Summarization on Summarization (Strong User)
0.1845Mean Total Edit DistanceEarlyEnsemble
Evaluation Results
| Method | Links | |
|---|---|---|
| EarlyEnsembleTest-time User Model=Llama-3.3-70B-Instruct, Online learning phase=true2026.01 | 0.1845 | |
| EarlyEnsembleUser LLM=Qwen-3 32B, Seeds=3, Learning Phase=Online2026.01 | 0.2092 | |
| LateEnsembleTest-time User Model=Llama-3.3-70B-Instruct, Online learning phase=true2026.01 | 0.2509 | |
| LateEnsembleUser LLM=Qwen-3 32B, Seeds=3, Learning Phase=Online2026.01 | 0.2768 | |
| SFTTest-time User Model=Llama-3.3-70B-Instruct, Online learning phase=true2026.01 | 0.489 | |
| SFTUser LLM=Qwen-3 32B, Seeds=3, Learning Phase=Online2026.01 | 0.5377 | |
| BaseTest-time User Model=Llama-3.3-70B-Instruct, Online learning phase=true2026.01 | 0.8498 | |
| BaseUser LLM=Qwen-3 32B, Seeds=3, Learning Phase=Online2026.01 | 0.9455 | |
| DPOTest-time User Model=Llama-3.3-70B-Instruct, Online learning phase=true2026.01 | 0.965 | |
| DPOUser LLM=Qwen-3 32B, Seeds=3, Learning Phase=Online2026.01 | 1.079 |