theory-of-mind

AI tutor powered by Theory-of-Mind reasoning

An awesome & curated list for Artificial General Intelligence, an emerging inter-discipline field that combines artificial intelligence and computational cognitive sciences.

[🏆Outstanding Paper Award at ACL 2024] MMToM-QA: Multimodal Theory of Mind Question Answering

[ACL 2024] Exploring Collaboration Mechanisms for LLM Agents: A Social Psychology View

The offcial implementation of "ToM2C: Target-oriented Multi-agent Communication and Cooperation with Theory of Mind" (ICLR 2022) .

ToMBench: Benchmarking Theory of Mind in Large Language Models, ACL 2024.

[ICML 2024] Language Models Represent Beliefs of Self and Others

VU-CLTL Pepper/Nao Application Repository (Python 2)

Code for our SIGGRAPH 2023 paper, "Acting as Inverse Inverse Planning"

Testing Theory of Mind (ToM) in language models with epistemic logic

Code for the paper "Symmetric Machine Theory of Mind", presented at ICML 2022.

ToMATO: Verbalizing the Mental States of Role-Playing LLMs for Benchmarking Theory of Mind (AAAI2025)

IMQ (interactive mentalizing questionnaire)

A research project into the underlying mechanisms of consciousness, and an emulation of a calculator that is conscious.

Code used to produce the figures of the academic paper "Risk Sensitivity and Theory of Mind in Human Coordination".

Material for the Moral Dynamics project -- a computational model of moral judgment by Felix Sosa, Tomer Ullman, Josh Tenenbaum, Sam Gershman, and Tobias Gerstenberg. A standing manuscript can be found here: www.mit.edu/~fsosa/papers/md.pdf

Selfplay is a Python framework designed for simulating role-based dialogues using chatbot interactions. It provides a flexible environment for creating and testing conversational scenarios, making it ideal for research, training, and interactive applications. With built-in templates for various role-play situations, Selfplay helps automate dialogue

🐵 Recreation of Theory of Mind experiment done on chimpanzees as an RL environment

Un cadre formel et axiomatisé pour décrire au mieux l'univers et servir de base à l'élaboration de théories computationelles du tout, dont la construction est libre, ouverte et contributive.

Use Spiking Neural Networks to solve Cart Pole control problem with a solution inspired by ToM. This code is part of my Master’s Thesis code.