Unverified50% confidenceFactExact time
A recent public benchmark for AI research agents covers nine agents and fifteen open research questions, and all 540 judging results have been released.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/30/2026
First Seen
Valid until: 1/28/2027
Sources
Related Entities
Related Claims
UnverifiedThe research team selected 8 models spanning different capability tiers and conducted cross-evaluations across 54 benchmarks71% similarUnverifiedThe COMPL-AI project provides a technical interpretation of the EU AI Act, organizing compliance evidence into six principle groups and eighteen technical requirements, twelve of which can be measured through existing benchmarks.71% similarUnverifiedThe research team evaluated SPLIT on seven benchmarks, covering math, science QA, code, and software engineering agents, involving 26 models, 175 agent systems, and six real verifiers, all from public releases.70% similarUnverified评估AI攻克难题类报道的四步判断法为：题目是否准确、证明是否公开、有没有独立审阅、能否形式化验算70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1114262API
curl https://kongchang.com/api/v1/knowledge/claims/1114262MCP
get_claim(id=1114262)