SPLADE-v3: New baselines for SPLADE
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_ | 1866917610579296256 |
|---|---|
| author | Lassance, Carlos Déjean, Hervé Formal, Thibault Clinchant, Stéphane |
| author_facet | Lassance, Carlos Déjean, Hervé Formal, Thibault Clinchant, Stéphane |
| contents | A companion to the release of the latest version of the SPLADE library. We describe changes to the training structure and present our latest series of models -- SPLADE-v3. We compare this new version to BM25, SPLADE++, as well as re-rankers, and showcase its effectiveness via a meta-analysis over more than 40 query sets. SPLADE-v3 further pushes the limit of SPLADE models: it is statistically significantly more effective than both BM25 and SPLADE++, while comparing well to cross-encoder re-rankers. Specifically, it gets more than 40 MRR@10 on the MS MARCO dev set, and improves by 2% the out-of-domain results on the BEIR benchmark. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_06789 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | SPLADE-v3: New baselines for SPLADE Lassance, Carlos Déjean, Hervé Formal, Thibault Clinchant, Stéphane Information Retrieval Computation and Language A companion to the release of the latest version of the SPLADE library. We describe changes to the training structure and present our latest series of models -- SPLADE-v3. We compare this new version to BM25, SPLADE++, as well as re-rankers, and showcase its effectiveness via a meta-analysis over more than 40 query sets. SPLADE-v3 further pushes the limit of SPLADE models: it is statistically significantly more effective than both BM25 and SPLADE++, while comparing well to cross-encoder re-rankers. Specifically, it gets more than 40 MRR@10 on the MS MARCO dev set, and improves by 2% the out-of-domain results on the BEIR benchmark. |
| title | SPLADE-v3: New baselines for SPLADE |
| topic | Information Retrieval Computation and Language |
| url | https://arxiv.org/abs/2403.06789 |