Grounding Gaps in Language Model Generations

• URL: http://arxiv.org/abs/2311.09144v2
• Date: Tue, 2 Apr 2024 19:34:15 GMT
• Title: Grounding Gaps in Language Model Generations
• Authors: Omar Shaikh, Kristina Gligorić, Ashna Khetan, Matthias Gerstgrasser, Diyi Yang, Dan Jurafsky,
• Abstract summary: We study whether large language models generate text that reflects human grounding. We find that -- compared to humans -- LLMs generate language with less conversational grounding. To understand the roots of the identified grounding gap, we examine the role of instruction tuning and preference optimization.
• Score: 67.79817087930678
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Effective conversation requires common ground: a shared understanding between the participants. Common ground, however, does not emerge spontaneously in conversation. Speakers and listeners work together to both identify and construct a shared basis while avoiding misunderstanding. To accomplish grounding, humans rely on a range of dialogue acts, like clarification (What do you mean?) and acknowledgment (I understand.). However, it is unclear whether large language models (LLMs) generate text that reflects human grounding. To this end, we curate a set of grounding acts and propose corresponding metrics that quantify attempted grounding. We study whether LLM generations contain grounding acts, simulating turn-taking from several dialogue datasets and comparing results to humans. We find that -- compared to humans -- LLMs generate language with less conversational grounding, instead generating text that appears to simply presume common ground. To understand the roots of the identified grounding gap, we examine the role of instruction tuning and preference optimization, finding that training on contemporary preference data leads to a reduction in generated grounding acts. Altogether, we highlight the need for more research investigating conversational grounding in human-AI interaction.

Related papers

• Empathic Grounding: Explorations using Multimodal Interaction and Large Language Models with Conversational Agents [0.6990493129893112]
  Empathic grounding is required whenever the speaker's emotions are foregrounded. We describe a model that takes as input user speech and facial expression to generate multimodal grounding moves for a listening agent. Our work highlights the role of emotion awareness and multimodality in generating appropriate grounding moves for conversational agents.
  arXiv  Detail & Related papers  (2024-07-01T21:46:30Z)
• It Couldn't Help But Overhear: On the Limits of Modelling Meta-Communicative Grounding Acts with Supervised Learning [19.812562421377706]
  Overhearers are deprived of the privilege of performing grounding acts and can only conjecture about intended meanings. We show that there is evidence pointing to the impossibility of properly modelling human meta-communicative acts with data-driven learning models. Most importantly, we wish to bring this topic back to the community's table, encouraging discussion on the consequences of having models designed to only "listen in"
  arXiv  Detail & Related papers  (2024-05-02T09:55:19Z)
• Conversational Grounding: Annotation and Analysis of Grounding Acts and Grounding Units [3.805394793605586]
  We present the annotation of two dialog corpora employing Grounding Acts, Grounding Units, and a measure of their degree of grounding. Our work aims to make conversations with machines better understood and more reliable in natural day-to-day collaborative dialogs.
  arXiv  Detail & Related papers  (2024-03-25T10:39:18Z)
• A Glitch in the Matrix? Locating and Detecting Language Model Grounding with Fakepedia [57.31074448586854]
  Large language models (LLMs) have an impressive ability to draw on novel information supplied in their context. Yet the mechanisms underlying this contextual grounding remain unknown. We present a novel method to study grounding abilities using Fakepedia.
  arXiv  Detail & Related papers  (2023-12-04T17:35:42Z)
• How Well Do Large Language Models Truly Ground? [39.39062385290276]
  A common method is to generate responses by grounding on external contexts given as input, known as knowledge-augmented models. Previous research often narrowly defines "grounding" as just having the correct answer, which does not ensure the reliability of the entire response. We propose a stricter definition of grounding: a model is truly grounded if it (1) fully utilizes the necessary knowledge from the provided context, and (2) stays within the limits of that knowledge.
  arXiv  Detail & Related papers  (2023-11-15T16:11:27Z)
• The Vector Grounding Problem [0.0]
  We argue that referential grounding is the one that lies at the heart of the Vector Grounding Problem. We also argue that, perhaps unexpectedly, multimodality and embodiment are neither necessary nor sufficient conditions for referential grounding in artificial systems.
  arXiv  Detail & Related papers  (2023-04-04T02:54:04Z)
• KPT: Keyword-guided Pre-training for Grounded Dialog Generation [82.68787152707455]
  We propose KPT (guided Pre-Training), a novel self-supervised pre-training method for grounded dialog generation. Specifically, we use a pre-trained language model to extract the most uncertain tokens in the dialog as keywords. We conduct extensive experiments on various few-shot knowledge-grounded generation tasks, including grounding on dialog acts, knowledge graphs, persona descriptions, and Wikipedia passages.
  arXiv  Detail & Related papers  (2022-12-04T04:05:01Z)
• Towards Language Modelling in the Speech Domain Using Sub-word Linguistic Units [56.52704348773307]
  We propose a novel LSTM-based generative speech LM based on linguistic units including syllables and phonemes. With a limited dataset, orders of magnitude smaller than that required by contemporary generative models, our model closely approximates babbling speech. We show the effect of training with auxiliary text LMs, multitask learning objectives, and auxiliary articulatory features.
  arXiv  Detail & Related papers  (2021-10-31T22:48:30Z)
• Grounding 'Grounding' in NLP [59.28887479119075]
  As a community, we use the term broadly to reference any linking of text to data or non-textual modality. Cognitive Science more formally defines "grounding" as the process of establishing what mutual information is required for successful communication.
  arXiv  Detail & Related papers  (2021-06-04T00:40:59Z)
• Provable Limitations of Acquiring Meaning from Ungrounded Form: What will Future Language Models Understand? [87.20342701232869]
  We investigate the abilities of ungrounded systems to acquire meaning. We study whether assertions enable a system to emulate representations preserving semantic relations like equivalence. We find that assertions enable semantic emulation if all expressions in the language are referentially transparent. However, if the language uses non-transparent patterns like variable binding, we show that emulation can become an uncomputable problem.
  arXiv  Detail & Related papers  (2021-04-22T01:00:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.