Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Towards Harnessing Large Language Models for Comprehension of Conversational Grounding

Kristiina Jokinen, Phillip Schneider, Taiga Mori

TL;DR

The paper investigates how large language models comprehend conversational grounding by classifying turns as explicit or implicit grounding and predicting grounded knowledge elements in information seeking dialogues. It evaluates LLM capabilities on these tasks, identifying challenges and failure modes. It discusses pipeline architectures and knowledge bases as pathways to improve grounding comprehension, including retrieval-augmented approaches. The work highlights limitations and provides guidance for building more effective grounding-aware dialogue systems with practical implications for deployment in knowledge-rich conversations.

Abstract

Conversational grounding is a collaborative mechanism for establishing mutual knowledge among participants engaged in a dialogue. This experimental study analyzes information-seeking conversations to investigate the capabilities of large language models in classifying dialogue turns related to explicit or implicit grounding and predicting grounded knowledge elements. Our experimental results reveal challenges encountered by large language models in the two tasks and discuss ongoing research efforts to enhance large language model-based conversational grounding comprehension through pipeline architectures and knowledge bases. These initiatives aim to develop more effective dialogue systems that are better equipped to handle the intricacies of grounded knowledge in conversations.

Towards Harnessing Large Language Models for Comprehension of Conversational Grounding

TL;DR

The paper investigates how large language models comprehend conversational grounding by classifying turns as explicit or implicit grounding and predicting grounded knowledge elements in information seeking dialogues. It evaluates LLM capabilities on these tasks, identifying challenges and failure modes. It discusses pipeline architectures and knowledge bases as pathways to improve grounding comprehension, including retrieval-augmented approaches. The work highlights limitations and provides guidance for building more effective grounding-aware dialogue systems with practical implications for deployment in knowledge-rich conversations.

Abstract

Conversational grounding is a collaborative mechanism for establishing mutual knowledge among participants engaged in a dialogue. This experimental study analyzes information-seeking conversations to investigate the capabilities of large language models in classifying dialogue turns related to explicit or implicit grounding and predicting grounded knowledge elements. Our experimental results reveal challenges encountered by large language models in the two tasks and discuss ongoing research efforts to enhance large language model-based conversational grounding comprehension through pipeline architectures and knowledge bases. These initiatives aim to develop more effective dialogue systems that are better equipped to handle the intricacies of grounded knowledge in conversations.
Paper Structure (15 sections, 2 theorems, 4 equations, 3 figures, 2 tables)

This paper contains 15 sections, 2 theorems, 4 equations, 3 figures, 2 tables.

Table of Contents

  1. Part Title
  2. Section Heading
  3. Section Heading
  4. Subsection Heading
  5. Subsubsection Heading
  6. Paragraph Heading
  7. Subparagraph Heading
  8. Section Heading
  9. Subsection Heading
  10. Subsubsection Heading
  11. Paragraph Heading
  12. Chapter Heading
  13. Section Heading
  14. Subsection Heading
  15. Subsubsection Heading

Key Result

theorem 1

Theorem text goes here.

Figures (3)

  • Figure 1: If the width of the figure is less than 7.8 cm use the sidecapion command to flush the caption on the left side of the page. If the figure is positioned at the top of the page, align the sidecaption with the top of the figure -- to achieve this you simply need to use the optional argument [t] with the sidecaption command
  • Figure 2: If the width of the figure is less than 7.8 cm use the sidecapion command to flush the caption on the left side of the page. If the figure is positioned at the top of the page, align the sidecaption with the top of the figure -- to achieve this you simply need to use the optional argument [t] with the sidecaption command
  • Figure 3: Please write your figure caption here

Theorems & Definitions (6)

  • theorem 1
  • definition 1
  • proof
  • theorem 2
  • definition 2
  • proof