Summary of Social Orientation: a New Feature For Dialogue Analysis, by Todd Morrill et al.
Social Orientation: A New Feature for Dialogue Analysis
by Todd Morrill, Zhaoyuan Deng, Yanda Chen, Amith Ananthram, Colin Wayne Leach, Kathleen McKeown
First submitted to arxiv on: 26 Feb 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 Our paper introduces a novel approach to predicting the success or failure of dialogues by leveraging social orientations. We draw from Circumplex theory in psychology, which categorizes conversation participants into different social orientations (e.g., Warm-Agreeable, Arrogant-Calculating). To apply this framework to dialogue modeling, we created a new dataset of machine-labeled dialogue utterances with social orientation tags. Our results demonstrate that incorporating these tags improves task performance, especially in low-resource settings, on both English and Chinese language benchmarks. Moreover, our findings show that social orientation tags can be used to explain the outcomes of social interactions when integrated into neural models. To facilitate further research, we release our datasets, code, and fine-tuned models for predicting social orientation tags on dialogue utterances. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Imagine trying to guess how a conversation will go based on who’s talking. Our paper helps make this kind of prediction by using special labels that describe the personalities of people in conversations. We created a new set of labeled conversations that shows these labels can improve our accuracy at predicting the success or failure of dialogues, even when we don’t have much data to work with. We also found that these labels can help us understand why some conversations go well and others don’t. To make it easier for other researchers to use our approach, we’re releasing our labeled conversations, computer code, and trained models. |
