Unverified50% confidenceFactExact time
The researchers re-analyzed 540 evaluation results and manually annotated 256 numerical claims across 18 submissions.
1
Sources
50%
Confidence
Long-term
Relevance
10/30/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe article reviewed 232 inference providers and 470 models.73% similarUnverifiedBy re-tallying the results files, readers found a row of data that the authors had skipped: the maintainer file lists 9 of 212 tests judged WEAK (4%), while the agent file lists 79 of 591 (13%).70% similarUnverifiedAmong the approximately 2,000 papers examined, more than 1,100 had at least one key claim independently verified, of which 266 were fully reproduced, 632 were partially reproduced through small-scale experiments, and approximately 4,000 individual claims were confirmed.69% similarUnverifiedThis benchmark uses 10 original short documents of 200 to 250 words each, with 18 planted anomalies, 37 scoring entries in total (18 dirty and 19 clean), testing 13 models with 74 prompts per model.69% similarUnverifiedOne historical predecessor path reproduced 17 frozen authorization-successor outcomes, instance blind tests matched all 183 scoring expectations, and software packages reproduced 240 archived observation results.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1114265API
curl https://kongchang.com/api/v1/knowledge/claims/1114265MCP
get_claim(id=1114265)