Summary of Fine-grained Behavior Simulation with Role-playing Large Language Model on Social Media, by Kun Li et al.
Fine-Grained Behavior Simulation with Role-Playing Large Language Model on Social Media
by Kun Li, Chenwei Dai, Wei Zhou, Songlin Hu
First submitted to arxiv on: 4 Dec 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
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 A novel fine-grained behavior simulation dataset called FineRob is introduced, consisting of 78.6k QA records from the behavioral history of 1,866 distinct users across three social media platforms. This dataset aims to improve large language models’ (LLMs) capability to simulate user behavior in real-world scenarios. Two dominant reasoning patterns are identified in LLMs’ behavior simulation processes, and a fine-tuning method called OM-CoT is proposed to enhance the capability. Comprehensive experiments demonstrate the effectiveness of this approach. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Large language models can do many things, like chat with us or play games. But they don’t always understand how people really behave online. This paper tries to fix that by creating a big dataset of real people’s behavior on social media. They took data from 1,866 users across three platforms and broke it down into smaller pieces. Then, they tested if these models can really simulate how people act online. They found two main ways the models work and came up with a new way to make them better at this. |
Keywords
» Artificial intelligence » Fine tuning
