Dialogue Generation on Wizard of Wikipedia (WoW) (dev)
16.4F1 ScoreKID
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| KIDBackbone=GPT2-medium, Parameters=345M, Evaluation Protocol=fine-tuned, Decoding Strategy=KID2022.04 | 16.4 | 15.9 | — | |
| KIDBackbone=BART-large, Parameters=406M, Evaluation Protocol=fine-tuned, Decoding Strategy=KID2022.04 | 15.7 | 16.6 | — | |
| BART+DPRRole=Published SotA2022.04 | 13.5 | 15.5 | — | |
| GPT2-mediumBackbone=GPT2-medium, Parameters=345M, Evaluation Protocol=fine-tuned, Decoding Strategy=Sampling2022.04 | 12.6 | 11.9 | — | |
| BART-largeBackbone=BART-large, Parameters=406M, Evaluation Protocol=fine-tuned, Decoding Strategy=Sampling2022.04 | 12.2 | 15 | — | |
| BIDERrefinement_type=Abstractive refinement2024.02 | 12.2 | — | 69 | |
| LLM-Embedderrefinement_type=Extractive refinement2024.02 | 11.8 | — | 179 | |
| Bge-Rerankerrefinement_type=Extractive refinement, additional_training=true2024.02 | 11.7 | — | 180 | |
| BART-largeBackbone=BART-large, Parameters=406M, Evaluation Protocol=fine-tuned, Decoding Strategy=Beam2022.04 | 10.9 | 11.9 | — | |
| GPT2-mediumBackbone=GPT2-medium, Parameters=345M, Evaluation Protocol=fine-tuned, Decoding Strategy=Beam2022.04 | 10.3 | 12.1 | — | |
| GPT2-mediumBackbone=GPT2-medium, Parameters=345M, Evaluation Protocol=zero-shot2022.04 | 9.3 | 10.8 | — | |
| Original Promptrefinement_type=Without refinement2024.02 | 8.6 | — | 710 | |
| SBERTrefinement_type=Extractive refinement2024.02 | 8.6 | — | 178 | |
| Zero-shotrefinement_type=Without refinement2024.02 | 8.5 | — | 0 | |
| BM25refinement_type=Extractive refinement2024.02 | 8.5 | — | 177 | |
| BART-Summarizerrefinement_type=Abstractive refinement2024.02 | 8.5 | — | 194 | |
| BART-largeBackbone=BART-large, Parameters=406M, Evaluation Protocol=zero-shot2022.04 | 8.1 | 8.5 | — | |
| Selective-Contextrefinement_type=Abstractive refinement2024.02 | 8.1 | — | 220 | |
| LongLLMLinguarefinement_type=Abstractive refinement2024.02 | 7.7 | — | 124 |