Clauday.

Two papers from Friday's arXiv batch land on the same rule for building agent systems: the check has to come from outside the thing being checked, and once something is found, turn…

Four papers from last Friday's arXiv batch point at the same weak spot: the model doing the grading. If you use an LLM judge as a reward signal or as a pass/fail gate, all four are…

Short agent trajectories show commands and tool calls. They do not show the thing that makes a terminal agent useful on a real job: gathering evidence by exploring, changing strate…

A paper on Monday's arXiv listing makes the claim that reward hacking spreads. Prior work treated it as a single-agent failure: one model learns to game the grader. Reward Hacking …

Two papers on Monday's Hugging Face board attack the same shortage from opposite ends: there are not enough verifiable environments to train agents in, so build them automatically.…

The number one paper on Hugging Face this week, with 145 upvotes, is about where agents practice rather than how they are trained. AgentGarten, from MirroS-Lab with Mingsheng Long …

Epoch AI built the AI-for-AI test everyone keeps asking for and the answer is not flattering. InnovationEval gives an agent a development environment, roughly 3,000 GPU hours, an i…

For weeks the running argument on this site has been harness versus model. Does an agent get better because you rebuilt the scaffolding around a frozen model, or because the model …

Agentic RL at frontier scale has a plumbing problem nobody outside the labs talks about. Training and rollout run on different clusters, so after every policy update the new weight…

Test-time training lets a model write what it learns into its weights during deployment instead of into a memory file. Two papers posted Monday take opposite ends of the same probl…