Eric Hanchen Jiang

Reasoning via Test-Time Instance-Level Policy Gradient in Latent Space

Beyond Magic Words: Sharpness-Aware Prompt Evolving for Robust Large Language Models with TARE

Learning to Rank Chain-of-Thought: Using a Small Model

GTD: Dynamic Generation of Multi LLM Agents Communication Topologies with Graph Diffusion Models

Two Lenses are Better Than One: Dual Vector Quantization for Self-Supervised Graph Learning

Distortion Regularization for Distance-Preserving Graph Embedding

Diagnose, Localize, Align: A Full-Stack Framework for Reliable LLM Multi-Agent Systems under Instruction Conflicts

Energy-Driven Steering: Reducing False Refusals in Large Language Models

Conditional Trajectories in Diffusion Models - Modeling Galaxy Evolution from Redshift