Singh, Aaditya K;
Ding, David;
Saxe, Andrew;
Hill, Felix;
Lampinen, Andrew Kyle;
(2023)
Know your audience: specializing grounded language models with listener subtraction.
In:
EACL 2023 - 17th Conference of the European Chapter of the Association for Computational Linguistics, Proceedings of the Conference.
(pp. pp. 3884-3911).
Association for Computational Linguistics: Dubrovnik, Croatia.
Preview |
PDF
2023.eacl-main.279.pdf - Published Version Download (9MB) | Preview |
Abstract
Effective communication requires adapting to the idiosyncrasies of each communicative context—such as the common ground shared with each partner. Humans demonstrate this ability to specialize to their audience in many contexts, such as the popular game Dixit. We take inspiration from Dixit to formulate a multiagent image reference game where a (trained) speaker model is rewarded for describing a target image such that one (pretrained) listener model can correctly identify it among distractors, but another listener cannot. To adapt, the speaker must exploit differences in the knowledge it shares with the different listeners. We show that finetuning an attention-based adapter between a CLIP vision encoder and a large language model in this contrastive, multi-agent setting gives rise to context-dependent natural language specialization from rewards only, without direct supervision. Through controlled experiments, we show that training a speaker with two listeners that perceive differently, using our method, allows the speaker to adapt to the idiosyncracies of the listeners. Furthermore, we show zero-shot transfer of the specialization to real-world data. Our experiments demonstrate a method for specializing grounded language models without direct supervision and highlight the interesting research challenges posed by complex multi-agent communication
| Type: | Proceedings paper |
|---|---|
| Title: | Know your audience: specializing grounded language models with listener subtraction |
| Event: | EACL 2023 - 17th Conference of the European Chapter of the Association for Computational Linguistics |
| ISBN-13: | 9781959429449 |
| Open access status: | An open access version is available from UCL Discovery |
| Publisher version: | https://aclanthology.org/2023.eacl-main.279/ |
| Language: | English |
| Additional information: | © The Author(s). This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. https://creativecommons.org/licenses/by/4.0/ |
| UCL classification: | UCL UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Life Sciences UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Life Sciences > Gatsby Computational Neurosci Unit |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10171251 |
Archive Staff Only
 |
View Item |
