OpenDS-World launches live-web deep search benchmark, frontier models trail humans by large margins
Xiaomin Yu · hf · 2026-10-23
A multi-institution team (HKUST-GZ, Fudan, UW, Stanford, Yale and more) released OpenDS-World, a live-web benchmark for evaluating autonomous multi-step deep search agents on the real internet, with code and dataset open-sourced. The paper reports significant performance gaps between frontier models and humans, exposing weaknesses in multi-step retrieval and synthesis on real web pages.
More from coding & agent
- Dev shares two AGENTS.md lines to make AI-generated UIs look less like AI — pjausovec · 2026-10-23
- Meta, Google, OpenAI and xAI all converge on cloud computers for personal agents — sujingshen · 2026-10-23
- diffy semantic diff app to auto-discover Git worktrees in next release — seatedro · 2026-10-23
- How a 5-person AI film crew produced an Odyssey pilot scene in Imagine — chaitu · 2026-10-23
- "One prompt built it all!" Frontier dev: we refactor like this several times a day — fkasummer · 2026-10-23
- Study: LLMs chime in at unwelcome moments 46.4% of the time in group chats — windx0303 · 2026-10-23