AI Briefing#autoresearch
The latest AI and developer news about #autoresearch, with the original source and a short summary.
The latest AI and developer news about #autoresearch, with the original source and a short summary.
Jump Trading leverages GPT-6 Astra to enable agents to perform multi-day, recursive quantitative analysis with minimal human intervention.
darwin-skill 2.0 iteratively improves agent skills through evaluation, verification, and rollback.
The AIDE² system surpassed a model hand-tuned over 2 years through 8 days of recursive self-improvement.
It explains the 'outer loop' structure through which AI agents improve their own performance using feedback signals and human input.
LLMs proved that hyperparameter optimization performance is maximized when combined with traditional algorithms rather than used alone.
When an agent read papers and forks before writing code, llama.cpp optimizations improved significantly.