Papers by Tushar Goswamy

1 papers
Adapting a Language Model for Controlled Affective Text Generation (2020.coling-main)

Copied to clipboard

Challenge: Existing models for affective text generation fail to capture emotional aspects of conversations without explicit affective information.
Approach: They propose to incorporate emotion as prior for the probabilistic state-of-the-art text generation model such as GPT-2 and incorporate emotion into the model to ensure grammatical correctness.
Outcome: The proposed model outperforms existing models in all intensities and is robust to human evaluations.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations