Clauday.

A paper from PhAI Labs, submitted September 16 and sitting near the top of Hugging Face daily papers, names a problem the agent field has been dancing around: the scientific experi…

Luo Fuli, who runs Xiaomi's MiMo team and came out of DeepSeek, broke roughly six months of silence on September 16 by doing the single most un-lab-like thing available: turning th…

Postgres has had thirty years of very smart people tuning its query planner. Rohan Bansal trained a 4B model to write hints that beat it by 1.81x geometric mean on the Join Order B…

The expensive part of a self-improving agent is not the improving, it is the trying. Every policy change has to be tested against the real world, and in algorithm engineering or GP…

DataFlex-RL sat at the top of HuggingFace's daily papers board with 96 upvotes, and the reason is that it is a negative result, which almost nobody publishes. Paper at https://arxi…

RadixArk dropped Miles v0.1 and called it production-level post-training, which is a boring name for the least boring thing on Hugging Face's board today. This is the plumbing unde…

While a researcher was quitting Anthropic over self-improving AI, a team called TokenRhythm put a working prototype of it on Hugging Face under Apache 2.0. NeoHorse-1 is two checkp…

Two papers landed on arXiv on September 3 saying the same thing from opposite directions: the bottleneck in agent RL is no longer models or compute, it is environments to train the…

Self-improvement loops have a standard cause of death: the judge. Fix a reward model or an LLM judge in place, let the policy improve against it, and within a couple of rounds the …

The top paper on Hugging Face's daily board (118 upvotes) inverts the usual relationship between agents and environments. Instead of training agents inside games, "Agentic Game Dev…