MT
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Machine Translation | MT en -> fr | BLEU39.01 | 14 | |
| Machine Translation | MT en -> de | BLEU29.41 | 14 | |
| Machine Translation | MT (test) | Average Win Rate58.24 | 12 | |
| Generative Recommendation | MT | Latency (ms)11.7 | 9 | |
| Link Prediction | MT2 org | MRR10.6 | 9 | |
| Makeup Transfer | MT | L2M Score0.183 | 8 | |
| Retrieval | MT Small | Recall@101.12 | 6 | |
| Retrieval | MT Large | Recall@100.97 | 6 | |
| Graph classification | MT standard | Accuracy88.4 | 6 | |
| retrieval | MT-Other Cities High-Freq | Recall@50.5342 | 5 | |
| Makeup Transfer | MT (test) | Avg User Rating4.35 | 5 | |
| Machine Translation | MT Qwen2-7B-Instruct v1 (test) | Acceptance Length (τ)2.2 | 4 | |
| Inductive Link Prediction | MT2 sci | MRR25.8 | 4 | |
| Makeup Transfer | MT | FID12.07 | 4 | |
| Industrial Defect Detection | MT (test) | mAP93.79 | 3 | |
| Machine Translation | MT | Throughput (tokens/s)188.69 | 3 | |
| Machine Translation | MT | Performance Loss11.2 | 1 |
Showing 17 of 17 rows