48 citations · 175 across the 29 of their papers we have counts for
31 papers
How AI Writing Assistance Distorts Perceptions of AI Users and Their Opinions
Paul Röttger, Kobi Hackenburg, Hannah Rose Kirk +1
Hundreds of millions of people use AI for writing assistance. We asked how AI distorts perceptions of AI users and their opinions. Writers (N=2,939) wrote political opinion paragra…
Diffusion Language Models Are Natively Length-Aware
Vittorio Rossi, Giacomo Cirò, Davide Beltrame +3
Unlike autoregressive language models, which terminate variable-length generation upon predicting an End-of-Sequence (EoS) token, Diffusion Language Models (DLMs) operate over a fi…
No for Some, Yes for Others: Persona Prompts and Other Sources of False Refusal in Language Models
Flor Miriam Plaza-del-Arco, Paul Röttger, Nino Scherrer +3
Large language models (LLMs) are increasingly integrated into our daily lives and personalized. However, LLM personalization might also increase unintended side effects. Recent wor…
The Pluralistic Moral Gap: Understanding Judgment and Value Differences between Humans and Large Language Models
Giuseppe Russo, Debora Nozza, Paul Röttger +1
People increasingly rely on Large Language Models (LLMs) for moral advice, which may influence humans' decisions. Yet, little is known about how closely LLMs align with human moral…
TrojanStego: Your Language Model Can Secretly Be A Steganographic Privacy Leaking Agent
Dominik Meier, Jan Philip Wahle, Paul Röttger +2
As large language models (LLMs) become integrated into sensitive workflows, concerns grow over their potential to leak confidential information. We propose TrojanStego, a novel thr…
AILuminate: Introducing v1.0 of the AI Risk and Reliability Benchmark from MLCommons
Shaona Ghosh, Heather Frase, Adina Williams +99
The rapid advancement and deployment of AI systems have created an urgent need for standard safety-evaluation frameworks. This paper introduces AILuminate v1.0, the first comprehen…
IssueBench: Millions of Realistic Prompts for Measuring Issue Bias in LLM Writing Assistance
Paul Röttger, Musashi Hinck, Valentin Hofmann +4
Large language models (LLMs) are helping millions of users write texts about diverse issues, and in doing so expose users to different ideas and perspectives. This creates concerns…
MSTS: A Multimodal Safety Test Suite for Vision-Language Models
Paul Röttger, Giuseppe Attanasio, Felix Friedrich +19
Vision-language models (VLMs), which process image and text inputs, are increasingly integrated into chat assistants and other consumer AI applications. Without proper safeguards,…
Evidence of a log scaling law for political persuasion with large language models
Kobi Hackenburg, Ben M. Tappin, Paul Röttger +3
Large language models can now generate political messages as persuasive as those written by humans, raising concerns about how far this persuasiveness may continue to increase with…
Beyond Flesch-Kincaid: Prompt-based Metrics Improve Difficulty Classification of Educational Texts
Donya Rooein, Paul Rottger, Anastassia Shaitarova +1
Using large language models (LLMs) for educational applications like dialogue-based teaching is a hot topic. Effective teaching, however, requires teachers to adapt the difficulty…
Introducing v0.5 of the AI Safety Benchmark from MLCommons
Bertie Vidgen, Adarsh Agrawal, Ahmed M. Ahmed +97
This paper introduces v0.5 of the AI Safety Benchmark, which has been created by the MLCommons AI Safety Working Group. The AI Safety Benchmark has been designed to assess the safe…
Look at the Text: Instruction-Tuned Language Models are More Robust Multiple Choice Selectors than You Think
Xinpeng Wang, Chengzhi Hu, Bolei Ma +2
Multiple choice questions (MCQs) are commonly used to evaluate the capabilities of large language models (LLMs). One common way to evaluate the model response is to rank the candid…
SafetyPrompts: a Systematic Review of Open Datasets for Evaluating and Improving Large Language Model Safety
Paul Röttger, Fabio Pernisi, Bertie Vidgen +1
The last two years have seen a rapid growth in concerns around the safety of large language models (LLMs). Researchers and practitioners have met these concerns by creating an abun…
Improving Adversarial Data Collection by Supporting Annotators: Lessons from GAHD, a German Hate Speech Dataset
Janis Goldzycher, Paul Röttger, Gerold Schneider
Hate speech detection models are only as good as the data they are trained on. Datasets sourced from social media suffer from systematic gaps and biases, leading to unreliable mode…
Evaluating the Elementary Multilingual Capabilities of Large Language Models with MultiQ
Carolin Holtermann, Paul Röttger, Timm Dill +1
Large language models (LLMs) need to serve everyone, including a global majority of non-English speakers. However, most LLMs today, and open LLMs in particular, are often intended…
Political Compass or Spinning Arrow? Towards More Meaningful Evaluations for Values and Opinions in Large Language Models
Paul Röttger, Valentin Hofmann, Valentina Pyatkin +4
Much recent work seeks to evaluate values and opinions in large language models (LLMs) using multiple-choice surveys and questionnaires. Most of this work is motivated by concerns…
"My Answer is C": First-Token Probabilities Do Not Match Text Answers in Instruction-Tuned Language Models
Xinpeng Wang, Bolei Ma, Chengzhi Hu +5
The open-ended nature of language generation makes the evaluation of autoregressive large language models (LLMs) challenging. One common evaluation approach uses multiple-choice qu…
The Empty Signifier Problem: Towards Clearer Paradigms for Operationalising "Alignment" in Large Language Models
Hannah Rose Kirk, Bertie Vidgen, Paul Röttger +1
In this paper, we address the concept of "alignment" in large language models (LLMs) through the lens of post-structuralist socio-political theory, specifically examining its paral…
SimpleSafetyTests: a Test Suite for Identifying Critical Safety Risks in Large Language Models
Bertie Vidgen, Nino Scherrer, Hannah Rose Kirk +4
The past year has seen rapid acceleration in the development of large language models (LLMs). However, without proper steering and safeguards, LLMs will readily follow malicious in…
The Past, Present and Better Future of Feedback Learning in Large Language Models for Subjective Human Preferences and Values
Hannah Rose Kirk, Andrew M. Bean, Bertie Vidgen +2
Human feedback is increasingly used to steer the behaviours of Large Language Models (LLMs). However, it is unclear how to collect and incorporate feedback in a way that is efficie…
Safety-Tuned LLaMAs: Lessons From Improving the Safety of Large Language Models that Follow Instructions
Federico Bianchi, Mirac Suzgun, Giuseppe Attanasio +4
Training large language models to follow instructions makes them perform better on a wide range of tasks and generally become more helpful. However, a perfectly helpful model will…
XSTest: A Test Suite for Identifying Exaggerated Safety Behaviours in Large Language Models
Paul Röttger, Hannah Rose Kirk, Bertie Vidgen +3
Without proper safeguards, large language models will readily follow malicious instructions and generate toxic content. This risk motivates safety efforts such as red-teaming and l…
The Ecological Fallacy in Annotation: Modelling Human Label Variation goes beyond Sociodemographics
Matthias Orlikowski, Paul Röttger, Philipp Cimiano +1
Many NLP tasks exhibit human label variation, where different annotators give different labels to the same texts. This variation is known to depend, at least in part, on the sociod…
Personalisation within bounds: A risk taxonomy and policy framework for the alignment of large language models with personalised feedback
Hannah Rose Kirk, Bertie Vidgen, Paul Röttger +1
Large language models (LLMs) are used to generate content for a wide range of tasks, and are set to reach a growing audience in coming years due to integration in product interface…