Enregistré dans:
Détails bibliographiques
Auteurs principaux: Chiyah-Garcia, Javier, Suglia, Alessandro, Eshghi, Arash
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:https://arxiv.org/abs/2409.14247
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866929527642390528
author Chiyah-Garcia, Javier
Suglia, Alessandro
Eshghi, Arash
author_facet Chiyah-Garcia, Javier
Suglia, Alessandro
Eshghi, Arash
contents In dialogue, the addressee may initially misunderstand the speaker and respond erroneously, often prompting the speaker to correct the misunderstanding in the next turn with a Third Position Repair (TPR). The ability to process and respond appropriately to such repair sequences is thus crucial in conversational AI systems. In this paper, we first collect, analyse, and publicly release BlockWorld-Repairs: a dataset of multi-modal TPR sequences in an instruction-following manipulation task that is, by design, rife with referential ambiguity. We employ this dataset to evaluate several state-of-the-art Vision and Language Models (VLM) across multiple settings, focusing on their capability to process and accurately respond to TPRs and thus recover from miscommunication. We find that, compared to humans, all models significantly underperform in this task. We then show that VLMs can benefit from specialised losses targeting relevant tokens during fine-tuning, achieving better performance and generalising better to new scenarios. Our results suggest that these models are not yet ready to be deployed in multi-modal collaborative settings where repairs are common, and highlight the need to design training regimes and objectives that facilitate learning from interaction. Our code and data are available at www.github.com/JChiyah/blockworld-repairs
format Preprint
id arxiv_https___arxiv_org_abs_2409_14247
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Repairs in a Block World: A New Benchmark for Handling User Corrections with Multi-Modal Language Models
Chiyah-Garcia, Javier
Suglia, Alessandro
Eshghi, Arash
Computation and Language
Human-Computer Interaction
In dialogue, the addressee may initially misunderstand the speaker and respond erroneously, often prompting the speaker to correct the misunderstanding in the next turn with a Third Position Repair (TPR). The ability to process and respond appropriately to such repair sequences is thus crucial in conversational AI systems. In this paper, we first collect, analyse, and publicly release BlockWorld-Repairs: a dataset of multi-modal TPR sequences in an instruction-following manipulation task that is, by design, rife with referential ambiguity. We employ this dataset to evaluate several state-of-the-art Vision and Language Models (VLM) across multiple settings, focusing on their capability to process and accurately respond to TPRs and thus recover from miscommunication. We find that, compared to humans, all models significantly underperform in this task. We then show that VLMs can benefit from specialised losses targeting relevant tokens during fine-tuning, achieving better performance and generalising better to new scenarios. Our results suggest that these models are not yet ready to be deployed in multi-modal collaborative settings where repairs are common, and highlight the need to design training regimes and objectives that facilitate learning from interaction. Our code and data are available at www.github.com/JChiyah/blockworld-repairs
title Repairs in a Block World: A New Benchmark for Handling User Corrections with Multi-Modal Language Models
topic Computation and Language
Human-Computer Interaction
url https://arxiv.org/abs/2409.14247