Leveraging natural language processing to study emotional coherence in psychotherapy.

The association between emotional experience and expression, known as emotional coherence, is considered important for individual functioning. Recent advances in natural language processing (NLP) make it possible to automatically recognize verbally expressed emotions in psychotherapy dialogues and to explore emotional coherence with larger samples and finer granularity than previously. The present study used state-of-the-art emotion recognition models to automatically label clients' emotions at the utterance level, employed these labeled data to examine the coherence between verbally expressed emotions and self-reported emotions, and examined the associations between emotional coherence and clients' improvement in functioning throughout treatment. The data comprised 872 transcribed sessions from 68 clients. Clients self-reported their functioning before each session and their emotions after each. A subsample of 196 sessions were manually coded. A transformer-based approach was used to automatically label the remaining data for a total of 139,061 utterances. Multilevel modeling was used to assess emotional coherence and determine whether it was associated with changes in clients' functioning throughout treatment. The emotion recognition model demonstrated moderate performance. The findings indicated a significant association between verbally expressed emotions and self-reported emotions. Coherence in clients' negative emotions was associated with improvement in functioning. The results suggest an association between clients' subjective experience and their verbal expression of emotions and underscore the importance of this coherence to functioning. NLP may uncover crucial emotional processes in psychotherapy. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Label modification and bootstrapping for zero-shot cross-lingual hate speech detection.

The goal of hate speech detection is to filter negative online content aiming at certain groups of people. Due to the easy accessibility and multilinguality of social media platforms, it is crucial to protect everyone which requires building hate speech detection systems for a wide range of languages. However, the available labeled hate speech datasets are limited, making it difficult to build systems for many languages. In this paper we focus on cross-lingual transfer learning to support hate speech detection in low-resource languages, while highlighting label issues across application scenarios, such as inconsistent label sets of corpora or differing hate speech definitions, which hinder the application of such methods. We leverage cross-lingual word embeddings to train our neural network systems on the source language and apply them to the target language, which lacks labeled examples, and show that good performance can be achieved. We then incorporate unlabeled target language data for further model improvements by bootstrapping labels using an ensemble of different model architectures. Furthermore, we investigate the issue of label imbalance in hate speech datasets, since the high ratio of non-hate examples compared to hate examples often leads to low model performance. We test simple data undersampling and oversampling techniques and show their effectiveness.

Learning to reason over scene graphs: a case study of finetuning GPT-2 into a robot language model for grounded task planning.

Long-horizon task planning is essential for the development of intelligent assistive and service robots. In this work, we investigate the applicability of a smaller class of large language models (LLMs), specifically GPT-2, in robotic task planning by learning to decompose tasks into subgoal specifications for a planner to execute sequentially. Our method grounds the input of the LLM on the domain that is represented as a scene graph, enabling it to translate human requests into executable robot plans, thereby learning to reason over long-horizon tasks, as encountered in the ALFRED benchmark. We compare our approach with classical planning and baseline methods to examine the applicability and generalizability of LLM-based planners. Our findings suggest that the knowledge stored in an LLM can be effectively grounded to perform long-horizon task planning, demonstrating the promising potential for the future application of neuro-symbolic planning methods in robotics.

A taxonomy of anti-vaccination arguments from a systematic literature review and text modelling.

The proliferation of anti-vaccination arguments is a threat to the success of many immunization programmes. Effective rebuttal of contrarian arguments requires an approach that goes beyond addressing flaws in the arguments, by also considering the attitude roots-that is, the underlying psychological attributes driving a person's belief-of opposition to vaccines. Here, through a pre-registered systematic literature review of 152 scientific articles and thematic analysis of anti-vaccination arguments, we developed a hierarchical taxonomy that relates common arguments and themes to 11 attitude roots that explain why an individual might express opposition to vaccination. We further validated our taxonomy on coronavirus disease 2019 anti-vaccination misinformation, through a combination of human coding and machine learning using natural language processing algorithms. Overall, the taxonomy serves as a theoretical framework to link expressed opposition of vaccines to their underlying psychological processes. This enables future work to develop targeted rebuttals and other interventions that address the underlying motives of anti-vaccination arguments.