Dongfang Liu

Productive LLM Hallucinations: Conditions, Mechanisms, and Benefits

TokenSeek: Memory Efficient Fine Tuning via Instance-Aware Token Ditching

One-shot Learning for Robot Manipulation through Egocentric Video Demonstration

Beyond Next-Token Prediction: Diffusion vs. Autoregressive Reasoning in LLMs

Probabilistic Token Alignment for Large Language Model Fusion

Robust Ego-Exo Correspondence with Long-Term Memory

Visual Agents as Fast and Slow Thinkers

Diff-PIC: Revolutionizing Particle-In-Cell Nuclear Fusion Simulation with Diffusion Models

All You Need is One: Capsule Prompt Tuning with a Single Vector

Probabilistic Token Alignment for Large Language Model Fusion

DS-LLM: Leveraging Dynamical Systems to Enhance Both Training and Inference of Large Language Models

Re-Imagining Multimodal Instruction Tuning: A Representation View

Inertial Confinement Fusion Forecasting via Large Language Models

Hallucinating LLM Could Be Creative