Zeming Liu

SEFS achieves superior artistic stylization by leveraging low-resolution image crops, enhancing content consistency while avoiding unwanted style transfer artifacts.

Existing smart home assistants miss user intentions in 60% of cases when interpreting elliptical commands, revealing a critical gap in their design.

Self-conditioning on verified trajectories boosts reinforcement learning performance by over 8%, revealing the power of internal feedback in credit assignment.

LLM agents can get 18% better at tasks by co-evolving their skills and tools, instead of learning them separately.

Forget monolithic policies – splitting your LLM's RL policy into accuracy-focused and exploration-driven modes unlocks better performance and diversity.

Open-source 7B LLMs can now rival GPT-4o performance on validation tasks, thanks to a novel reinforcement learning approach that leverages calibrated self-evaluation as a dense reward signal.