Chuan Li

Latent Particle World Models: Self-supervised Object-centric Stochastic Dynamics Modeling

To Compress or Not? Pushing the Frontier of Lossless GenAI Model Weights Compression with Exponent Concentration

Measuring and Mitigating Rapport Bias of Large Language Models under Multi-Agent Social Interactions

Randomized Antipodal Search Done Right for Data Pareto Improvement of LLM Unlearning

TangoFlux: Super Fast and Faithful Text to Audio Generation with Flow Matching and Clap-Ranked Preference Optimization

OffTopicEval: When Large Language Models Enter the Wrong Chat, Almost Always!

InstructPro: Natural Language Guided Ligand-Binding Protein Design

Bifrost-1: Bridging Multimodal LLMs and Diffusion Models with Patch-level CLIP Latents

BLEUBERI: BLEU is a surprisingly effective reward for instruction following

Latent Thought Models with Variational Bayes Inference-Time Computation