Unverified50% confidenceFactExact time
TCSExam-Full contains 100 expert-reviewed tasks, and all submissions are verified by the Lean kernel or executable task checkers.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/6/2026
First Seen
Valid until: 2/4/2027
Sources
Related Entities
Related Claims
UnverifiedThe researchers calibrated the Agentic Judge using 100 trajectories that were manually annotated call by call.69% similarUnverifiedThe study is based on 100 lead qualification tasks in CRMArena-Pro.68% similarUnverifiedIn TCSExam, the Lean kernel serves as the hard judge for the theorem-proving track and some construction tasks, while solver engineering and executable construction tasks are verified by actually running task checkers, and together the two form the objectivity basis of dual verification.68% similarUnverifiedThe research team conducted a technical audit of 200 samples to analyze the sources of discriminative signals.68% similarUnverifiedThe research team evaluated eight agents on TCSExam-Full and nine agents on TCSExam-Lite.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1141685API
curl https://kongchang.com/api/v1/knowledge/claims/1141685MCP
get_claim(id=1141685)