3 papers
cs.AI2025
Synthetic Error Injection Fails to Elicit Self-Correction In Language Models
David X. Wu, Shreyas Kapur, Anant Sahai +1
Reinforcement learning has become the dominant paradigm for eliciting reasoning and self-correction capabilities in large language models, but its computational expense motivates e…
cs.LG2024
Evidence of Learned Look-Ahead in a Chess-Playing Neural Network
Erik Jenner, Shreyas Kapur, Vasil Georgiev +3
Do neural networks learn to implement algorithms such as look-ahead or search "in the wild"? Or do they rely purely on collections of simple heuristics? We present evidence of lear…
cs.AI2024
Diffusion On Syntax Trees For Program Synthesis
Shreyas Kapur, Erik Jenner, Stuart Russell
Large language models generate code one token at a time. Their autoregressive generation process lacks the feedback of observing the program's output. Training LLMs to suggest edit…