NovBench (Few-Shot) - Coverage: leaderboard

NovBench novelty assessment of EMNLP 2023 papers with a few-shot prompt with two analogous NovBench examples. Coverage is the share of the expert reviewers novelty comments that a generated statement matches by embedding similarity, from 0 to 1, higher is better.

Metric: Coverage (0-1): the share of the consolidated human novelty comments matched by a generated statement with embedding cosine similarity above 0.7; few-shot prompt with two analogous NovBench examples; 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: Not forecast. 18 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1SEA-E0.2483
2SEA-S0.2454
3Gemini 2.5 Flash0.2364#463
4DeepSeek R1 Distill Qwen 14B0.2252#2771
5Qwen 3 32B0.2242#972
6Qwen 3 8B0.224#1873
7Qwen 3 14B0.2238#1269
8CycleReviewer-70B0.2209#1880
9DeepSeek-R1-8B (NovBench checkpoint unspecified)0.219
10DeepSeek R1 Distill Llama 70B0.2112#1669
11GPT-OSS-20B0.2098#1183
12CycleReviewer-8B0.1892
13GPT-OSS-120B0.183#704
14GPT-50.1806#178
15DeepReviewer-14B0.114#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-few-shot-coverage · How It Works · Data refreshed daily, snapshot 2026-11-21.