Multi-Aspect Reviewed-Item Retrieval via LLM Query Decomposition and Aspect Fusion
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866929446556008448 |
|---|---|
| author | Korikov, Anton Saad, George Baron, Ethan Khan, Mustafa Shah, Manav Sanner, Scott |
| author_facet | Korikov, Anton Saad, George Baron, Ethan Khan, Mustafa Shah, Manav Sanner, Scott |
| contents | While user-generated product reviews often contain large quantities of information, their utility in addressing natural language product queries has been limited, with a key challenge being the need to aggregate information from multiple low-level sources (reviews) to a higher item level during retrieval. Existing methods for reviewed-item retrieval (RIR) typically take a late fusion (LF) approach which computes query-item scores by simply averaging the top-K query-review similarity scores for an item. However, we demonstrate that for multi-aspect queries and multi-aspect items, LF is highly sensitive to the distribution of aspects covered by reviews in terms of aspect frequency and the degree of aspect separation across reviews. To address these LF failures, we propose several novel aspect fusion (AF) strategies which include Large Language Model (LLM) query extraction and generative reranking. Our experiments show that for imbalanced review corpora, AF can improve over LF by a MAP@10 increase from 0.36 to 0.52, while achieving equivalent performance for balanced review corpora. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_00878 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Multi-Aspect Reviewed-Item Retrieval via LLM Query Decomposition and Aspect Fusion Korikov, Anton Saad, George Baron, Ethan Khan, Mustafa Shah, Manav Sanner, Scott Information Retrieval While user-generated product reviews often contain large quantities of information, their utility in addressing natural language product queries has been limited, with a key challenge being the need to aggregate information from multiple low-level sources (reviews) to a higher item level during retrieval. Existing methods for reviewed-item retrieval (RIR) typically take a late fusion (LF) approach which computes query-item scores by simply averaging the top-K query-review similarity scores for an item. However, we demonstrate that for multi-aspect queries and multi-aspect items, LF is highly sensitive to the distribution of aspects covered by reviews in terms of aspect frequency and the degree of aspect separation across reviews. To address these LF failures, we propose several novel aspect fusion (AF) strategies which include Large Language Model (LLM) query extraction and generative reranking. Our experiments show that for imbalanced review corpora, AF can improve over LF by a MAP@10 increase from 0.36 to 0.52, while achieving equivalent performance for balanced review corpora. |
| title | Multi-Aspect Reviewed-Item Retrieval via LLM Query Decomposition and Aspect Fusion |
| topic | Information Retrieval |
| url | https://arxiv.org/abs/2408.00878 |