Quark: Controllable text generation with reinforced unlearning

X Lu, S Welleck, J Hessel, L Jiang… - Advances in neural …, 2022 - proceedings.neurips.cc
Large-scale language models often learn behaviors that are misaligned with user
expectations. Generated text may contain offensive or toxic language, contain significant …

A large-scale dataset for empathetic response generation

A Welivita, Y **e, P Pu - Proceedings of the 2021 Conference on …, 2021 - aclanthology.org
Recent development in NLP shows a strong trend towards refining pre-trained models with a
domain-specific dataset. This is especially the case for response generation where emotion …

Immersive experience with non-player characters dynamic dialogue

MF Hasani, Y Udjaja - 2021 1st International Conference on …, 2021 - ieeexplore.ieee.org
Non-Player Character (NPC) is one of the important elements in the game, because NPCs
can liven up the atmosphere in the game by means of intense interaction with players with …

Use of a taxonomy of empathetic response intents to control and interpret empathy in neural chatbots

A Welivita, P Pu - arxiv preprint arxiv:2305.10096, 2023 - arxiv.org
A recent trend in the domain of open-domain conversational agents is enabling them to
converse empathetically to emotional prompts. Current approaches either follow an end-to …

Empathetic Conversational Agents for Distress Support

KA Welivita - 2023 - infoscience.epfl.ch
Due to the increasing demands of today's fast-paced world, mental health concerns are on
the rise, which necessitates innovative approaches to provide support to those in need …

Towards Empathetic Open-Domain Chatbots

Y **e - 2022 - infoscience.epfl.ch
The research community of dialog generation has been interested in incorporating
emotional information into the design of open-domain dialog systems ever since neural …