ACL Findings2026
BRIEF-Pro: Universal Context Compression with Short-to-Long Synthesis for Fast and Accurate Multi-Hop Reasoning
Jia-Chen Gu, Junyi Zhang, Di Wu, Yuankai Li, Kai-Wei Chang, Nanyun Peng
Abstract
As retrieval-augmented generation (RAG) tackles complex tasks, increasingly expanded contexts offer richer information, but at the cost of higher latency and increased cognitive load on the model. To mitigate this bottleneck, especially for intricate multi-hop questions, we introduce BRIEF-PRO. It is a universal, lightweight compressor that distills relevant evidence for a given query from retrieved documents into a concise summary for seamless integration into in-context RAG. Using seed data consisting of relatively short contexts (fewer than 1k words), BRIEF-PRO is trained to perform abstractive compression of extended contexts exceeding 10k words across a wide range of scenarios. Furthermore, BRIEF-PRO offers flexible user control over summary length by allowing users to specify the desired number of sentences. Experiments on four open-domain multi-hop question-answering datasets show that BRIEF-PRO generates more concise and relevant summaries, enhancing performance across small, large, and proprietary language models. With the 70B reader model, 32× compression by BRIEF-PRO improves QA performance by 4.67% on average over LongLLMLingua's 9×, while requiring only 23% of its computational overhead 1 .
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
- Learning to Compress Prompts with Gist Tokens cites LoRA: Low-Rank Adaptation of Large Language Models
- When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric Memories cites Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
Builds on18
- Retrieval-Augmented Generation for Knowledge-Intensive NLP TasksPatrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni et al.NeurIPS 2020 · 19,162 citations
- LoRA: Low-Rank Adaptation of Large Language ModelsEdward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu et al.ICLR 2022 · 18,833 citations
- Large Language Models Can Be Easily Distracted by Irrelevant ContextFreda Shi, Xinyun Chen, Kanishka Misra, Nathan Scales et al.ICML 2023 · 970 citations
- Learning to Compress Prompts with Gist TokensJesse Mu, Xiang Li, Noah D. GoodmanNeurIPS 2023 · 488 citations
- When Not to Trust Language Models: Investigating Effectiveness of Parametric and Non-Parametric MemoriesAlex Mallen, Akari Asai, Victor Zhong, Rajarshi Das et al.ACL 2023 · 233 citations
Related papers
- EXIT: Context-Aware Extractive Compression for Enhancing Retrieval-Augmented GenerationTaeho Hwang, Sukmin Cho, Soyeong Jeong, Hoyun Song et al.ACL Findings 2025 · 10 citations
- CompAct: Compressing Retrieved Documents Actively for Question AnsweringChanwoong Yoon, Taewhoo Lee, Hyeon Hwang, Minbyul Jeong et al.EMNLP 2024 · 10 citations
- RECOMP: Improving Retrieval-Augmented LMs with Context Compression and Selective AugmentationFangyuan Xu, Weijia Shi, Eunsol ChoiICLR 2024 · 260 citations
- Enhancing RAG Efficiency with Adaptive Context CompressionShuyu Guo, Shuo Zhang, Zhaochun RenEMNLP Findings 2025 · 5 citations
- ECoRAG: Evidentiality-guided Compression for Long Context RAGYeonseok Jeong, Jinsu Kim, Dohyeon Lee, Seung-won HwangACL Findings 2025 · 4 citations
