@dosco: i'm seeing a lot of industry papers that are karpathy's auto research loop (not cited) or a codex optimization goal for…
Summary
A critical observation about recent industry AI papers lacking novelty, citing examples like SkillOpt that treat natural-language skills as trainable external parameters.
View Cached Full Text
Cached at: 05/26/26, 07:04 AM
i’m seeing a lot of industry papers that are karpathy’s auto research loop (not cited) or a codex optimization goal for improving one specific thing turned into a system and a paper. at the risk of sounding negative but trying to provide honest feedback i fail to see the novelty in this whole genre of papers eg. skill-opt, skill-forge, and skills-coach
Yifan Yang (@Yif_Yang): 🚀 Introducing SkillOpt — an optimizer for agent skills.
Instead of finetuning model weights, we treat a natural-language skill as a trainable external parameter.
Think of it as deep learning for the frontier-model + agent era: learning rate, LR schedule, mini-batch, batch
Similar Articles
@dair_ai: https://x.com/dair_ai/status/2061104052818108476
A roundup of three notable AI papers: SkillOpt treats skill documents as trainable parameters to optimize frozen agents; a new method compiles agentic workflows into model weights for 100x cost reduction; and AutoScientists introduces a decentralized agent team for long-running science without a central planner.
@omarsar0: Karpathy's autoresearch repo started an impressive trend. Agents can now train AI models to build SoTA agentic systems.…
Karpathy's autoresearch repository has sparked a trend where agents train AI models to build state-of-the-art agentic systems, highlighting current limitations in LLM-driven hypothesis generation.
@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.
@lftherios: 1/ Autoresearch from @karpathy has been one of the most interesting agentic patterns to emerge this year. The challenge…
Andrej Karpathy's autoresearch pattern highlights how current AI agents run experiments in isolation, wasting compute by duplicating work and rediscovering dead ends.
@dair_ai: https://x.com/dair_ai/status/2068724104815890889
Highlights three recent AI papers: SpatialClaw (training-free spatial reasoning via code), SkillWeaver (compositional skill routing with decompose-retrieve-compose pipeline), and PreAct (compiling agent runs into fast state machines for repeated tasks).