Tag
The article discusses how human-AI collaboration in hypothesis-driven research is advancing mathematical problem-solving and AI development, citing examples like progress on the Riemann Hypothesis and the AIRA-Compose system for architecture search.
SkillHEX proposes a closed-loop framework for autonomous skill evolution in LLM agents, using hypothesis-driven self-verification and evidence-guided tree search to overcome sparse reward challenges. It outperforms existing self-evolving methods on SkillsBench with limited interaction budgets.
PyCC.id is a Python library for hypothesis-driven equation discovery from time-series data, leveraging structural identifiability to help filter candidate models.