Jiantao Jiao

Emergence of Superposition: Unveiling the Training Dynamics of Chain of Continuous Thought

Sample Complexity and Representation Ability of Test-time Scaling Paradigms

GSM-Agent: Understanding Agentic Reasoning Using Controllable Environments

How Do Transformers Perform Two-Hop Reasoning in Context?

EmbedLLM: Learning Compact Representations of Large Language Models

Reasoning by Superposition: A Theoretical Perspective on Chain of Continuous Thought

Information-Driven Design of Imaging Systems

Generalization or Hallucination? Understanding Out-of-Context Reasoning in Transformers

Token Assorted: Mixing Latent and Text Tokens for Improved Language Model Reasoning

How to Evaluate Reward Models for RLHF

Watermarking using Semantic-aware Speculative Sampling: from Theory to Practice

Active-Dormant Attention Heads: Mechanistically Demystifying Extreme-Token Phenomena in LLMs

Thinking LLMs: General Instruction Following with Thought Generation

Meta-Rewarding Language Models: Self-Improving Alignment with LLM-as-a-Meta-Judge