Tagged articles

Harness Evolution

3 articles · Page 1 of 1

Oct 29, 2026 · Artificial Intelligence

CMU's Hermit: Unified Framework for Fair Harness Evolution Comparison

CMU researchers open-source Hermit, a unified training framework that enables fair comparison of recursive self-improvement methods by externalizing learning via pluggable Harness Adapters, supporting algorithms like GEPA and SkillOpt, and providing a tutorial for continuous agent improvement in tools like Codex and Claude Code.

Agent Continuous LearningCMU ResearchCodex Integration

0 likes · 8 min read

CMU's Hermit: Unified Framework for Fair Harness Evolution Comparison

Machine Heart

Sep 17, 2026 · Artificial Intelligence

Beyond Prompts: Harness-Policy Co-Evolution for Agent Safety by Shanghai AI Lab

Shanghai AI Lab and university collaborators propose SHE and SafeEvolve, two frameworks that evolve agent safety by learning from execution trajectories: SHE updates a modular safety harness via trajectory-driven evolution, while SafeEvolve distills verified harness experience into the policy model through SFT and RL, reducing attack success rates on benchmarks.

Agent SafetyHarness EvolutionLLM agents

0 likes · 12 min read

Beyond Prompts: Harness-Policy Co-Evolution for Agent Safety by Shanghai AI Lab

Machine Heart

Sep 16, 2026 · Artificial Intelligence

Harness Evolution vs. Test-Time Scaling: Simple Retries Outperform Complex Self-Improvement

A study from AI2 and University of Washington finds that complex Harness Evolution for AI agents fails to consistently outperform simple test-time scaling methods like parallel sampling under equal compute budgets, and improvements rarely transfer to unseen tasks, questioning whether observed gains stem from genuine self-improvement or just extra attempts.

AI agentsAgent EvaluationBenchmarking

0 likes · 14 min read

Harness Evolution vs. Test-Time Scaling: Simple Retries Outperform Complex Self-Improvement