An important assumption that comes with using LLMs on psycholinguistic data has gone unverified. LLM-based predictions are based on subword tokenization, not decomposition of words into morphemes. Does that matter? We carefully test this by comparing surprisal estimates using orthographic, morphological, and BPE tokenization against reading time data. Our results replicate previous findings and provide evidence that in the aggregate, predictions using BPE tokenization do not suffer relative to morphological and orthographic segmentation. However, a finer-grained analysis points to potential issues with relying on BPE-based tokenization, as well as providing promising results involving morphologically-aware surprisal estimates and suggesting a new method for evaluating morphological prediction.
This study focuses on how different modalities of human communication can be used to distinguish between healthy controls and subjects with schizophrenia who exhibit strong positive symptoms. We developed a multi-modal schizophrenia classification system using audio, video, and text. Facial action units and vocal tract variables were extracted as low-level features from video and audio respectively, which were then used to compute high-level coordination features that served as the inputs to the audio and video modalities. Context-independent text embeddings extracted from transcriptions of speech were used as the input for the text modality. The multi-modal system is developed by fusing a segment-to-session-level classifier for video and audio modalities with a text model based on a Hierarchical Attention Network (HAN) with cross-modal attention. The proposed multi-modal system outperforms the previous state-of-the-art multi-modal system by 8.53% in the weighted average F1 score.
Language is multifaceted. A given utterance can be re-expressed in equivalent forms, and its implicit and explicit content support various logical and pragmatic inferences. When processing an utterance, we consider these different aspects, as mediated by our interpretive goals -- understanding that "it's dark in here" may be a veiled direction to turn on a light. Nonetheless, NLP methods typically operate over the surface form alone, eliding this nuance. In this work, we represent language with language, and direct an LLM to decompose utterances into logical and plausible inferences. The reduced complexity of the decompositions makes them easier to embed, opening up novel applications. Variations on our technique lead to state-of-the-art improvements on sentence embedding benchmarks, a substantive application in computational political science, and to a novel construct-discovery process, which we validate with human annotations.
Stressors are related to depression, but this relationship is complex. We investigate the relationship between open-ended text responses about stressors and depressive symptoms across gender and racial/ethnic groups. First, we use topic models and other NLP tools to find thematic and vocabulary differences when reporting stressors across demographic groups. We train language models using self-reported stressors to predict depressive symptoms, finding a relationship between stressors and depression. Finally, we find that differences in stressors translate to downstream performance differences across demographic groups.
Recently, the relationship between automated and human evaluation of topic models has been called into question. Method developers have staked the efficacy of new topic model variants on automated measures, and their failure to approximate human preferences places these models on uncertain ground. Moreover, existing evaluation paradigms are often divorced from real-world use. Motivated by content analysis as a dominant real-world use case for topic modeling, we analyze two related aspects of topic models that affect their effectiveness and trustworthiness in practice for that purpose: the stability of their estimates and the extent to which the model's discovered categories align with human-determined categories in the data. We find that neural topic models fare worse in both respects compared to an established classical method. We take a step toward addressing both issues in tandem by demonstrating that a straightforward ensembling method can reliably outperform the members of the ensemble.
Topic model evaluation, like evaluation of other unsupervised methods, can be contentious. However, the field has coalesced around automated estimates of topic coherence, which rely on the frequency of word co-occurrences in a reference corpus. Recent models relying on neural components surpass classical topic models according to these metrics. At the same time, unlike classical models, the practice of neural topic model evaluation suffers from a validation gap: automatic coherence for neural models has not been validated using human experimentation. In addition, as we show via a meta-analysis of topic modeling literature, there is a substantial standardization gap in the use of automated topic modeling benchmarks. We address both the standardization gap and the validation gap. Using two of the most widely used topic model evaluation datasets, we assess a dominant classical model and two state-of-the-art neural models in a systematic, clearly documented, reproducible way. We use automatic coherence along with the two most widely accepted human judgment tasks, namely, topic rating and word intrusion. Automated evaluation will declare one model significantly different from another when corresponding human evaluations do not, calling into question the validity of fully automatic evaluations independent of human judgments.
The records of a clinical encounter can be extensive and complex, thus placing a premium on tools that can extract and summarize relevant information. This paper introduces the task of generating discharge summaries for a clinical encounter. Summaries in this setting need to be faithful, traceable, and scale to multiple long documents, motivating the use of extract-then-abstract summarization cascades. We introduce two new measures, faithfulness and hallucination rate for evaluation in this task, which complement existing measures for fluency and informativeness. Results across seven medical sections and five models show that a summarization architecture that supports traceability yields promising results, and that a sentence-rewriting approach performs consistently on the measure used for faithfulness (faithfulness-adjusted $F_3$) over a diverse range of generated sections.
Topic models are often used to identify human-interpretable topics to help make sense of large document collections. We use knowledge distillation to combine the best attributes of probabilistic topic models and pretrained transformers. Our modular method can be straightforwardly applied with any neural topic model to improve topic quality, which we demonstrate using two models having disparate architectures, obtaining state-of-the-art topic coherence. We show that our adaptable framework not only improves performance in the aggregate over all estimated topics, as is commonly reported, but also in head-to-head comparisons of aligned topics.
The vast majority of research in computer assisted medical coding focuses on coding at the document level, but a substantial proportion of medical coding in the real world involves coding at the level of clinical encounters, each of which is typically represented by a potentially large set of documents. We introduce encounter-level document attention networks, which use hierarchical attention to explicitly take the hierarchical structure of encounter documentation into account. Experimental evaluation demonstrates improvements in coding accuracy as well as facilitation of human reviewers in their ability to identify which documents within an encounter play a role in determining the encounter level codes.
An important component of achieving language understanding is mastering the composition of sentence meaning, but an immediate challenge to solving this problem is the opacity of sentence vector representations produced by current neural sentence composition models. We present a method to address this challenge, developing tasks that directly target compositional meaning information in sentence vector representations with a high degree of precision and control. To enable the creation of these controlled tasks, we introduce a specialized sentence generation system that produces large, annotated sentence sets meeting specified syntactic, semantic and lexical constraints. We describe the details of the method and generation system, and then present results of experiments applying our method to probe for compositional information in embeddings from a number of existing sentence composition models. We find that the method is able to extract useful information about the differing capacities of these models, and we discuss the implications of our results with respect to these systems' capturing of sentence information. We make available for public use the datasets used for these experiments, as well as the generation system.