vermashresth / awesome-emergent-languages

Paper Reading in Neural Emergent Communication Literature

Emergent Languages

Paper Collection and Summaries

Emergent Languages is a very interesting research area to study how languages emerge artificially, what properties they showcase and whether the functional nature of language and communication can be used to improve upon otherwise supervised learning setting for natural language tasks. This area has strong connections with Computational Linguisics, Artificial Intelligence, Cognitive Science and Behaviourial Theory.

I have tried to include as many papers as possible which study emergent languagees through deep learning (language as in discrete messeages, and not mere continious communication). Most of these works explore emergent languages in referential game setting.

Overview

• Complete List
• Novel Environments or Training
• Exploring Linguistic Properties
  • Communication Effectiveness
  • Compositionality
  • Language Drift
  • Misc
• Emergent Language in Populations
• Priors and Biases in Emergent Language
• Shortcomings of Emergent Language
• Applications of Emergent Language
• Frameworks

Complete List

Novel Environments or Training

• Multi-Agent Cooperation and the Emergence of (Natural) Language | ICLR 2017

  • Angeliki Lazaridou, Alexander Peysakhovich, Marco Baroni | Paper

• Emergence of Language with Multi-agent Games: Learning to Communicate with Sequences of Symbols | NIPS 2017

  • Serhii Havrylov, Ivan Titov | paper

• Emergence of Linguistic Communication from Referential Games with Symbolic and Pixel Input | ICLR 2018

  • Angeliki Lazaridou, Karl Moritz Hermann, Karl Tuyls, Stephen Clark | paper

• Emergent Communication in a Multi-Modal, Multi-Step Referential Game | ICLR 2018

  • Katrina Evtimova, Andrew Drozdov, Douwe Kiela, Kyunghyun Cho | paper

• Emergent Communication through Negotiation | ICLR 2018

  • Kris Cao, Angeliki Lazaridou, Marc Lanctot, Joel Z Leibo, Karl Tuyls, Stephen Clark | | paper

• Emergent Linguistic Phenomena in Multi-Agent Communication Games | EMNLP 2019

  • Laura Graesser, Kyunghyun Cho, Douwe Kiela | paper

• Miss Tools and Mr Fruit: Emergent communication in agents learning about object affordances | ACL 2019

  • Diane Bouchacourt, Marco Baroni | paper

• Emergence of Writing Systems Through Multi-Agent Cooperation (Student Abstract) | AAAI 2020

  • Shresth Verma, Joydip Dhar | paper

Exploring Linguistic Properties

Communication Effectiveness

• On the Pitfalls of Measuring Emergent Communication | AAMAS 2019

  • Ryan Lowe, Jakob Foerster, Y-Lan Boureau, Joelle Pineau, Yann Dauphin | paper

• Social Influence as Intrinsic Motivation for Multi-Agent Deep Reinforcement Learning | ICML 2019

  • Natasha Jaques, Angeliki Lazaridou, Edward Hughes, Caglar Gulcehre, Pedro A. Ortega, DJ Strouse, Joel Z. Leibo, Nando de Freitas | paper

Compositionality

• Compositional Obverter Communication Learning From Raw Visual Input | ICLR 2018

  • Edward Choi, Angeliki Lazaridou, Nando de Freitas | paper

• Measuring Compositionality in Representation Learning | ICLR 2019

  • Jacob Andreas | paper

• Compositionality and Generalization in Emergent Languages | ACL 2020

  • Rahma Chaabouni, Eugene Kharitonov, Diane Bouchacourt, Emmanuel Dupoux, Marco Baroni | paper

• Emergence of Grounded Compositional Language in Multi-Agent Populations | AAAI 2018

  • Igor Mordatch, Pieter Abbeel | paper

• Compositional Languages Emerge in a Neural Iterated Learning Model | ICLR 2020

  • Yi Ren, Shangmin Guo, Matthieu Labeau, Shay B. Cohen, Simon Kirby | paper

• Natural Language Does Not Emerge 'Naturally' in Multi-Agent Dialog | EMNPLP 2017

  • Satwik Kottur, José Moura, Stefan Lee, Dhruv Batra | paper

• Capacity, Bandwidth, and Compositionality in Emergent Language Learning | AAMAS 2020

  • Cinjon Resnick, Abhinav Gupta, Jakob Foerster, Andrew M. Dai, Kyunghyun Cho | paper

• Ease-of-Teaching and Language Structure from Emergent Communication | NeurIPS 2019

  • Fushan Li, Michael Bowling | paper

• Emergence of Language with Multi-agent Games: Learning to Communicate with Sequences of Symbols | NIPS 2017

  • Serhii Havrylov, Ivan Titov | paper

• Emergence of Linguistic Communication from Referential Games with Symbolic and Pixel Input | ICLR 2018

  • Angeliki Lazaridou, Karl Moritz Hermann, Karl Tuyls, Stephen Clark | paper

Language Drift

• Countering Language Drift via Visual Grounding | EMNLP 2019

  • Jason Lee, Kyunghyun Cho, Douwe Kiela | paper

• Countering Language Drift with Seeded Iterated Learning | ICML 2020

  • Yuchen Lu, Soumye Singhal, Florian Strub, Olivier Pietquin, Aaron Courville | paper

• Multi-agent Communication meets Natural Language: Synergies between Functional and Structural Language Learning | ACL 2020

  • Angeliki Lazaridou, Anna Potapenko, Olivier Tieleman | paper

Misc

• Emergence of Communication in an Interactive World with Consistent Speakers | NeurIPS 2018 Emergent Language Workshop

  • Ben Bogin, Mor Geva, Jonathan Berant | paper

• Natural Language Does Not Emerge 'Naturally' in Multi-Agent Dialog | EMNPLP 2017

  • Satwik Kottur, José Moura, Stefan Lee, Dhruv Batra | paper

• How agents see things: On visual representations in an emergent language game | EMNLP 2018

  • Diane Bouchacourt, Marco Baroni | paper

• Entropy Minimization In Emergent Languages | ICML 2020

  • Eugene Kharitonov, Rahma Chaabouni, Diane Bouchacourt, Marco Baroni | paper

• Anti-efficient encoding in emergent communication | NeurIPS 2019

  • Rahma Chaabouni, Eugene Kharitonov, Emmanuel Dupoux, Marco Baroni | paper

Emergent Language in Populations

• To Populate is To Regulate | NeurIPS 2019 Emergent Communication Workshop

  • Nicole Fitzgerald | paper

• Shaping representations through communication: community size effect in artificial learning systems | NeurIPS 2019 workshop on visually grounded interaction and language

  • Olivier Tieleman, Angeliki Lazaridou, Shibl Mourad, Charles Blundell, Doina Precup | paper

• Ease-of-Teaching and Language Structure from Emergent Communication | NeurIPS 2019

  • Fushan Li, Michael Bowling | paper

• Compositional Languages Emerge in a Neural Iterated Learning Model | ICLR 2020

  • Yi Ren, Shangmin Guo, Matthieu Labeau, Shay B. Cohen, Simon Kirby | paper

• Emergence of Grounded Compositional Language in Multi-Agent Populations | AAAI 2018

  • Igor Mordatch, Pieter Abbeel | paper

• Emergent Linguistic Phenomena in Multi-Agent Communication Games | EMNLP 2019

  • Laura Graesser, Kyunghyun Cho, Douwe Kiela | paper

Priors and Biases in Emergent Language

• Biases for Emergent Communication in Multi-agent Reinforcement Learning | NeurIPS 2019

  • Tom Eccles, Yoram Bachrach, Guy Lever, Angeliki Lazaridou, Thore Graepel | paper

• Word-order Biases in Deep-agent Emergent Communication | ACL 2019

  • Rahma Chaabouni, Eugene Kharitonov, Alessandro Lazaric, Emmanuel Dupoux, Marco Baroni | paper

Shortcomings of Emergent Language

• On the Pitfalls of Measuring Emergent Communication | AAMAS 2019

  • Ryan Lowe, Jakob Foerster, Y-Lan Boureau, Joelle Pineau, Yann Dauphin | paper(https://arxiv.org/abs/1810.08647)

• How agents see things: On visual representations in an emergent language game | EMNLP 2018

  • Diane Bouchacourt, Marco Baroni | paper

• Entropy Minimization In Emergent Languages | ICML 2020

  • Eugene Kharitonov, Rahma Chaabouni, Diane Bouchacourt, Marco Baroni | paper

• Anti-efficient encoding in emergent communication | NeurIPS 2019

  • Rahma Chaabouni, Eugene Kharitonov, Emmanuel Dupoux, Marco Baroni | paper

Applications of Emergent Language

• Multi-agent Communication meets Natural Language: Synergies between Functional and Structural Language Learning | ACL 2020

  • Angeliki Lazaridou, Anna Potapenko, Olivier Tieleman | paper

Frameworks

Complete List

• Multi-Agent Cooperation and the Emergence of (Natural) Language | ICLR 2017

  • Angeliki Lazaridou, Alexander Peysakhovich, Marco Baroni | Paper

• Emergence of Language with Multi-agent Games: Learning to Communicate with Sequences of Symbols | NIPS 2017

  • Serhii Havrylov, Ivan Titov | paper

• Emergence of Linguistic Communication from Referential Games with Symbolic and Pixel Input | ICLR 2018

  • Angeliki Lazaridou, Karl Moritz Hermann, Karl Tuyls, Stephen Clark | paper

• Emergent Communication in a Multi-Modal, Multi-Step Referential Game | ICLR 2018

  • Katrina Evtimova, Andrew Drozdov, Douwe Kiela, Kyunghyun Cho | paper

• Emergent Communication through Negotiation | ICLR 2018

  • Kris Cao, Angeliki Lazaridou, Marc Lanctot, Joel Z Leibo, Karl Tuyls, Stephen Clark | | paper

• Emergent Linguistic Phenomena in Multi-Agent Communication Games | EMNLP 2019

  • Laura Graesser, Kyunghyun Cho, Douwe Kiela | paper

• Miss Tools and Mr Fruit: Emergent communication in agents learning about object affordances | ACL 2019

  • Diane Bouchacourt, Marco Baroni | paper

• Emergence of Writing Systems Through Multi-Agent Cooperation (Student Abstract) | AAAI 2020

  • Shresth Verma, Joydip Dhar | paper

• Emergence of Communication in an Interactive World with Consistent Speakers | NeurIPS 2018 Emergent Language Workshop

  • Ben Bogin, Mor Geva, Jonathan Berant | paper

• Natural Language Does Not Emerge 'Naturally' in Multi-Agent Dialog | EMNPLP 2017

  • Satwik Kottur, José Moura, Stefan Lee, Dhruv Batra | paper

• Capacity, Bandwidth, and Compositionality in Emergent Language Learning | AAMAS 2020

  • Cinjon Resnick, Abhinav Gupta, Jakob Foerster, Andrew M. Dai, Kyunghyun Cho | paper

• How agents see things: On visual representations in an emergent language game | EMNLP 2018

  • Diane Bouchacourt, Marco Baroni | paper

• Entropy Minimization In Emergent Languages | ICML 2020

  • Eugene Kharitonov, Rahma Chaabouni, Diane Bouchacourt, Marco Baroni | paper

• Anti-efficient encoding in emergent communication | NeurIPS 2019

  • Rahma Chaabouni, Eugene Kharitonov, Emmanuel Dupoux, Marco Baroni | paper

• On the Pitfalls of Measuring Emergent Communication | AAMAS 2019

  • Ryan Lowe, Jakob Foerster, Y-Lan Boureau, Joelle Pineau, Yann Dauphin | paper

• Social Influence as Intrinsic Motivation for Multi-Agent Deep Reinforcement Learning | ICML 2019

  • Natasha Jaques, Angeliki Lazaridou, Edward Hughes, Caglar Gulcehre, Pedro A. Ortega, DJ Strouse, Joel Z. Leibo, Nando de Freitas | paper

• Compositional Obverter Communication Learning From Raw Visual Input | ICLR 2018

  • Edward Choi, Angeliki Lazaridou, Nando de Freitas | paper

• Measuring Compositionality in Representation Learning | ICLR 2019

  • Jacob Andreas | paper

• Compositionality and Generalization in Emergent Languages | ACL 2020

  • Rahma Chaabouni, Eugene Kharitonov, Diane Bouchacourt, Emmanuel Dupoux, Marco Baroni | paper

• Emergence of Grounded Compositional Language in Multi-Agent Populations | AAAI 2018

  • Igor Mordatch, Pieter Abbeel | paper

• Ease-of-Teaching and Language Structure from Emergent Communication | NeurIPS 2019

  • Fushan Li, Michael Bowling | paper

• Compositional Languages Emerge in a Neural Iterated Learning Model | ICLR 2020

  • Yi Ren, Shangmin Guo, Matthieu Labeau, Shay B. Cohen, Simon Kirby | paper

• Shaping representations through communication: community size effect in artificial learning systems | NeurIPS 2019 workshop on visually grounded interaction and language

  • Olivier Tieleman, Angeliki Lazaridou, Shibl Mourad, Charles Blundell, Doina Precup | paper

• Biases for Emergent Communication in Multi-agent Reinforcement Learning | NeurIPS 2019

  • Tom Eccles, Yoram Bachrach, Guy Lever, Angeliki Lazaridou, Thore Graepel | paper

• Word-order Biases in Deep-agent Emergent Communication | ACL 2019

  • Rahma Chaabouni, Eugene Kharitonov, Alessandro Lazaric, Emmanuel Dupoux, Marco Baroni | paper

• Countering Language Drift via Visual Grounding | EMNLP 2019

  • Jason Lee, Kyunghyun Cho, Douwe Kiela | paper

• Countering Language Drift with Seeded Iterated Learning | ICML 2020

  • Yuchen Lu, Soumye Singhal, Florian Strub, Olivier Pietquin, Aaron Courville | paper

• Multi-agent Communication meets Natural Language: Synergies between Functional and Structural Language Learning | ACL 2020

  • Angeliki Lazaridou, Anna Potapenko, Olivier Tieleman | paper

• On the interaction between supervision and self-play in emergent communication | ICLR 2020

  • Ryan Lowe, Abhinav Gupta, Jakob Foerster, Douwe Kiela, Joelle Pineau | paper

• Learning to learn to communicate | ICML 2019 Workshop on Adaptive and Multitask learning

  • Ryan Lowe, Abhinav Gupta, Jakob Foerster, Douwe Kiela, Joelle Pineau | paper

• To Populate is To Regulate | NeurIPS 2019 Emergent Communication Workshop

  • Nicole Fitzgerald | paper

• Emergent Language Generalization and Acquisition Speed are not tied to Compositionality | Arxiv 2020

• Biology and Compositionality: Empirical Considerations for Emergent-Communication Protocols | NeurIPS emergent Commm workshop 2019

• Few-shot Language Coordination by Modeling Theory of Mind| ICML 2021

  • Hao Zhu, Graham Neubig, Yonatan Bisk | paper

• Interpretable agent communication from scratch (with a generic visual processor emerging on the side) | NeurIPS 2021

  • Roberto_Dessi, Eugene Kharitonov, Marco Baroni | paper

• The Grammar of Emergent Languages | EMNLP 2020

  • Oskar van der Wal, Silvan de Boer, Elia Bruni, Dieuwke Hupkes | paper

• Emergent Communication under Competition | AAMAS 2021

  • Michael Noukhovitch, Travis LaCroix, Angeliki Lazaridou, Aaron Courville | paper

• Emergent Discrete Communication in Semantic Spaces | NeurIPS 2021

  • Mycal Tucker, Huao Li, Siddharth Agrawal, Dana Hughes, Katia P. Sycara, Michael Lewis, Julie Shah | paper

• Learning to Draw: Emergent Communication through Sketching | NeurIPS 2021

  • Daniela Mihai, Jonathon Hare | paper

• Emergent Communication of Generalizations | NeurIPS 2021

  • Jesse Mu, Noah Goodman | paper

• Emergent Communication under Varying Sizes and Connectivities | NeurIPS 2021

  • Jooyeon Kim, Alice Oh | paper

• Emergent communication at scale | ICLR 2022

  • Rahma Chaabouni, Florian Strub, Florent Altché, Eugene Tarassov, Corentin Tallec, Elnaz Davoodi, Kory Wallace Mathewson, Olivier Tieleman, Angeliki Lazaridou, Bilal Piot | paper