@FinanceYF5: 2/ SkillOpt: Treating Documents as Trainable Parameters Microsoft treats SKILL.md as trainable model parameters—without changing weights, only optimizing natural language documents, with a validation gate filtering each change. 6 Benchmarks, 52 consecutive wins, GPT-5.5 conversation boost…
Summary
Microsoft proposes the SkillOpt method, which treats documents as trainable parameters. By optimizing natural language documents without modifying weights, it improves model performance. It achieves 52 consecutive wins across 6 benchmarks, with GPT-5.5 improving by 23.5 points and Claude Code by 19.1 points.
View Cached Full Text
Cached at: 06/01/26, 11:18 AM
2/📄 SkillOpt: Treating Documents as Trainable Parameters
Microsoft treats SKILL.md as a trainable model parameter — without modifying weights, only optimizing natural language documentation, with a validation gate filtering each change.
52 consecutive wins across 6 benchmarks, GPT-5.5 direct chat improved by +23.5 points, Claude Code improved by +19.1 points.
Similar Articles
@omarsar0: New research from Microsoft Research I see a lot of AI engineers handwriting agent skill docs and hope they generalize.…
Microsoft Research introduces SkillOpt, a method that treats agent skill documents as trainable external state, using an optimizer model to make bounded edits validated by a held-out set. The approach achieves best or tied results across 52 evaluation cells and improves accuracy by over 23 points on GPT-5.5, with zero extra inference cost and transferable skills.
@Yif_Yang: Introducing SkillOpt — an optimizer for agent skills. Instead of finetuning model weights, we treat a natural-language …
Introducing SkillOpt, an optimizer that treats natural-language skills as trainable external parameters instead of finetuning model weights. It uses bounded edits and validation gating to enable stable, controllable skill updates, achieving best or tied-best results across 52 settings on 6 benchmarks with 7 models.
SkillOpt treats markdown skill files as trainable parameters with proper optimization machinery
A new paper formalizes skill optimization for agents by treating markdown skill files as trainable parameters, using bounded edits validated against holdout sets. The approach transfers well between models and improves performance on procedural benchmarks.
@Xudong07452910: This SkillOpt paper is quite interesting—it actually addresses a very important point: AI agents in the future won't just rely on humans writing prompts; they can train their own 'job descriptions'. Currently, many skills/prompts are written one-off, and when real tasks pile up, various edge cases start to fail...
SkillOpt introduces a systematic controllable text-space optimizer that enables AI agents to train and improve their own skills (like 'work instructions') through iterative edits and validation, outperforming human-crafted and one-shot prompts across multiple benchmarks and models.
@DAIEvolutionHub: MICROSOFT JUST OPEN-SOURCED A WAY TO “TRAIN” AI AGENTS WITHOUT TOUCHING MODEL WEIGHTS SkillOpt treats a simple markdown…
Microsoft open-sourced SkillOpt, a method that treats markdown skill files like neural network parameters to train AI agents without modifying model weights, using learning rates, validation checks, minibatches, and epochs.