Unverified50% confidenceBenchmarkExact time
In the author's published 25-question small-sample evaluation, ZY-S1 improved over the base model on ARC-Challenge from 76 to 80 (+4 percentage points) and on MMLU from 48 to 52 (+4 percentage points).
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/13/2026
First Seen
Valid until: 10/27/2026
Sources
Related Entities
Related Claims
UnverifiedIn the author's published 25-question small-sample evaluation, ZY-S1 dropped from 4 to 0 on MATH-500 compared with the base model, showing degradation in mathematical ability.75% similarUnverifiedThe score gap among multiple frontier models on MMLU has narrowed to the order of one percentage point, driving the emergence of more challenging benchmarks such as MMLU-Pro and ARC-AGI.67% similarUnverifiedMid-tier models typically achieve more than 80% of flagship model scores on benchmarks such as MMLU and HumanEval, while requiring less than 30% of the inference cost.66% similarUnverifiedIn the intervention experiment, the donor-control AUC rose from 0.254 to 0.540, with a paired change of +0.286 (95% confidence interval [+0.272, +0.300]), and an increase was observed across all 18 concepts.66% similarVerifiedOn mainstream evaluations such as MMLU, HumanEval, and GSM8K, the score gaps among leading large models are often only a few percentage points.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1014223API
curl https://kongchang.com/api/v1/knowledge/claims/1014223MCP
get_claim(id=1014223)