Summary of Concept — An Evaluation Protocol on Conversational Recommender Systems with System-centric and User-centric Factors, by Chen Huang et al.
Concept – An Evaluation Protocol on Conversational Recommender Systems with System-centric and User-centric Factors
by Chen Huang, Peixin Qin, Yang Deng, Wenqiang Lei, Jiancheng Lv, Tat-Seng Chua
First submitted to arxiv on: 4 Apr 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 Concept evaluation protocol for conversational recommendation systems (CRS) addresses the limitation of existing protocols by incorporating both system-centric and user-centric factors. The protocol is designed to assess three key characteristics: understanding, consistency, and coherence, which are further divided into six primary abilities. A LLM-based user simulator and evaluator with tailored scoring rubrics are used to implement Concept. This protocol serves as a reference guide for evaluating CRS, providing an overview of the strengths and weaknesses of current models, while also highlighting the issue of low usability in popular chatbots like ChatGPT. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A new way to test how well conversational AI systems work is being developed. The existing methods only look at whether these systems are good at having conversations or not. But this new approach considers how users feel when they use these systems, too. It’s called Concept and it has three main parts: understanding, consistency, and coherence. These parts are further broken down into six smaller parts that test different skills. To make sure this works, a special computer program is being used to simulate human conversations with the AI system and rate its performance. This new way of testing will help us understand what’s good or bad about these systems and how we can make them better. |
