NovBench (RAG) - Relevance: leaderboard
NovBench novelty assessment of EMNLP 2023 papers with a retrieval-augmented prompt with the five most similar ACL, EMNLP and NAACL 2019-2022 titles and abstracts. Relevance is how well the generated assessment covers each novelty claim, as an information-matching score from 0 to 5, higher is better.
Metric: Relevance (AvgIMS, 0-5; the paper states a 1-5 scale but prints values below 1): for each novelty-description sentence the best information-matching score against any sentence of the generated evaluation, averaged; retrieval-augmented prompt with the five most similar ACL, EMNLP and NAACL 2019-2022 titles and abstracts; 1,684 EMNLP 2023 papers: from the novelty-description sentences of the paper introduction (selected by GPT-5) the model writes a novelty evaluation as positive, neutral and negative statements, compared with the expert reviewers novelty comments (extracted by GPT-4o-mini, then deduplicated, consolidated and sorted by sentiment by GPT-4o); greedy decoding, at most 4,096 tokens; higher is better. Source: arxiv.org. Saturation forecast: Around July 2028. 19 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | SEA-S | 3.52 | |
| 2 | Gemini 2.5 Flash | 3.51 | #463 |
| 3 | GPT-4o | 3.45 | #751 |
| 4 | SEA-E | 3.38 | |
| 5 | Qwen 3 8B | 3.3 | #1873 |
| 6 | Qwen 3 32B | 3.26 | #972 |
| 7 | GPT-5 | 3.23 | #178 |
| 8 | Qwen 3 14B | 3.2 | #1269 |
| 9 | GPT-OSS-120B | 3.14 | #704 |
| 10 | GPT-OSS-20B | 3.14 | #1183 |
| 11 | DeepSeek R1 Distill Qwen 14B | 3.07 | #2771 |
| 12 | DeepSeek R1 Distill Llama 70B | 3.04 | #1669 |
| 13 | CycleReviewer-70B | 3.03 | #1880 |
| 14 | CycleReviewer-8B | 2.89 | |
| 15 | DeepReviewer-14B | 2.73 | #844 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=novbench-rag-relevance · How It Works · Data refreshed daily, snapshot 2026-11-21.