Archiki Prasad

One Life to Learn: Inferring Symbolic World Models for Stochastic Environments from Unguided Exploration

Executable Functional Abstractions: Inferring Generative Programs for Advanced Math Problems

Think Right: Learning to Mitigate Under-Over Thinking via Adaptive, Attentive Compression

LASeR: Learning to Adaptively Select Reward Models with Multi-Arm Bandits

Retrieval-Augmented Generation with Conflicting Evidence

Self-Consistency Preference Optimization

Learning to Generate Unit Tests for Automated Debugging

LASeR: Learning to Adaptively Select Reward Models with Multi-Armed Bandits

MAgICoRe: Multi-Agent, Iterative, Coarse-to-Fine Refinement for Reasoning

System 1.x: Learning to Balance Fast and Slow Planning with Language Models