Topic
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Attribute Steering | Topic (T) | Confidence100 | 26 | |
| Reproduction Report Generation | Topic 4 PhyCustom Flux | Score83 | 6 | |
| Classification | Topic | Coverage Loss1 | 5 | |
| Topic-controlled generation | Topic Human Evaluation (test) | Fluency4.01 | 3 |
Showing 4 of 4 rows