The Inner Sentiments of a Thought
This work provides a novel method for understanding and manipulating emotional content in language, with potential applications in fields like psychiatry, though it is incremental in building on existing sentiment analysis techniques.
The authors explored how transformer-based language models implicitly represent sentiments in text, training predictors to estimate the distribution of sentiments like valence and admiration from sentence prefixes. They demonstrated that these predictors are well-calibrated and can be used to analyze emotional trajectories in sentences and generate text with extreme sentiments.
Transformer-based large-scale language models (LLMs) are able to generate highly realistic text. They are duly able to express, and at least implicitly represent, a wide range of sentiments and color, from the obvious, such as valence and arousal to the subtle, such as determination and admiration. We provide a first exploration of these representations and how they can be used for understanding the inner sentimental workings of single sentences. We train predictors of the quantiles of the distributions of final sentiments of sentences from the hidden representations of an LLM applied to prefixes of increasing lengths. After showing that predictors of distributions of valence, determination, admiration, anxiety and annoyance are well calibrated, we provide examples of using these predictors for analyzing sentences, illustrating, for instance, how even ordinary conjunctions (e.g., "but") can dramatically alter the emotional trajectory of an utterance. We then show how to exploit the distributional predictions to generate sentences with sentiments in the tails of distributions. We discuss the implications of our results for the inner workings of thoughts, for instance for psychiatric dysfunction.