Team UTSA-NLP at SemEval 2024 Task 5: Prompt Ensembling for Argument Reasoning in Civil Procedures with GPT4

In this paper, we present our system for the SemEval Task 5, The Legal Argument Reasoning Task in Civil Procedure Challenge. Legal argument reasoning is an essential skill that all law students must master. Moreover, it is important to develop natural language processing solutions that can reason about a question given terse domain-specific contextual information. Our system explores a prompt-based solution using GPT4 to reason over legal arguments. We also evaluate an ensemble of prompting strategies, including chain-of-thought reasoning and in-context learning. Overall, our system results in a Macro F1 of .8095 on the validation dataset and .7315 (5th out of 21 teams) on the final test set. Code for this project is available at https://github.com/danschumac1/CivilPromptReasoningGPT4.

Extracting Biomedical Entities from Noisy Audio Transcripts

Automatic Speech Recognition (ASR) technology is fundamental in transcribing spoken language into text, with considerable applications in the clinical realm, including streamlining medical transcription and integrating with Electronic Health Record (EHR) systems. Nevertheless, challenges persist, especially when transcriptions contain noise, leading to significant drops in performance when Natural Language Processing (NLP) models are applied. Named Entity Recognition (NER), an essential clinical task, is particularly affected by such noise, often termed the ASR-NLP gap. Prior works have primarily studied ASR's efficiency in clean recordings, leaving a research gap concerning the performance in noisy environments. This paper introduces a novel dataset, BioASR-NER, designed to bridge the ASR-NLP gap in the biomedical domain, focusing on extracting adverse drug reactions and mentions of entities from the Brief Test of Adult Cognition by Telephone (BTACT) exam. Our dataset offers a comprehensive collection of almost 2,000 clean and noisy recordings. In addressing the noise challenge, we present an innovative transcript-cleaning method using GPT4, investigating both zero-shot and few-shot methodologies. Our study further delves into an error analysis, shedding light on the types of errors in transcription software, corrections by GPT4, and the challenges GPT4 faces. This paper aims to foster improved understanding and potential solutions for the ASR-NLP gap, ultimately supporting enhanced healthcare documentation practices.

Deciphering Textual Authenticity: A Generalized Strategy through the Lens of Large Language Semantics for Detecting Human vs. Machine-Generated Text

With the recent proliferation of Large Language Models (LLMs), there has been an increasing demand for tools to detect machine-generated text. The effective detection of machine-generated text face two pertinent problems: First, they are severely limited in generalizing against real-world scenarios, where machine-generated text is produced by a variety of generators, including but not limited to GPT-4 and Dolly, and spans diverse domains, ranging from academic manuscripts to social media posts. Second, existing detection methodologies treat texts produced by LLMs through a restrictive binary classification lens, neglecting the nuanced diversity of artifacts generated by different LLMs. In this work, we undertake a systematic study on the detection of machine-generated text in real-world scenarios. We first study the effectiveness of state-of-the-art approaches and find that they are severely limited against text produced by diverse generators and domains in the real world. Furthermore, t-SNE visualizations of the embeddings from a pretrained LLM's encoder show that they cannot reliably distinguish between human and machine-generated text. Based on our findings, we introduce a novel system, T5LLMCipher, for detecting machine-generated text using a pretrained T5 encoder combined with LLM embedding sub-clustering to address the text produced by diverse generators and domains in the real world. We evaluate our approach across 9 machine-generated text systems and 9 domains and find that our approach provides state-of-the-art generalization ability, with an average increase in F1 score on machine-generated text of 19.6\% on unseen generators and domains compared to the top performing existing approaches and correctly attributes the generator of text with an accuracy of 93.6\%.

BabyStories: Can Reinforcement Learning Teach Baby Language Models to Write Better Stories?

Language models have seen significant growth in the size of their corpus, leading to notable performance improvements. Yet, there has been limited progress in developing models that handle smaller, more human-like datasets. As part of the BabyLM shared task, this study explores the impact of reinforcement learning from human feedback (RLHF) on language models pretrained from scratch with a limited training corpus. Comparing two GPT-2 variants, the larger model performs better in storytelling tasks after RLHF fine-tuning. These findings suggest that RLHF techniques may be more advantageous for larger models due to their higher learning and adaptation capacity, though more experiments are needed to confirm this finding. These insights highlight the potential benefits of RLHF fine-tuning for language models within limited data, enhancing their ability to maintain narrative focus and coherence while adhering better to initial instructions in storytelling tasks. The code for this work is publicly at https://github.com/Zephyr1022/BabyStories-UTSA.

Towards Understanding the Generalization of Medical Text-to-SQL Models and Datasets

Electronic medical records (EMRs) are stored in relational databases. It can be challenging to access the required information if the user is unfamiliar with the database schema or general database fundamentals. Hence, researchers have explored text-to-SQL generation methods that provide healthcare professionals direct access to EMR data without needing a database expert. However, currently available datasets have been essentially "solved" with state-of-the-art models achieving accuracy greater than or near 90%. In this paper, we show that there is still a long way to go before solving text-to-SQL generation in the medical domain. To show this, we create new splits of the existing medical text-to-SQL dataset MIMICSQL that better measure the generalizability of the resulting models. We evaluate state-of-the-art language models on our new split showing substantial drops in performance with accuracy dropping from up to 92% to 28%, thus showing substantial room for improvement. Moreover, we introduce a novel data augmentation approach to improve the generalizability of the language models. Overall, this paper is the first step towards developing more robust text-to-SQL models in the medical domain.\footnote{The dataset and code will be released upon acceptance.

Bike Frames: Understanding the Implicit Portrayal of Cyclists in the News

Increasing the number of cyclists, whether for general transport or recreation, can provide health improvements and reduce the environmental impact of vehicular transportation. However, the public's perception of cycling may be driven by the ideologies and reporting standards of news agencies. For instance, people may identify cyclists on the road as "dangerous" if news agencies overly report cycling accidents, limiting the number of people that cycle for transportation. Moreover, if fewer people cycle, there may be less funding from the government to invest in safe infrastructure. In this paper, we explore the perceived perception of cyclists within news headlines. To accomplish this, we introduce a new dataset, "Bike Frames", that can help provide insight into how headlines portray cyclists and help detect accident-related headlines. Next, we introduce a multi-task (MT) regularization approach that increases the detection accuracy of accident-related posts, demonstrating improvements over traditional MT frameworks. Finally, we compare and contrast the perceptions of cyclists with motorcyclist-related headlines to ground the findings with another related activity for both male- and female-related posts. Our findings show that general news websites are more likely to report accidents about cyclists than other events. Moreover, cyclist-specific websites are more likely to report about accidents than motorcycling-specific websites, even though there is more potential danger for motorcyclists. Finally, we show substantial differences in the reporting about male vs. female-related persons, e.g., more male-related cyclists headlines are related to accidents, but more female-related motorcycling headlines about accidents. WARNING: This paper contains descriptions of accidents and death.

Linguistic Elements of Engaging Customer Service Discourse on Social Media

Customers are rapidly turning to social media for customer support. While brand agents on these platforms are motivated and well-intentioned to help and engage with customers, their efforts are often ignored if their initial response to the customer does not match a specific tone, style, or topic the customer is aiming to receive. The length of a conversation can reflect the effort and quality of the initial response made by a brand toward collaborating and helping consumers, even when the overall sentiment of the conversation might not be very positive. Thus, through this study, we aim to bridge this critical gap in the existing literature by analyzing language's content and stylistic aspects such as expressed empathy, psycho-linguistic features, dialogue tags, and metrics for quantifying personalization of the utterances that can influence the engagement of an interaction. This paper demonstrates that we can predict engagement using initial customer and brand posts.

A Marker-based Neural Network System for Extracting Social Determinants of Health

Objective. The impact of social determinants of health (SDoH) on patients' healthcare quality and the disparity is well-known. Many SDoH items are not coded in structured forms in electronic health records. These items are often captured in free-text clinical notes, but there are limited methods for automatically extracting them. We explore a multi-stage pipeline involving named entity recognition (NER), relation classification (RC), and text classification methods to extract SDoH information from clinical notes automatically. Materials and Methods. The study uses the N2C2 Shared Task data, which was collected from two sources of clinical notes: MIMIC-III and University of Washington Harborview Medical Centers. It contains 4480 social history sections with full annotation for twelve SDoHs. In order to handle the issue of overlapping entities, we developed a novel marker-based NER model. We used it in a multi-stage pipeline to extract SDoH information from clinical notes. Results. Our marker-based system outperformed the state-of-the-art span-based models at handling overlapping entities based on the overall Micro-F1 score performance. It also achieved state-of-the-art performance compared to the shared task methods. Conclusion. The major finding of this study is that the multi-stage pipeline effectively extracts SDoH information from clinical notes. This approach can potentially improve the understanding and tracking of SDoHs in clinical settings. However, error propagation may be an issue, and further research is needed to improve the extraction of entities with complex semantic meanings and low-resource entities using external knowledge.

A Comprehensive Study of Gender Bias in Chemical Named Entity Recognition Models

Objective. Chemical named entity recognition (NER) models have the potential to impact a wide range of downstream tasks, from identifying adverse drug reactions to general pharmacoepidemiology. However, it is unknown whether these models work the same for everyone. Performance disparities can potentially cause harm rather than the intended good. Hence, in this paper, we measure gender-related performance disparities of chemical NER systems. Materials and Methods. We develop a framework to measure gender bias in chemical NER models using synthetic data and a newly annotated dataset of over 92,405 words with self-identified gender information from Reddit. We applied and evaluated state-of-the-art biomedical NER models. Results. Our findings indicate that chemical NER models are biased. The results of the bias tests on the synthetic dataset and the real-world data multiple fairness issues. For example, for synthetic data, we find that female-related names are generally classified as chemicals, particularly in datasets containing many brand names rather than standard ones. For both datasets, we find consistent fairness issues resulting in substantial performance disparities between female- and male-related data. Discussion. Our study highlights the issue of biases in chemical NER models. For example, we find that many systems cannot detect contraceptives (e.g., birth control). Conclusion. Chemical NER models are biased and can be harmful to female-related groups. Therefore, practitioners should carefully consider the potential biases of these models and take steps to mitigate them.

Measuring Geographic Performance Disparities of Offensive Language Classifiers

Text classifiers are applied at scale in the form of one-size-fits-all solutions. Nevertheless, many studies show that classifiers are biased regarding different languages and dialects. When measuring and discovering these biases, some gaps present themselves and should be addressed. First, ``Does language, dialect, and topical content vary across geographical regions?'' and secondly ``If there are differences across the regions, do they impact model performance?''. We introduce a novel dataset called GeoOLID with more than 14 thousand examples across 15 geographically and demographically diverse cities to address these questions. We perform a comprehensive analysis of geographical-related content and their impact on performance disparities of offensive language detection models. Overall, we find that current models do not generalize across locations. Likewise, we show that while offensive language models produce false positives on African American English, model performance is not correlated with each city's minority population proportions. Warning: This paper contains offensive language.