United States

This paper investigates biases of Large Language Models (LLMs) through the lens of grammatical gender. Drawing inspiration from seminal works in psycholinguistics, particularly the study of gender&#39;s influence on language perception, we leverage multilingual LLMs to revisit and expand upon the foundational experiments of Boroditsky (2003). Employing LLMs as a novel method for examining psycholinguistic biases related to grammatical gender, we prompt a model to describe nouns with adjectives in various languages, focusing specifically on languages with grammatical gender. In particular, we look at adjective co-occurrences across gender and languages, and train a binary classifier to predict grammatical gender given adjectives an LLM uses to describe a noun. Surprisingly, we find that a simple classifier can not only predict noun gender above chance but also exhibit cross-language transferability. We show that while LLMs may describe words differently in different languages, they are biased similarly.

EMNLP 2024

What an Elegant Bridge: Multilingual LLMs are Biased Similarly in Different Languages

llms

psycholinguistics

bias

## Welcome to EMNLP 2024! 
We are excited to welcome you to one of the most prominent conferences in the field of Natural Language Processing. This year, EMNLP 2024 is being held in a hybrid format,
offering both virtual and in-person participation in beautiful Miami. Due to a record-breaking number of submissions, we've expanded the total number of accepted papers to accommodate more cutting-edge research from around the globe.
### [Conference Handbook](https://drive.google.com/file/d/1WPROgxjLAC96AJL7Ugy0tEnYm7dkrbHt/view?usp=sharing)

You are required to register for this event. **Please register [here](https://2024.emnlp.org/registration/).** The EMNLP 2024 event page on Underline will be open to public one week prior to the event.

Please register!

EMNLP 2024 will take place in Miami, Florida from Nov 12th to Nov 16th, 2024, at the Hyatt Regency Miami Hote and on Underline for remote participants.

This paper investigates biases of Large Language Models (LLMs) through the lens of grammatical gender. Drawing inspiration from seminal works in psycholinguistics, particularly the study of gender's influence on language perception, we leverage multilingual LLMs to revisit and expand upon the foundational experiments of Boroditsky (2003). Employing LLMs as a novel method for examining psycholinguistic biases related to grammatical gender, we prompt a model to describe nouns with adjectives in various languages, focusing specifically on languages with grammatical gender. In particular, we look at adjective co-occurrences across gender and languages, and train a binary classifier to predict grammatical gender given adjectives an LLM uses to describe a noun. Surprisingly, we find that a simple classifier can not only predict noun gender above chance but also exhibit cross-language transferability. We show that while LLMs may describe words differently in different languages, they are biased similarly.

This paper explores the intersection of psychology and artificial intelligence through the development and evaluation of specialized Large Language Models (LLMs). We introduce PsychoLex , a suite of resources designed to enhance LLMs' proficiency in psycho-logical tasks in both Persian and English. Key contributions include the PsychoLexQA dataset for instructional content and the PsychoLexEval dataset for rigorous evaluation of LLMs in complex psychological scenarios. Additionally, we present the PsychoLexLLaMA model, optimized specifically for psychological applications, demonstrating superior performance compared to general-purpose models. The findings underscore the potential of tailored LLMs for advancing psychological research and applications, while also highlighting areas for further refinement. This research offers a foundational step towards integrating LLMs into specialized psychological domains, with implications for future advancements in AI-driven psychological practice.

PsychoLex: Unveiling the Psychological Mind of Large Language Models

Automatic text summarization helps to digest the vast and ever-growing amount of scientific publications. While transformer-based solutions like BERT and SciBERT have advanced scientific summarization, lengthy documents pose a challenge due to the token limits of these models. To address this issue, we introduce and evaluate a two-stage model that combines an extract-then-compress framework. Our model incorporates a “graph-augmented extraction module” to select order-based salient sentences and an “abstractive compression module” to generate concise summaries. Additionally, we introduce the BioConSumm dataset, which focuses on biodiversity conservation, to support underrepresented domains and explore domain-specific summarization strategies. Out of the tested models, our model achieves the highest ROUGE-2 and ROUGE-L scores on our newly created dataset (BioConSumm) and on the SUMPUBMED dataset, which serves as a benchmark in the field of biomedicine.

Two-Stage Graph-Augmented Summarization for Scientific Documents

Instruction finetuning is a popular paradigm to align large language models (LLM) with human intent. Despite its popularity, this idea is less explored in improving LLMs to align existing foundation models with scien- tific disciplines, concepts and goals. In this work, we present SciTune as a tuning frame- work to improve the ability of LLMs to follow multimodal instructions generated from scien- tific publications. To test our methodology, we train a large multimodal model LLaMA- SciTune that connects a vision encoder and LLM for science-focused visual and language understanding. LLaMA-SciTune significantly outperforms the state-of-the-art models in the generated figure types and captions in SciCap and VisText benchmarks. In comparison to the models that are finetuned with synthetic data only, LLaMA-SciTune surpasses human performance on average and in many sub- categories on the ScienceQA benchmark. Our results demonstrate that human-generated sci- entific multimodal instructions remain highly valuable in tuning LLMs to perform well on science tasks, despite their lower volume and relative scarcity compared to synthetic data. We publicly release the SciTune codebase.

SCITUNE: Aligning Large Language Models with Human-Curated Scientific Multimodal Instructions

Semi-structured interviews (SSIs) are a commonly employed data-collection method in healthcare research, offering in-depth qualitative insights into subject experiences. Despite their value, manual analysis of SSIs is notoriously time-consuming and labor-intensive, in part due to the difficulty of extracting and categorizing emotional responses, and challenges in scaling human evaluation for large populations. In this study, we develop RACER, a Large Language Model (LLM) based expert-guided automated pipeline that efficiently converts raw interview transcripts into insightful domain-relevant themes and sub-themes. We used RACER to analyze SSIs conducted with 93 healthcare professionals and trainees to assess the broad personal and professional mental health impacts of the COVID-19 crisis. RACER achieves moderately high agreement with two human evaluators (72\%), which approaches the human inter-rater agreement (77\%). Interestingly, LLMs and humans struggle with similar content involving nuanced emotional, ambivalent/dialectical, and psychological statements. Our study highlights the opportunities and challenges in using LLMs to improve research efficiency and opens new avenues for scalable analysis of SSIs in healthcare research.

RACER: An LLM-powered Methodology for Scalable Analysis of Semi-structured Mental Health Interviews

Understanding and modeling collective intelligence is essential for addressing complex social systems. Directed graphs called fuzzy cognitive maps (FCMs) offer a powerful tool for encoding causal mental models, but extracting high-integrity FCMs from text is challenging. This study presents an approach using large language models (LLMs) to automate FCM extraction. We introduce novel graph-based similarity measures and evaluate them by correlating their outputs with human judgments through the Elo rating system. Results show positive correlations with human evaluations, but even the best-performing measure exhibits limitations in capturing FCM nuances. Fine-tuning LLMs improves performance, but existing measures still fall short. This study highlights the need for soft similarity measures tailored to FCM extraction, advancing collective intelligence modeling with NLP.

Soft Measures for Extracting Causal Collective Intelligence

Effective generation of novel hypotheses is instrumental to scientific progress. So far, researchers have been the main powerhouse behind hypothesis generation by painstaking data analysis and thinking (also known as the Eureka moment). In this paper, we examine the potential of large language models (LLMs) to generate hypotheses. We focus on hypothesis generation based on data (i.e., labeled examples). To enable LLMs to handle long contexts, we generate initial hypotheses from a small number of examples and then update them iteratively to improve the quality of hypotheses. Inspired by multi-armed bandits, we design a reward function to inform the exploitation-exploration tradeoff in the update process. Our algorithm is able to generate hypotheses that enable much better predictive performance than few-shot prompting in classification tasks, improving accuracy by 31.7% on a synthetic dataset and by 13.9%, 3.3% and, 24.9% on three real-world datasets. We also outperform supervised learning by 12.1% and 11.6% on two challenging real-world datasets. Furthermore, we find that the generated hypotheses not only corroborate human-verified theories but also uncover new insights for the tasks.

Hypothesis Generation with Large Language Models

Large Language Model (LLM) agents have been increasingly adopted as simulation tools to model humans in applications such as social science. However, one fundamental question remains: can LLM agents really simulate human behaviors? In this paper, we focus on one of the most critical behaviors in human interactions, trust, and aim to investigate whether or not LLM agents can simulate human trust behaviors. We first find that LLM agents generally exhibit trust behaviors, referred to as agent trust, under the framework of Trust Games, which are widely recognized in behavioral economics. Then, we discover that LLM agents can have high behavioral alignment with humans regarding trust behaviors, particularly for GPT-4, indicating the feasibility to simulate human trust behaviors with LLM agents. In addition, we probe into the biases in agent trust and the differences in agent trust towards agents and humans. We also explore the intrinsic properties of agent trust under conditions including advanced reasoning strategies and external manipulations. We further offer important implications of our discoveries for various scenarios where trust is paramount. Our study provides new insights into the behaviors of LLM agents and the fundamental analogy between LLMs and humans.

Can Large Language Model Agents Simulate Human Trust Behaviors?

Moral sentiments expressed in natural language significantly influence both online and offline environments, shaping behavioral styles and in teraction patterns, including social media self presentation, cyberbullying, adherence to so cial norms, and ethical decision-making. To effectively measure moral sentiments in natural language processing texts, it is crucial to utilize large, annotated datasets that provide nuanced understanding for accurate analysis and model training. However, existing corpora, while valuable, often face linguistic limitations. To address this gap in the Chinese language domain, we introduce the Moral Foundations Weibo Corpus. This corpus consists of 25,671 Chinese comments on Weibo, encompassing six diverse topic areas. Each comment is manually annotated by at least three systematically trained an notators based on ten moral categories derived from a grounded theory of morality. To assess annotator reliability, we present the kappa test results, a gold standard for measuring consis tency. Additionally, we apply several the latest large language models to supplement the manual annotations, conducting analytical experiments to compare their performance and report baseline results for moral sentiment classification.

The Moral Foundations Weibo Corpus


Flavor development in the food industry is increasingly challenged by the need for rapid innovation and precise flavor profile creation. Traditional flavor research methods typically rely on iterative, subjective testing, which lacks the efficiency and scalability required for modern demands. This paper proposes three contributions to address the challenges. Firstly, we define a new problem domain for autonomous scientific agents in flavor science, conceptualized as the generation of hypotheses for flavor development using a vast database of chemical and sensory attributes.

FOODPUZZLE: Toward Developing Large Language Models as Flavor Scientist

Humor is an important social construct with various roles in human communication, yet clinicians remain divided on its appropriateness and effectiveness. Despite its importance, empirical research on humor in psychotherapy is limited. This study explores the theoretical concept of ``humor'' by examining the operational variable of "laughs" within psychotherapy. Method: We analyzed transcriptions from 872 psychotherapy sessions involving 68 clients treated by 59 therapists. Clients self-reported their symptoms and state of well-being before each session, while both clients and therapists provided self-reports on their therapeutic alliance after each session. Through text analysis, we extracted the number of laughs and words for each session. We investigated the within-client associations between laughs and symptoms, well-being, therapeutic alliance, and clients' number of words. Results: We found session-level associations between laughs and well-being, symptoms, and the number of words. However, no significant associations were observed between laughs and the therapeutic alliance, either from the perspective of the therapist or the client.

Downloads

Next from EMNLP 2024

PsychoLex: Unveiling the Psychological Mind of Large Language Models

Similar lecture

What an Elegant Bridge: Multilingual LLMs are Biased Similarly in Different Languages

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from EMNLP 2024

PsychoLex: Unveiling the Psychological Mind of Large Language Models

Similar lecture

What an Elegant Bridge: Multilingual LLMs are Biased Similarly in Different Languages

Downloads