Beyond Prompting: An Efficient Embedding Framework for Open-Domain Question Answering

TY - CHAP

T1 - Beyond Prompting

T2 - An Efficient Embedding Framework for Open-Domain Question Answering

AU - Hu, Zhanghao

AU - Yan, Hanqi

AU - Zhu, Qinglin Zhu

AU - Shen, Zhenyi

AU - He, Yulan

AU - Gui, Lin

PY - 2025

Y1 - 2025

N2 - Large language models have recently pushed open domain question answering (ODQA) to new frontiers. However, prevailing retriever-reader pipelines often depend on multiple rounds of prompt level instructions, leading to high computational overhead, instability, and suboptimal retrieval coverage. In this paper, we propose EmbQA, an embedding-level framework that alleviates these shortcomings by enhancing both the retriever and the reader. Specifically, we refine query representations via lightweight linear layers under an unsupervised contrastive learning objective, thereby reordering retrieved passages to highlight those most likely to contain correct answers. Additionally, we introduce an exploratory embedding that broadens the model's latent semantic space to diversify candidate generation and employs an entropy-based selection mechanism to choose the most confident answer automatically. Extensive experiments across three open-source LLMs, three retrieval methods, and four ODQA benchmarks demonstrate that EmbQA substantially outperforms recent baselines in both accuracy and efficiency.

AB - Large language models have recently pushed open domain question answering (ODQA) to new frontiers. However, prevailing retriever-reader pipelines often depend on multiple rounds of prompt level instructions, leading to high computational overhead, instability, and suboptimal retrieval coverage. In this paper, we propose EmbQA, an embedding-level framework that alleviates these shortcomings by enhancing both the retriever and the reader. Specifically, we refine query representations via lightweight linear layers under an unsupervised contrastive learning objective, thereby reordering retrieved passages to highlight those most likely to contain correct answers. Additionally, we introduce an exploratory embedding that broadens the model's latent semantic space to diversify candidate generation and employs an entropy-based selection mechanism to choose the most confident answer automatically. Extensive experiments across three open-source LLMs, three retrieval methods, and four ODQA benchmarks demonstrate that EmbQA substantially outperforms recent baselines in both accuracy and efficiency.

M3 - Conference paper

BT - The 63rd Annual Meeting of the Association for Computational Linguistics, ACL 2025 - Proceedings of the Conference

PB - Association for Computational Linguistics

ER -