Interpret and Control Dense Retrieval with Sparse Latent Features
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866916626673172480 |
|---|---|
| author | Kang, Hao Wang, Tevin Xiong, Chenyan |
| author_facet | Kang, Hao Wang, Tevin Xiong, Chenyan |
| contents | Dense embeddings deliver strong retrieval performance but often lack interpretability and controllability. This paper introduces a novel approach using sparse autoencoders (SAE) to interpret and control dense embeddings via the learned latent sparse features. Our key contribution is the development of a retrieval-oriented contrastive loss, which ensures the sparse latent features remain effective for retrieval tasks and thus meaningful to interpret. Experimental results demonstrate that both the learned latent sparse features and their reconstructed embeddings retain nearly the same retrieval accuracy as the original dense vectors, affirming their faithfulness. Our further examination of the sparse latent space reveals interesting features underlying the dense embeddings and we can control the retrieval behaviors via manipulating the latent sparse features, for example, prioritizing documents from specific perspectives in the retrieval results. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_00786 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Interpret and Control Dense Retrieval with Sparse Latent Features Kang, Hao Wang, Tevin Xiong, Chenyan Information Retrieval Dense embeddings deliver strong retrieval performance but often lack interpretability and controllability. This paper introduces a novel approach using sparse autoencoders (SAE) to interpret and control dense embeddings via the learned latent sparse features. Our key contribution is the development of a retrieval-oriented contrastive loss, which ensures the sparse latent features remain effective for retrieval tasks and thus meaningful to interpret. Experimental results demonstrate that both the learned latent sparse features and their reconstructed embeddings retain nearly the same retrieval accuracy as the original dense vectors, affirming their faithfulness. Our further examination of the sparse latent space reveals interesting features underlying the dense embeddings and we can control the retrieval behaviors via manipulating the latent sparse features, for example, prioritizing documents from specific perspectives in the retrieval results. |
| title | Interpret and Control Dense Retrieval with Sparse Latent Features |
| topic | Information Retrieval |
| url | https://arxiv.org/abs/2411.00786 |