14 citations · 39 across the 29 of their papers we have counts for
33 papers
In-Context Learning for the Imputation of Public Opinion Data with Large Language Models
Tobias Holtdirk, Georg Ahnert, Joseph W Sakshaug +1
Large language models have been widely evaluated as simulators of individual survey responses. In practice, however, fully unobserved responses are rare; the dominant problem is pa…
ChinaHeritaQA: A Culturally-Grounded Visual Question Answering Dataset for World Heritage Sites in China
Yi Zhang, Bolei Ma, Yong Cao +3
We introduce ChinaHeritaQA, a multimodal benchmark dataset for evaluating the cultural reasoning abilities of vision-language models (VLMs) on UNESCO World Heritage sites in China.…
Beyond Correctness: Evaluating and Improving LLM Feedback in Statistical Education
Niklas Ippisch, Markus Herklotz, Anna-Carolina Haensch +1
Large language models (LLMs) have been proposed as scalable tools to address the gap between the importance of individualized written feedback and the practical challenges of provi…
Can we trust LLMs as a tutor for our students? Evaluating the Quality of LLM-generated Feedback in Statistics Exams
Markus Herklotz, Niklas Ippisch, Anna-Carolina Haensch
One of the central challenges for instructors is offering meaningful individual feedback, especially in large courses. Faced with limited time and resources, educators are often fo…
Systematic Evaluation of Uncertainty Estimation Methods in Large Language Models
Christian Hobelsberger, Theresa Winner, Andreas Nawroth +2
Large language models (LLMs) produce outputs with varying levels of uncertainty, and, just as often, varying levels of correctness; making their practical reliability far from guar…
Capabilities and Evaluation Biases of Large Language Models in Classical Chinese Poetry Generation: A Case Study on Tang Poetry
Bolei Ma, Yina Yao, Anna-Carolina Haensch
Large Language Models (LLMs) are increasingly applied to creative domains, yet their performance in classical Chinese poetry generation and evaluation remains poorly understood. We…
Too Open for Opinion? Embracing Open-Endedness in Large Language Models for Social Simulation
Bolei Ma, Yong Cao, Indira Sen +4
Large Language Models (LLMs) are increasingly used to simulate public opinion and other social phenomena. Most current studies constrain these simulations to multiple-choice or sho…
Survey Response Generation: Generating Closed-Ended Survey Responses In-Silico with Large Language Models
Georg Ahnert, Anna-Carolina Haensch, Barbara Plank +1
Many in-silico simulations of human survey responses with large language models (LLMs) focus on generating closed-ended survey responses, whereas LLMs are typically trained to gene…
Mexico 2021: Psychological Intimate Partner Violence Against Women and the Role of Childhood Violence Exposure -- A Machine Learning Approach
Clara Strasser Ceballos, Anna-Carolina Haensch
In 2021, psychological violence was the most prevalent form of intimate partner violence (IPV) suffered by women in Mexico. The consequences of psychological IPV can include low se…
AIn't Nothing But a Survey? Using Large Language Models for Coding German Open-Ended Survey Responses on Survey Motivation
Leah von der Heyde, Anna-Carolina Haensch, Bernd Weiß +1
The recent development and wider accessibility of LLMs have spurred discussions about how they can be used in survey research, including classifying open-ended survey responses. Du…
Safer Prompts: Reducing Risks from Memorization in Visual Generative AI
Lena Reissinger, Yuanyuan Li, Anna-Carolina Haensch +1
Visual Generative AI models have demonstrated remarkable capability in generating high-quality images from user inputs like text prompts. However, because these models have billion…
"It Listens Better Than My Therapist": Exploring Social Media Discourse on LLMs as Mental Health Tool
Anna-Carolina Haensch
The emergence of generative AI chatbots such as ChatGPT has prompted growing public and academic interest in their role as informal mental health support tools. While early rule-ba…
Cracking the Code: Evaluating Zero-Shot Prompting Methods for Providing Programming Feedback
Niklas Ippisch, Anna-Carolina Haensch, Jan Simson +3
Despite the growing use of large language models (LLMs) for providing feedback, limited research has explored how to achieve high-quality feedback. This case study introduces an ev…
Algorithmic Fidelity of Large Language Models in Generating Synthetic German Public Opinions: A Case Study
Bolei Ma, Berk Yoztyurk, Anna-Carolina Haensch +5
In recent research, large language models (LLMs) have been increasingly used to investigate public opinions. This study investigates the algorithmic fidelity of LLMs, i.e., the abi…
United in Diversity? Contextual Biases in LLM-Based Predictions of the 2024 European Parliament Elections
Leah von der Heyde, Anna-Carolina Haensch, Alexander Wenz +1
"Synthetic samples" based on large language models (LLMs) have been argued to serve as efficient alternatives to surveys of humans, assuming that their training data includes infor…
Vox Populi, Vox AI? Using Language Models to Estimate German Public Opinion
Leah von der Heyde, Anna-Carolina Haensch, Alexander Wenz
The recent development of large language models (LLMs) has spurred discussions about whether LLM-generated "synthetic samples" could complement or replace traditional surveys, cons…
The Potential and Challenges of Evaluating Attitudes, Opinions, and Values in Large Language Models
Bolei Ma, Xinpeng Wang, Tiancheng Hu +4
Recent advances in Large Language Models (LLMs) have sparked wide interest in validating and comprehending the human-like cognitive-behavioral traits LLMs may capture and convey. T…
ODEs and Mandatory Voting
Christoph Börgers, Natasa Dragovic, Anna Haensch +2
This paper presents mathematics relevant to the question whether voting should be mandatory. Assuming a static distribution of voters' political beliefs, we model how politicians m…
30 Years of Synthetic Data
Joerg Drechsler, Anna-Carolina Haensch
The idea to generate synthetic data as a tool for broadening access to sensitive microdata has been proposed for the first time three decades ago. While first applications of the i…
Seeing ChatGPT Through Students' Eyes: An Analysis of TikTok Data
Anna-Carolina Haensch, Sarah Ball, Markus Herklotz +1
Advanced large language models like ChatGPT have gained considerable attention recently, including among students. However, while the debate on ChatGPT in academia is making waves,…
A blue sky bifurcation in the dynamics of political candidates
Christoph Börgers, Bruce Boghosian, Natasa Dragovic +1
Political candidates often shift their positions opportunistically in hopes of capturing more votes. When there are only two candidates, the best strategy for each of them is often…
A particle method for continuous Hegselmann-Krause opinion dynamics
Bruce Boghosian, Christoph Börgers, Natasa Dragovic +2
We derive a differential-integral equation akin to the Hegselmann-Krause model of opinion dynamics, and propose a particle method for solving the equation. Numerical experiments de…
A geospatial bounded confidence model including mega-influencers with an application to Covid-19 vaccine hesitancy
Anna Haensch, Natasa Dragovic, Christoph Börgers +1
We introduce a geospatial bounded confidence model with mega-influencers, inspired by Hegselmann and Krause. The inclusion of geography gives rise to large-scale geospatial pattern…
An Equity-Aware Recommender System for Curating Art Exhibits Based on Locally-Constrained Graph Matching
Anna Haensch, Dina Deitsch
Public art shapes our shared spaces. Public art should speak to community and context, and yet, recent work has demonstrated numerous instances of art in prominent institutions fav…