Empirical study on BlenderBot 2.0’s Errors Analysis in terms of Model, Data and User-Centric Approach

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Language Models that Seek for Knowledge: Modular Search & Generation for Dialogue and Prompt Completion (2022, background): “They, as well as others (Lee et al., 2022), report that their method can have the problems of either mixing up facts together incorrectly or generating a generic response that ignores the knowledge, which our method attempts to address.”