Let Skills Evolve Collectively with Agentic Evolver
The long-horizon computer-use harness. Run AI agents across desktop apps and the CLI for extended periods while preserving task state and making reliable progress on complex workflows. Features fresh-context execution, durable verified state, independent auditing, recoverable progress, and native Claude Code / Codex / OpenClaw integration.
[ICLR 2026] Tree Search for LLM Agent Reinforcement Learning
[CVPR 2026] Beyond Generation: Advancing Image Editing Priors for Depth and Normal Estimation
[ACL 2026 Findings] Thinking with Map: Reinforced Parallel Map-Augmented Agent for Geolocalization
Benchmarking models as runtime Controllers for Loop Engineering