Ebisu - JF-TE - Hit Rate@5: leaderboard
Evaluates Japanese financial implicit commitment recognition (JF-ICR) and terminology extraction (JF-TE). This is the paper v1 result table, whose stated generation limit is 1,024; current repository task defaults differ. JF-ICR uses accuracy, JF-TE uses maximal-term F1 and hit rates at 1/5/10, and the mean combines five metrics. Published fractions are converted to percentages. Explicit zero scores are retained; malformed predictions can receive zero under the public evaluator. Source-specified dated checkpoints are retained. This series reports Ebisu - JF-TE - Hit Rate@5 using Hit Rate@5 (%).
Metric: Hit Rate@5 (%). Source: arxiv.org. Saturation forecast: Around July 2027. 22 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Llama 3.3 70B Instruct | 36.57 | #541 |
| 2 | Qwen 2.5 32B Instruct | 34.62 | #509 |
| 3 | Llama-3.3-Swallow-70B-Instruct-v0.4 | 32.7 | #540 |
| 4 | Kimi K2 | 32.51 | #252 |
| 5 | Llama 4 Scout | 30.94 | #648 |
| 6 | Qwen 3 235B A22B 2507 Instruct FP8 | 30.53 | #234 |
| 7 | Claude Sonnet 4.5 | 30.22 | #148 |
| 8 | GPT-4o | 28.45 | #357 |
| 9 | DeepSeek V3.1 | 26.63 | #275 |
| 10 | Ministral-3-14B-Instruct-2512 | 23.77 | #625 |
| 11 | Qwen 3 32B | 10.52 | #449 |
| 12 | Qwen 3 14B | 7.87 | #542 |
| 13 | Qwen-14B | 6.5 | #1153 |
| 14 | Qwen 3 8B | 1.42 | #678 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=ebisu-jf-te-hit-rate-5 · How It Works · Data refreshed daily, snapshot 2026-11-05.