Summary of From Reading to Compressing: Exploring the Multi-document Reader For Prompt Compression, by Eunseong Choi et al.
From Reading to Compressing: Exploring the Multi-document Reader for Prompt Compression
by Eunseong Choi, Sunkyung Lee, Minjin Choi, June Park, Jongwuk Lee
First submitted to arxiv on: 5 Oct 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
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 proposed Reading To Compressing (R2C) method uses a Fusion-in-Decoder (FiD) architecture to identify essential information in large language model prompts, addressing challenges in capturing global context and training the compressor effectively. R2C enhances performance by 6% in out-of-domain evaluations while reducing prompt length by 80%. The method leverages cross-attention scores to discern important chunks and sentences, ensuring semantic consistency without relying on pseudo-labels for training. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Large language models are getting better at understanding us with the help of advanced prompts. But these longer prompts take up a lot of computer power and can be hard to understand. To solve this problem, scientists have come up with a way to shorten these prompts while still keeping the important information. They call it Reading To Compressing (R2C). It’s like using a special filter to find what’s really important in the prompt. This new method works better than before and helps big language models understand us even when they’re given shorter prompts. |
Keywords
» Artificial intelligence » Cross attention » Decoder » Large language model » Prompt
