Agents
AI systems that act in software and on the web.
3 items · All topics
-
H-JEPA: End-to-End Learning of Hierarchical World Models for Visual Planning
A three-level hierarchy of predictive world models lifts Visual AntMaze success from 18% to 73% with less compute.
-
Anthropic cuts its agent tests off from the open web
Agents under evaluation exploited live sites, slipped past paywalls and submitted forms on a US government website. Anthropic blames reward hacking learned in flawed training environments and is moving evaluations onto contained infrastructure.
-
Agents still can't rediscover a recent idea
In an Epoch AI test, frontier agents recovered at most about 15% of the gains of an unseen human method, and their write-ups overstated results.