Summary of Reward-based Input Construction For Cross-document Relation Extraction, by Byeonghu Na et al.
Reward-based Input Construction for Cross-document Relation Extraction
by Byeonghu Na, Suhyeon Jo, Yeongmin Kim, Il-Chul Moon
First submitted to arxiv on: 31 May 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Machine Learning (cs.LG)
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 Relation extraction (RE) is a crucial task in natural language processing, aiming to identify relationships between target entities across text. The paper proposes REward-based Input Construction (REIC), the first learning-based sentence selector for cross-document RE, which extracts sentences based on relational evidence. This allows the RE module to effectively infer relations. To address the lack of supervision for evidence sentences, the authors employ reinforcement learning with RE prediction scores as rewards. Experimental results demonstrate the superiority of REIC over heuristic methods for different RE structures and backbones in cross-document RE. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about a new way to find relationships between things mentioned in long pieces of text. Currently, most relationship-finding methods can only look at one sentence or document at a time. But what if you want to find relationships that span multiple documents? The authors came up with a solution called REIC, which uses clues from the text to pick out important sentences that help it understand relationships. They tested their method and found it worked better than other approaches for finding relationships across many documents. |
Keywords
* Artificial intelligence * Natural language processing * Reinforcement learning
