ListT5: Listwise Reranking with Fusion-in-Decoder Improves Zero-shot Retrieval
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909217992998912 |
|---|---|
| author | Yoon, Soyoung Choi, Eunbi Kim, Jiyeon Yun, Hyeongu Kim, Yireun Hwang, Seung-won |
| author_facet | Yoon, Soyoung Choi, Eunbi Kim, Jiyeon Yun, Hyeongu Kim, Yireun Hwang, Seung-won |
| contents | We propose ListT5, a novel reranking approach based on Fusion-in-Decoder (FiD) that handles multiple candidate passages at both train and inference time. We also introduce an efficient inference framework for listwise ranking based on m-ary tournament sort with output caching. We evaluate and compare our model on the BEIR benchmark for zero-shot retrieval task, demonstrating that ListT5 (1) outperforms the state-of-the-art RankT5 baseline with a notable +1.3 gain in the average NDCG@10 score, (2) has an efficiency comparable to pointwise ranking models and surpasses the efficiency of previous listwise ranking models, and (3) overcomes the lost-in-the-middle problem of previous listwise rerankers. Our code, model checkpoints, and the evaluation framework are fully open-sourced at \url{https://github.com/soyoung97/ListT5}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_15838 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | ListT5: Listwise Reranking with Fusion-in-Decoder Improves Zero-shot Retrieval Yoon, Soyoung Choi, Eunbi Kim, Jiyeon Yun, Hyeongu Kim, Yireun Hwang, Seung-won Information Retrieval We propose ListT5, a novel reranking approach based on Fusion-in-Decoder (FiD) that handles multiple candidate passages at both train and inference time. We also introduce an efficient inference framework for listwise ranking based on m-ary tournament sort with output caching. We evaluate and compare our model on the BEIR benchmark for zero-shot retrieval task, demonstrating that ListT5 (1) outperforms the state-of-the-art RankT5 baseline with a notable +1.3 gain in the average NDCG@10 score, (2) has an efficiency comparable to pointwise ranking models and surpasses the efficiency of previous listwise ranking models, and (3) overcomes the lost-in-the-middle problem of previous listwise rerankers. Our code, model checkpoints, and the evaluation framework are fully open-sourced at \url{https://github.com/soyoung97/ListT5}. |
| title | ListT5: Listwise Reranking with Fusion-in-Decoder Improves Zero-shot Retrieval |
| topic | Information Retrieval |
| url | https://arxiv.org/abs/2402.15838 |