2 citations · 4 across the 6 of their papers we have counts for
5 papers · 1 filter
Trust the Critic More
Kaiyue Wen, Luke Bailey, Arvind Mahankali +1
Standard language model RL algorithms credit every token of a long rollout with the same advantage determined by the terminal reward. Actor-critic methods can provide finer-grained…
Divide-and-Conquer CoT: RL for Reducing Latency via Parallel Reasoning
Arvind Mahankali, Kaiyue Wen, Tengyu Ma
Long chain-of-thought reasoning (Long CoT) is now fundamental to state-of-the-art LLMs, especially in mathematical reasoning. However, LLM generation is highly sequential, and long…
Formal Theorem Proving by Rewarding LLMs to Decompose Proofs Hierarchically
Kefan Dong, Arvind Mahankali, Tengyu Ma
Mathematical theorem proving is an important testbed for large language models' deep and abstract reasoning capability. This paper focuses on improving LLMs' ability to write proof…
One Step of Gradient Descent is Provably the Optimal In-Context Learner with One Layer of Linear Self-Attention
Arvind Mahankali, Tatsunori B. Hashimoto, Tengyu Ma
Recent works have empirically analyzed in-context learning and shown that transformers trained on synthetic linear regression tasks can learn to implement ridge regression, which i…
Beyond NTK with Vanilla Gradient Descent: A Mean-Field Analysis of Neural Networks with Polynomial Width, Samples, and Time
Arvind Mahankali, Jeff Z. Haochen, Kefan Dong +2
Despite recent theoretical progress on the non-convex optimization of two-layer neural networks, it is still an open question whether gradient descent on neural networks without un…