Summary of A Reliable Common-sense Reasoning Socialbot Built Using Llms and Goal-directed Asp, by Yankai Zeng et al.
A Reliable Common-Sense Reasoning Socialbot Built Using LLMs and Goal-Directed ASP
by Yankai Zeng, Abhiramon Rajashekharan, Kinjal Basu, Huaduo Wang, Joaquín Arias, Gopal Gupta
First submitted to arxiv on: 26 Jul 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Logic in Computer Science (cs.LO)
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 development of large language models like GPT has led to the creation of socialbots that can simulate human-like conversations. However, these models lack goal-directedness and are hard to control. They often provide confusing answers and struggle to integrate multiple topics into a coherent response, causing them to deviate from the main topic. To address this, we propose AutoCompanion, a socialbot that uses an LLM model to translate natural language into predicates and employs commonsense reasoning based on Answer Set Programming (ASP) to hold a conversation with humans. Our framework relies on s(CASP), a goal-directed implementation of ASP as the backend. This paper presents the design of our framework and how an LLM is used to parse user messages and generate responses from the s(CASP) engine output. We validate our proposal by describing real conversations where AutoCompanion’s goal is to keep users entertained, ensuring correct answers, coherence, and no deviation from the main topic. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Imagine a chatbot that can have a conversation with you like a human would. But right now, these chatbots aren’t very good at staying on topic or giving accurate answers. They’re more like pattern recognizers than thinkers. We want to make a better chatbot that can have a fun and coherent conversation with you. Our idea is called AutoCompanion. It uses a special language model to translate what you say into something the computer can understand, and then it uses commonsense reasoning to come up with answers. This means it won’t just give confusing answers, but will also keep the conversation on track and make sure it’s interesting. |
Keywords
» Artificial intelligence » Gpt » Language model
