Osbert Bastani

Knowledgeable Language Models as Black-Box Optimizers for Personalized Medicine

A Fast, Reliable, and Secure Programming Language for LLM Agents with Code Actions

Scaling Multi-Task Bayesian Optimization with Large Language Models

Divide and Abstract: Autoformalization via Decomposition and Abstraction Learning

Autonomous Functional Play with Correspondence-Driven Trajectory Warping

Sparling: End-to-End Spatial Concept Learning via Extremely Sparse Activations

BrowserArena: Evaluating LLM Agents on Real-World Web Navigation Tasks

LLM Program Optimization via Retrieval Augmented Search

SeekerGym: Benchmarking Agentic Information Seeking under Uncertainty

RAPID: An Efficient Reinforcement Learning Algorithm for Small Language Models

Vision Language Models are In-Context Value Learners

Alignment of Large Language Models with Constrained Learning

Evaluating the Diversity and Quality of LLM Generated Content

Zeroth-Order Fine-Tuning of LLMs with Transferable Static Sparsity

Conformal Structured Prediction

Diversity By Design: Leveraging Distribution Matching for Offline Model-Based Optimization

Stochastic Online Conformal Prediction with Semi-Bandit Feedback

Stochastic Online Conformal Prediction with Semi-Bandit Feedback

Sparling: Learning Latent Representations with Extremely Sparse Activations

Does Instruction Tuning Reduce Diversity? A Case Study Using Code Generation