Unverified50% confidenceBenchmarkExact time
The evaluation covered nine text-based scientific benchmarks, sampling 4,531 questions and 100 matched profiles, with 4,488 items completing all five conditions in full.
1
Sources
50%
Confidence
Long-term
Relevance
10/2/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe evaluation comprised 108 checks across three major modules: backend API and full pipeline at 65 points, frontend pages and real interaction at 25 points, and build and packaging at 10 points.68% similarUnverifiedThe experiment used balanced accuracy as the evaluation metric, covering 8 classification datasets.68% similarUnverifiedThe paper distinguishes six levels of evaluation evidence: field-level, pair-level, list-level, case-level, trajectory-level, and outcome-level66% similarUnverifiedThe research team selected 8 models spanning different capability tiers and conducted cross-evaluations across 54 benchmarks66% similarUnverifiedThe experiments were tested across six heterogeneous benchmarks.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/966774API
curl https://kongchang.com/api/v1/knowledge/claims/966774MCP
get_claim(id=966774)