Publications de Robert West | EPFL Graph Search

On the conversational persuasiveness of GPT-4

Early work has found that large language models (LLMs) can generate persuasive content. However, evidence on whether they can also personalize arguments to individual attributes remains limited, despite being crucial for assessing misuse. This preregistere ...

Springer Science and Business Media LLC2025

Post Guidance for Online Communities

Robert West, Manoel Horta Ribeiro

Effective content moderation in online communities is often a delicate balance between maintaining content quality and fostering user participation. In this paper, we introduce post guidance, a novel approach to community moderation that proactively guides ...

Association for Computing Machinery (ACM)2025

Prevalence and Prevention of Large Language Model Use in Crowd Work

Robert West

ASSOC COMPUTING MACHINERY2025

Deplatforming Norm-Violating Influencers on Social Media Reduces Overall Online Attention Toward Them

Robert West, Manoel Horta Ribeiro

From politicians to podcast hosts, online platforms have systematically banned (''deplatformed'') influential users for breaking platform guidelines. Previous inquiries on the effectiveness of this intervention are inconclusive because 1) they consider onl ...

Association for Computing Machinery (ACM)2025

A Logical Fallacy-Informed Framework for Argument Generation

Boi Faltings, Robert West, Luca Mouchel, Antoine Bosselut, Debjit Paul

Despite the remarkable performance of large language models (LLMs), they still struggle with generating logically sound arguments, resulting in potential risks such as spreading misinformation. An important factor contributing to LLMs' suboptimal performan ...

Association for Computational Linguistics2025

The AI Alignment Paradox The better we align AI models with our values, the easier we may make it to realign them with opposing values

Robert West

ASSOC COMPUTING MACHINERY2025

Activation Scaling for Steering and Interpreting Language Models

Robert West

Given the prompt “Rome is in”, can we steer a language model to flip its prediction of an incorrect token “France” to a correct token “Italy” by only multiplying a few relevant activation vectors with scalars? We argue that successfully intervening on a mo ...

Association for Computational Linguistics (ACL)2024

Open access improves the dissemination of science: insights from Wikipedia

Robert West

Wikipedia is a well-known platform for disseminating knowledge, and scientific sources, such as journal articles, play a critical role in supporting its mission. The open access movement aims to make scientific knowledge openly available, and we might intu ...

2024

Self-Recognition in Language Models

Robert West, Giuseppe Russo, Caglar Gulcehre

A rapidly growing number of applications rely on a small set of closed-source language models (LMs). This dependency might introduce novel security risks if LMs develop self-recognition capabilities. Inspired by human identity verification methods, we prop ...

Association for Computational Linguistics2024

Making Reasoning Matter: Measuring and Improving Faithfulness of Chain-of-Thought Reasoning

Boi Faltings, Robert West, Antoine Bosselut, Debjit Paul

Large language models (LLMs) have been shown to perform better when asked to reason step-by-step before answering a question. However, it is unclear to what degree the model's final answer is faithful to the stated reasoning steps. In this paper, we perfor ...

Association for Computational Linguistics (ACL)2024

Measuring and shaping the nutritional environment via food sales logs: case studies of campus-wide food choice and a call to action

Robin Adrien Zbinden, Robert West, Kristina Gligoric

Although diets influence health and the environment, measuring and changing nutrition is challenging. Traditional measurement methods face challenges, and designing and conducting behavior-changing interventions is conceptually and logistically complicated ...

Frontiers Media Sa2024

Fly-Swat or Cannon? Cost-Effective Language Model Choice via Meta-Modeling

Maxime Jean Julien Peyrard, Robert West, Marija Sakota

Generative language models (LMs) have become omnipresent across data science. For a wide variety of tasks, inputs can be phrased as natural language prompts for an LM, from whose output the solution can then be extracted. LM performance has consistently be ...

Assoc Computing Machinery2024

Food choice mimicry on a large university campus

Robert West

Social influence is a strong determinant of food consumption, which in turn influences the environment and health. Purchasing mimicry, a phenomenon where a person copies another person’s purchases, has been identified as the key governing mechanism. Althou ...

2024

Can Language Models Recognize Convincing Arguments?

Robert West

The capabilities of large language models (LLMs) have raised concerns about their potential to create and propagate convincing narratives.Here, we study their performance in detecting convincing arguments to gain insights into LLMs' persuasive capabilities ...

Association for Computational Linguistics (ACL)2024

In-Class Data Analysis Replications: Teaching StudentsWhile Testing Science

Robert West

Science is facing a reproducibility crisis. Overcoming it would require concerted efforts to replicate prior studies, but the incentives for researchers are currently weak, as replicating prior studies requires considerable time and effort without providin ...

MIT PRESS2024

Causally estimating the effect of YouTube's recommender system using counterfactual bots

Robert West, Manoel Horta Ribeiro

the effects of the algorithm from a user's intentions. Here we propose a method that we call "counterfactual bots" to causally estimate the role of algorithmic recommendations on the consumption of highly partisan content on YouTube. By comparing bots that ...

National Academy of Sciences2024

A Large-Scale Characterization of How Readers Browse Wikipedia

Tiziano Piccardi, Robert West, Akhil Arora

Despite the importance and pervasiveness of Wikipedia as one of the largest platforms for open knowledge, surprisingly little is known about how people navigate its content when seeking information. To bridge this gap, we present the first systematic large ...

ASSOC COMPUTING MACHINERY2023

Language Model Decoding as Likelihood–Utility Alignment

Boi Faltings, Maxime Jean Julien Peyrard, Robert West, Martin Josifoski, Valentin Hartmann, Jiheng Wei, Frano Rajic, Debjit Paul

A critical component of a successful language generation pipeline is the decoding algorithm. However, the general principles that should guide the choice of a decoding algorithm re- main unclear. Previous works only compare decoding algorithms in narrow sc ...

Assoc Computational Linguistics-Acl2023

Distribution Inference Risks: Identifying and Mitigating Sources of Leakage

Léo Nicolas René Meynent, Maxime Jean Julien Peyrard, Robert West, Valentin Hartmann

A large body of work shows that machine learning (ML) models can leak sensitive or confidential information about their training data. Recently, leakage due to distribution inference (or property inference) attacks is gaining attention. In this attack, the ...

IEEE COMPUTER SOC2023

An Ordinal Latent Variable Model of Conflict Intensity

Robert West

Measuring the intensity of events is crucial for monitoring and tracking armed conflict. Advances in automated event extraction have yielded massive data sets of '' who did what to whom '' micro-records that enable datadriven approaches to monitoring confl ...

Assoc Computational Linguistics-Acl2023