Blog / 博客Blog

知识地图 / 推荐阅读顺序Knowledge Map / Recommended Reading Order

1. Agent Research Environments：先问任务对什么工件、何种反馈和 verifier 负责，再把逐步拦截、提交、证据、reward 和最终接纳拆开。Agent research environments: first ask what artifact, feedback, and verifier a task holds an agent accountable for, then separate step interception, commitment, evidence, reward, and final acceptance.
2. ClawBench / 网页 Agent：把 self-host 网站、Browser / Computer Use、受控提交拦截、证据轨与独立 verifier 合成一个可训练、可反驳的 RL 设计。ClawBench / web agents: combine self-hosted websites, Browser / Computer Use, controlled submission interception, evidence trails, and independent verifiers into a trainable, falsifiable RL design.
3. 长程 Agent：把 long horizon 定义为可恢复的状态机——什么必须 checkpoint，怎样记录 active/wait/recovery 时间，如何在故障和安全退化下仍验证最终完成。Long-horizon agents: define long horizon as a recoverable state machine—what must be checkpointed, how to record active/wait/recovery time, and how completion stays verified under faults and safety degradation.
4. AutoResearch：看开发搜索何时只是本地优化，以及候选冻结、独立接纳与复验怎样把结果变成可反驳的研究证据。AutoResearch: learn when development search is only local optimization, and how candidate freezing, independent acceptance, and reruns create falsifiable research evidence.
5. RSI：用改进算子、选择、迁移与完整性四条曲线，检验一条 lineage 是否真的从受限优化升级到“改进改进者”。RSI: use improvement-operator, selection, transfer, and integrity curves to test whether a lineage really moves from bounded optimization to “improving the improver.”
6. Harness 架构：真实 coding agents 怎样组织 loop、tools、skills、budget、sandbox 与训练轨迹。Harness architecture: how real coding agents organize loops, tools, skills, budgets, sandboxes, and training traces.
7. LLM RL 基础：PPO / DPO / GRPO、reward、critic 与 verifiable reward。LLM RL foundations: PPO / DPO / GRPO, reward, critic, and verifiable rewards.