Summary of Autoregressive Multi-trait Essay Scoring Via Reinforcement Learning with Scoring-aware Multiple Rewards, by Heejin Do et al.
Autoregressive Multi-trait Essay Scoring via Reinforcement Learning with Scoring-aware Multiple Rewards
by Heejin Do, Sangwon Ryu, Gary Geunbae Lee
First submitted to arxiv on: 26 Sep 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 Scoring-aware Multi-reward Reinforcement Learning (SaMRL) method integrates actual evaluation schemes into the training process by designing QWK-based rewards with a mean-squared error penalty for multi-trait AES. This approach leverages token generation probabilities for robust multi-trait score predictions, enhancing scoring of previously inferior prompts. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper proposes a new method called SaMRL that helps train neural networks to score essays better. The old way of training these networks didn’t work well when trying to score multiple traits in an essay, like grammar and content. SaMRL changes this by using real evaluation schemes as rewards during training, which makes the network learn to score more accurately. |
Keywords
» Artificial intelligence » Reinforcement learning » Token
