CycleReviewer-70B: benchmark results
Provider: Other.
Aggregate Rating 1530 ± 48, rank #1952 of 10916 rated models, from 22 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| CoCoReviewBench - Clarity | 0.48 | Category-level clarity score (coherence, readability and str | 100 |
| NovBench - Coverage | 0.22 | Coverage (0-1): the share of the consolidated human novelty | 88.2 |
| Beyond Rating - Weakness Focus Divergence | 0.09 | KL divergence of the distribution of the generated reviews' | 71.4 |
| NovBench - Relevance | 3.46 | Relevance (AvgIMS, 0-5; the paper states a 1-5 scale but pri | 66.7 |
| NovBench (Few-Shot) - Correctness | 0.65 | Correctness (DistAcc, 0-1): one minus half the L1 distance b | 58.8 |
| NovBench (Few-Shot) - Coverage | 0.22 | Coverage (0-1): the share of the consolidated human novelty | 58.8 |
| NovBench - Correctness | 0.63 | Correctness (DistAcc, 0-1): one minus half the L1 distance b | 52.9 |
| NovBench (Few-Shot) - Relevance | 3.34 | Relevance (AvgIMS, 0-5; the paper states a 1-5 scale but pri | 50 |
| CoCoReviewBench - Correctness | -0.15 | Category-level correctness score (agreement of each review c | 47.1 |
| CoCoReviewBench | -0.16 | Average category-level review score (the mean of correctness | 41.2 |
| CoCoReviewBench - Paper Level | -0.35 | Paper-level review score (the same judge prompt applied once | 41.2 |
| NovBench (RAG) - Coverage | 0.15 | Coverage (0-1): the share of the consolidated human novelty | 41.2 |
Interactive version: theaggregate.ai/model?slug=cyclereviewer-70b · How It Works · Data refreshed daily, snapshot 2026-11-08.