Summary of Instructcmp: Length Control in Sentence Compression Through Instruction-based Large Language Models, by Juseon-do et al.
InstructCMP: Length Control in Sentence Compression through Instruction-based Large Language Models
by Juseon-Do, Jingun Kwon, Hidetaka Kamigaito, Manabu Okumura
First submitted to arxiv on: 16 Jun 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 This paper proposes Instruction-based Compression (InstructCMP), an approach to sentence compression that considers constraints like desired summary lengths. By leveraging Large Language Models’ (LLMs) zero-shot task-solving abilities, InstructCMP can generate faithful summaries while adhering to length limits. The authors created new evaluation datasets by transforming traditional sentence compression datasets into instruction format and tested the effectiveness of their proposed “length priming” approach, which incorporates additional length information into instructions without external resources. Experimental results demonstrate that applying length priming significantly improves InstructCMP’s performance in both zero-shot and fine-tuning settings without requiring model modifications. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary In this paper, scientists developed a new way to summarize texts called Instruction-based Compression (InstructCMP). It helps machines generate accurate summaries while following specific rules. The researchers created special datasets for testing their approach and found that it works well in making summaries that are the right length. They also showed that fine-tuning their model with more training data makes it even better at controlling summary lengths. |
Keywords
» Artificial intelligence » Fine tuning » Zero shot
