Rethinking Soft Compression in Retrieval-Augmented Generation: A Query-Conditioned Selector Perspective

TL;DR

This paper introduces SeleCom, a query-conditioned selector framework for retrieval-augmented generation that improves efficiency and performance by selectively compressing relevant information, addressing limitations of full compression methods.

Contribution

The paper proposes a novel selector-based soft compression method for RAG, redefining the encoder as a query-conditioned information selector trained with curriculum learning.

Findings

SeleCom outperforms existing soft compression methods.

SeleCom achieves comparable or better performance than non-compression baselines.

SeleCom reduces computation and latency by up to 84.6%.

Abstract

Retrieval-Augmented Generation (RAG) effectively grounds Large Language Models (LLMs) with external knowledge and is widely applied to Web-related tasks. However, its scalability is hindered by excessive context length and redundant retrievals. Recent research on soft context compression aims to address this by encoding long documents into compact embeddings, yet they often underperform non-compressed RAG due to their reliance on auto-encoder-like full-compression that forces the encoder to compress all document information regardless of relevance to the input query. In this work, we conduct an analysis on this paradigm and reveal two fundamental limitations: (I) Infeasibility, full-compression conflicts with the LLM's downstream generation behavior; and (II) Non-necessity: full-compression is unnecessary and dilutes task-relevant information density. Motivated by these insights, we introduce SeleCom, a selector-based soft compression framework for RAG that redefines the encoder's role as query-conditioned information selector. The selector is decoder-only and is trained with a massive, diverse and difficulty-graded synthetic QA dataset with curriculum learning. Extensive experiments show that SeleCom significantly outperforms existing soft compression approaches and achieves competitive or superior performance to non-compression baselines, while reducing computation and latency by 33.8%~84.6%.