Unverified50% confidenceFactExact time
The research team evaluated eight agents on TCSExam-Full and nine agents on TCSExam-Lite.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/6/2026
First Seen
Valid until: 2/4/2027
Sources
Related Entities
Related Claims
UnverifiedTCSExam-Full contains 100 expert-reviewed tasks, and all submissions are verified by the Lean kernel or executable task checkers.65% similarUnverifiedThe research team tested nine proprietary and open-weight models across five coding agent frameworks, and concluded that even the strongest agents could not fully solve the task of digitizing protocols.64% similarUnverifiedThe research team conducted experimental validation on five widely used GUI agents.63% 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.63% similarUnverifiedA study published on arXiv (arXiv:2609.15995) built a unified inference gateway that runs the same evaluation suite for ten external auditing tools across a shared panel of ten frontier models.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1141688API
curl https://kongchang.com/api/v1/knowledge/claims/1141688MCP
get_claim(id=1141688)