Summary of Xrag: Extreme Context Compression For Retrieval-augmented Generation with One Token, by Xin Cheng et al.
xRAG: Extreme Context Compression for Retrieval-augmented Generation with One Token
by Xin Cheng, Xun Wang, Xingxing Zhang, Tao Ge, Si-Qing Chen, Furu Wei, Huishuai Zhang, Dongyan Zhao
First submitted to arxiv on: 22 May 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper introduces xRAG, a novel context compression method designed for retrieval-augmented generation. It repurposes document embeddings in dense retrieval as features from the retrieval modality, allowing for seamless integration with language model representation spaces. The only trainable component is the modality bridge, enabling reuse of offline-constructed document embeddings and preserving plug-and-play nature. Experimental results demonstrate xRAG’s average 10% improvement across six knowledge-intensive tasks on various language model backbones, while reducing FLOPs by a factor of 3.53. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary xRAG is a new way to make computers understand text better. It takes old information and makes it fit into special boxes that help the computer learn more efficiently. The best part is that it can work with different kinds of language models, which are like super smart computers that learn from lots of text. This innovation is important because it means we can train computers to do tasks faster and better, which could lead to amazing breakthroughs in areas like medicine, science, and technology. |
Keywords
» Artificial intelligence » Language model » Retrieval augmented generation
