@rohanpaul_ai: Self-improving agents have always needed a human to write the grader. A new AWS + HSBC paper removes that step: the gra…
Summary
AWS and HSBC researchers present Double Ratchet, a method where a grader evolves alongside a self-improving agent from just 10 labeled examples, matching ground-truth supervision across code, SQL, and report generation tasks.
View Cached Full Text
Cached at: 10/12/26, 01:31 PM
Self-improving agents have always needed a human to write the grader.
A new AWS + HSBC paper removes that step: the grader evolves alongside the agent from 10 labeled examples and trains it about as well as ground truth.
Double Ratchet evolves the grader alongside the agent’s skills, building it from small, mostly rule-based checks for 1 failure type each. A grader version is chosen for agreeing with 10 labeled examples, never for raising the agent’s score.
It kept 88% to 110% of the gain that ground truth or a rubric gave the same skill loop, across code, SQL, and reports. With the safeguards removed, the grader passed 97% to 100% of outputs yet trained skills as well.
Similar Articles
@rohanpaul_ai: New paper from Cambridge Univ+NVIDIA and other top labs teaches AI agents and AI judges to improve together, so neither…
A new paper from Cambridge, NVIDIA, and other labs introduces the Red Queen Gödel Machine, a method where AI agents and their evaluators co-evolve to prevent stagnation. The approach avoids fixed benchmarks by allowing judges to improve at safe handoff points, leading to better performance in coding and paper writing tasks.
@rohanpaul_ai: Great paper on Self-evolving agents. Enterprise agents cannot truly improve until their messy daily work becomes safe l…
A paper proposing a mechanism for enterprise agents to improve by safely converting messy daily work into learning data, using a data proxy and control layer, with AREAL2.0 demonstrating online RL from real interaction traces.
@rohanpaul_ai: Brilliant new paper from Meta, CMU and other labs. Shows that coding agents improve faster by manufacturing their own s…
A new paper from Meta, CMU, and other labs presents Self-play SWE-RL, a method where coding agents train themselves by manufacturing and fixing bugs in real codebases, achieving significant gains on SWE-bench benchmarks without relying on human-written tasks.
@rohanpaul_ai: New John Hopkins and Carnegie Mellon University Paper Shows that a small model can learn to improve an agent's harness …
A Johns Hopkins and Carnegie Mellon paper shows a 4B editor model can learn to rewrite an agent's harness code from failure reports, with edits transferring to unseen tasks and outperforming its 35B teacher on average edit scores.
@rvaniaaaa: Almost every self-improving agent i looked at had the same missing piece. People build the discovery part. Scout finds …
A tweet discusses the missing piece in self-improving agents: a strict review gate where an experienced engineer must approve each skill. It explains how a growing library improves discovery and extraction quality, with humans approving all merges to keep the system trustworthy.