Modified copy of irom-princeton/dppo (Diffusion Policy Policy Optimization) — swaps the PPO advantage for a GRPO-form group-relative advantage across the robomimic tasks. Research fork, not a maintained release.
原汁原昧 Claude Code 可运行,可构建, 可调试版; Typescript 类型全修复; 企业级可靠性; 安全无毒, lock 文件保真, 可直接 bun i; bun run dev 启动
Research fork for AI agent orchestration dashboards, task tracking, cost monitoring and workflow visibility.
Research fork: daily AI ecosystem digest from GitHub, arXiv, HN, HuggingFace, Product Hunt and other sources.
Research fork of BSK-RL extending spacecraft tasking environments toward satellite-terrestrial integrated network simulations.
Research fork for studying OpenClaw agent organizations across Codex, Claude Code, Cursor and other coding agents.
Research fork of OpenClaw for studying local AI assistants, tools, skills, agent teams, memory and automation use cases.
Research fork for cross-tool AI skill sharing across Codex, Claude Code, OpenClaw and other agent runtimes.
To associate your repository with the research-fork topic, visit your repo's landing page and select "manage topics."