Skill atrophy, the gradual decline of human capability under AI assistance, poses a safety risk in shared-control of semi-autonomous systems, where operators may be unable to disti…
Modeling Student Learning with 3.8 Million Program Traces
Alexis Ross, Megha Srivastava, Jeremiah Blanchard +1
As programmers write code, they often edit and retry multiple times, creating rich "interaction traces" that reveal how they approach coding tasks and provide clues about their lev…
Policy Learning with a Language Bottleneck
Megha Srivastava, Cedric Colas, Dorsa Sadigh +1
Modern AI systems such as self-driving cars and game-playing agents achieve superhuman performance, but often lack human-like generalization, interpretability, and inter-operabilit…
Shared Autonomy for Proximal Teaching
Megha Srivastava, Reihaneh Iranmanesh, Yuchen Cui +6
Motor skill learning often requires experienced professionals who can provide personalized instruction. Unfortunately, the availability of high-quality training can be limited for…
Optimistic Verifiable Training by Controlling Hardware Nondeterminism
Megha Srivastava, Simran Arora, Dan Boneh
The increasing compute demands of AI systems have led to the emergence of services that train models on behalf of clients lacking necessary resources. However, ensuring correctness…