Ge Liu

Consolidating specialist LLMs into a single, more accurate generalist is possible with *zero* gradient updates, relying solely on forward passes and input statistics.

Chain-of-thought token bloat can be slashed by 27% while actually boosting accuracy simply by recognizing that reasoning capacity lives almost entirely within the null space of a base model's dominant singular vectors.

Training voice agents in an audio-native environment can more than double their task success rates and enhance efficiency without external dependencies.

The Qwen3-Omni model can reconstruct complex narratives from garbled audio inputs, revealing a hidden layer of reasoning that challenges our understanding of audio language processing.

LLMs can achieve a "free lunch" in reasoning tasks, simultaneously improving accuracy and reducing token usage by simply training them to solve multiple problems in parallel.