Ninghao Liu

Recursive task synthesis not only slashes generation costs to $0.05 per task but also produces increasingly complex challenges that boost model performance by up to 10 points on key benchmarks.

Agents struggle with long-horizon tasks, achieving only a 15.2% success rate even with advanced models, highlighting a critical gap in current AI capabilities.

OPERA's intrinsic reward mechanism enables LLMs to achieve new heights in open-ended reasoning, outperforming proprietary models without the pitfalls of biased supervision.

SAERec achieves a breakthrough in intent-based recommendations by automatically constructing a nuanced intent space that significantly enhances both accuracy and interpretability.

Combining soft correctness-aware gating with teacher-probability scaling leads to a substantial boost in GUI grounding performance, revealing the critical importance of signal quality in self-distillation.

TRON enables an endless supply of tailored training instances, revolutionizing how we approach reinforcement learning for visual reasoning.