Empowering Language Model with Guided Knowledge Fusion for Biomedical Document Re-ranking
This work addresses the challenge of accurate document retrieval for biomedical and healthcare queries, which is incremental as it builds on existing PLM-based re-ranking methods.
The paper tackles the problem of biomedical document re-ranking by integrating knowledge with pre-trained language models to better interpret semantics, resulting in significant improvements over vanilla PLMs and existing methods on biomedical and open-domain datasets.
Pre-trained language models (PLMs) have proven to be effective for document re-ranking task. However, they lack the ability to fully interpret the semantics of biomedical and health-care queries and often rely on simplistic patterns for retrieving documents. To address this challenge, we propose an approach that integrates knowledge and the PLMs to guide the model toward effectively capturing information from external sources and retrieving the correct documents. We performed comprehensive experiments on two biomedical and open-domain datasets that show that our approach significantly improves vanilla PLMs and other existing approaches for document re-ranking task.