Xinyue Li

Vision-based tactile signals in the VTOUCH dataset significantly enhance bimanual manipulation capabilities, paving the way for more effective robotic interactions.

LLMs underperform traditional ML methods in software fairness tasks, challenging the assumption that they offer a silver bullet solution for bias mitigation.

Forget retraining massive models: DPC-VQA shows you can adapt MLLMs to new video quality tasks with just a tiny calibration layer and a fraction of the data.

LLM-based multi-agent systems are riddled with hidden failure modes that traditional testing misses, but FLARE uncovers them with coverage-guided fuzzing.

VLMs can be transformed into pixel-precise structural document parsing experts, achieving state-of-the-art OCR performance by enforcing syntactic validity and structural integrity through reinforcement learning.