Towards Collaborative Plan Acquisition through Theory of Mind Modeling in Situated Dialogue

This is the official code repository for the paper (anthology link). Will update when available.

An extended verion of the paper including supplemental material can be found at (ArXiv).

Dataset

You will need to download the original dataset released with the MindCraft: Theory of Mind Modeling for Situated Dialogue in Collaborative Tasks paper as well as our extension of the original dataset.

If you are having trouble accessing the data and models please contact Cristian-Paul Bara at cpbara@umich.edu

Name		Name	Last commit message	Last commit date
config		config
misc		misc
src		src
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
__init__.py		__init__.py
action_predictor.py		action_predictor.py
baselines_with_dialogue_moves.py		baselines_with_dialogue_moves.py
baselines_with_dialogue_moves.sh		baselines_with_dialogue_moves.sh
dataset_splits.json		dataset_splits.json
dev_baseline.sh		dev_baseline.sh
dev_eval.sh		dev_eval.sh
dev_plan.sh		dev_plan.sh
dialogue_act_classifier.py		dialogue_act_classifier.py
dialogue_act_predictor.py		dialogue_act_predictor.py
dialogue_move_classifier.py		dialogue_move_classifier.py
eval.py		eval.py
intermediate_representations.py		intermediate_representations.py
kill_job.sh		kill_job.sh
monitor.sh		monitor.sh
plan_predictor.py		plan_predictor.py
plan_stab.py		plan_stab.py
plan_stab.sh		plan_stab.sh
print_log.sh		print_log.sh
qwer.py		qwer.py
run_action_predictor.sh		run_action_predictor.sh
run_dialogue.sh		run_dialogue.sh
run_dialogue_act_classifier.sh		run_dialogue_act_classifier.sh
run_dialogue_act_predictor.sh		run_dialogue_act_predictor.sh
run_dialogue_move_classifier.sh		run_dialogue_move_classifier.sh
run_plan_predictor.sh		run_plan_predictor.sh
sandbox.py		sandbox.py