LLM
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Binary Inconsistency Detection | LLM | Accuracy70.27 | 10 | |
| Robust Steganography | LLM Generative Text | Embedding Capacity (bits / 1k tokens)84.08 | 5 | |
| Span Detection | LLM | F1 Score0.3322 | 5 | |
| Language Modeling | LLM (val) | Loss1.3364 | 4 | |
| Language | LLM-329M | Peak Performance (FP4/FP8)205 | 1 |
Showing 5 of 5 rows