Commonsense Reasoning on BBH (test)
159.89Throughput (Tokens/s)ES-dLLM
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| ES-dLLMBackbone=LLaDA-8B-Instruct, Number of shots=32026.03 | 159.89 | 14.5 | 54.51 | |
| ES-dLLM*Backbone=LLaDA-8B-Instruct, Number of shots=32026.03 | 133.48 | 12.1 | 56.66 | |
| DualCacheBackbone=LLaDA-8B-Instruct, Number of shots=32026.03 | 130.14 | 11.8 | 53.29 | |
| LLaDABackbone=LLaDA-8B-Instruct, Number of shots=32026.03 | 11.06 | 1 | 56.75 |