# Multi-task Learning of Negation and Speculation for Targeted Sentiment Classification

arXiv: https://arxiv.org/abs/2010.08318 · HTML: https://mlsanity.com/paper/2010.08318/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Oct 2020.

## Built on by (most cited first, 7 of 7)

- [Is ChatGPT a Good Sentiment Analyzer? A Preliminary Study](https://mlsanity.com/paper/2304.04339/) (2023)
- [CONDAQA: A Contrastive Reading Comprehension Dataset for Reasoning about Negation](https://mlsanity.com/paper/2211.00295/) (2022)
- [This is not correct! Negation-aware Evaluation of Language Generation Systems](https://mlsanity.com/paper/2307.13989/) (2023)
- [Beyond Positive Scaling: How Negation Impacts Scaling Trends of Language Models](https://mlsanity.com/paper/2305.17311/) (2023)
- [HausaNLP at SemEval-2023 Task 12: Leveraging African Low Resource TweetData for Sentiment Analysis](https://mlsanity.com/paper/2304.13634/) (2023)
- [Negation-Instance Based Evaluation of End-to-End Negation Resolution](https://mlsanity.com/paper/2109.10013/) (2021)
- [Human-in-the-Loop Disinformation Detection: Stance, Sentiment, or Something Else?](https://mlsanity.com/paper/2111.05139/) (2021)

## Abstract (arXiv)

The majority of work in targeted sentiment analysis has concentrated on finding better methods to improve the overall results. Within this paper we show that these models are not robust to linguistic phenomena, specifically negation and speculation. In this paper, we propose a multi-task learning method to incorporate information from syntactic and semantic auxiliary tasks, including negation and speculation scope detection, to create English-language models that are more robust to these phenomena. Further we create two challenge datasets to evaluate model performance on negated and speculative samples. We find that multi-task models and transfer learning via language modelling can improve performance on these challenge datasets, but the overall performances indicate that there is still much room for improvement. We release both the datasets and the source code at https://github.com/jerbarnes/multitask_negation_for_targeted_sentiment.