Tagged articles

skill distillation

4 articles · Page 1 of 1

Sep 14, 2026 · Frontend Development

AI Coding: Escaping the Blind Box — Contracts and Gates for Verifiable Frontend Development

The author argues that AI can generate 90% of complex calendar UIs but the remaining 10-20% becomes a blind-box guessing game because developers cannot verify fixes; they propose four workflow gates (contract, main path, link, container) and executable skills to replace random patching with provable correctness.

AI-assisted developmentState ManagementWebView integration

0 likes · 27 min read

AI Coding: Escaping the Blind Box — Contracts and Gates for Verifiable Frontend Development

Baobao Algorithm Notes

Aug 4, 2026 · Artificial Intelligence

Agentic RL: Cutting‑Edge Techniques from GLM‑5.2 and Qwen

The article dissects recent Agentic RL breakthroughs—including GLM‑5.2’s shift from GRPO to critic‑based PPO, Qwen’s multi‑dimensional verification system, the generative‑critic GenAC, and the on‑policy skill‑distillation method OPID—showing how each tackles long‑trajectory credit assignment, reward hacking, and scalable evaluation across software‑engineering, front‑end, and real‑world tasks.

Agentic RLGLM-5.2Generative Critic

0 likes · 36 min read

Agentic RL: Cutting‑Edge Techniques from GLM‑5.2 and Qwen

Digital Planet

Apr 10, 2026 · Industry Insights

How AI Turns Employees into Replicable Skills – From Storing Experience to Mastering Rules

The article examines why stagnant personal output, slowing corporate growth, and a slowing economy all share a common logic, explains how AI "old employee" systems evolve from merely recording experience to actively reproducing decision‑making rules, and outlines how individuals can maintain value by crossing the execution‑definition boundary that AI cannot replicate.

AIcareer growthexecution vs definition

0 likes · 14 min read

How AI Turns Employees into Replicable Skills – From Storing Experience to Mastering Rules

PaperAgent

Mar 2, 2026 · Artificial Intelligence

SKILLRL: Boosting LLM Agents with Skill Distillation and Recursive Evolution

SKILLRL introduces a novel framework that transforms raw LLM agent trajectories into compact, reusable skills via experience‑driven distillation, hierarchical skill banks, and recursive skill evolution, achieving up to 90% success on ALFWorld and 73% on WebShop while reducing token usage by over 10% compared to memory‑based baselines.

LLM agentsReinforcement LearningSKILLRL

0 likes · 10 min read

SKILLRL: Boosting LLM Agents with Skill Distillation and Recursive Evolution