Mehrnaz Mofakhami

Ananya Sahu

Thomas Euyang

Brittawnya Prince

Madeline Smith

Hangyu Lin

Acyr Locatelli

Sara Hooker

Tom Kocmi

Aidan Gomez

Ivan Zhang

Phil Blunsom … (voir 6 de plus)

Nick Frosst

Joelle Pineau

Beyza Ermis

Ahmet Üstün

Julia Kreutzer

Marzieh Fadaee

Tiny Aya redefines what a small multilingual language model can achieve. Trained on 70 languages and refined through region-aware posttraini… (voir plus)ng, it delivers state-of-the-art in translation quality, strong multilingual understanding, and high-quality target-language generation, all with just 3.35B parameters. The release includes a pretrained foundation model, a globally balanced instruction-tuned variant, and three region-specialized models targeting languages from Africa, South Asia, Europe, Asia-Pacific, and West Asia. This report details the training strategy, data composition, and comprehensive evaluation framework behind Tiny Aya, and presents an alternative scaling path for multilingual AI: one centered on efficiency, balanced performance across languages, and practical deployment.

2026-03-11

arXiv (prépublication)

arxiv.org

A Coin Flip for Safety: LLM Judges Fail to Reliably Measure Adversarial Robustness

Leo Schwinn

Moritz Ladenburger

Tim Beyer

Gauthier Gidel

Stephan Günnemann

Automated \enquote{LLM-as-a-Judge} frameworks have become the de facto standard for scalable evaluation across natural language processing. … (voir plus)For instance, in safety evaluation, these judges are relied upon to evaluate harmfulness in order to benchmark the robustness of safety against adversarial attacks. However, we show that existing validation protocols fail to account for substantial distribution shifts inherent to red-teaming: diverse victim models exhibit distinct generation styles, attacks distort output patterns, and semantic ambiguity varies significantly across jailbreak scenarios. Through a comprehensive audit using 6642 human-verified labels, we reveal that the unpredictable interaction of these shifts often causes judge performance to degrade to near random chance. This stands in stark contrast to the high human agreement reported in prior work. Crucially, we find that many attacks inflate their success rates by exploiting judge insufficiencies rather than eliciting genuinely harmful content. To enable more reliable evaluation, we propose ReliableBench, a benchmark of behaviors that remain more consistently judgeable, and JudgeStressTest, a dataset designed to expose judge failures. (Data in supplement).

2025-12-31

International Conference on Machine Learning (Accept (regular))

Tight Lower Bounds and Improved Convergence in Performative Prediction

Performative prediction is a framework accounting for the shift in the data distribution induced by the prediction of a model deployed in th… (voir plus)e real world. Ensuring rapid convergence to a stable solution where the data distribution remains the same after the model deployment is crucial, especially in evolving environments. This paper extends the Repeated Risk Minimization (RRM) framework by utilizing historical datasets from previous retraining snapshots, yielding a class of algorithms that we call Affine Risk Minimizers and enabling convergence to a performatively stable point for a broader class of problems. We introduce a new upper bound for methods that use only the final iteration of the dataset and prove for the first time the tightness of both this new bound and the previous existing bounds within the same regime. We also prove that utilizing historical datasets can surpass the lower bound for last iterate RRM, and empirically observe faster convergence to the stable point on various performative prediction benchmarks. We offer at the same time the first lower bound analysis for RRM within the class of Affine Risk Minimizers, quantifying the potential improvements in convergence speed that could be achieved with other variants in our framework.

2025-09-17

NeurIPS.cc/2025/Conference (poster)

Performative Prediction on Games and Mechanism Design

Fernando P. Santos

2025-04-22

Proceedings of The 28th International Conference on Artificial Intelligence and Statistics (publié)

proceedings.mlr.press

A Generative Approach to LLM Harmfulness Detection with Red Flag Tokens

Most safety training methods for large-language models (LLMs) based on fine-tuning rely on dramatically changing the output distribution of … (voir plus)the model when faced with a harmful request, shifting it from an unsafe answer to a refusal to respond. These methods inherently compromise model capabilities and might make auto-regressive models vulnerable to attacks that make likely an initial token of affirmative response. To avoid that, we propose to expand the model's vocabulary with a special token we call a *red flag token* (

2025-03-04

ICLR.cc/2025/Workshop/BuildingTrust (accepté)

A Generative Approach to LLM Harmfulness Mitigation with Red Flag Tokens

2024-12-31

arXiv.org (prépublication)

Performance Control in Early Exiting to Deploy Large Models at the Same Cost of Smaller Ones

Reza Bayat

Ioannis Mitliagkas

Joao Monteiro

Valentina Zantedeschi

2024-06-19

ICML.cc/2024/Workshop/ES-FoMo-II (poster)

Performative Prediction with Neural Networks

Ioannis Mitliagkas

Gauthier Gidel

Performative prediction is a framework for learning models that influence the data they intend to predict. We focus on finding classifiers t… (voir plus)hat are performatively stable, i.e. optimal for the data distribution they induce. Standard convergence results for finding a performatively stable classifier with the method of repeated risk minimization assume that the data distribution is Lipschitz continuous to the model's parameters. Under this assumption, the loss must be strongly convex and smooth in these parameters; otherwise, the method will diverge for some problems. In this work, we instead assume that the data distribution is Lipschitz continuous with respect to the model's predictions, a more natural assumption for performative systems. As a result, we are able to significantly relax the assumptions on the loss function. In particular, we do not need to assume convexity with respect to the model's parameters. As an illustration, we introduce a resampling procedure that models realistic distribution shifts and show that it satisfies our assumptions. We support our theory by showing that one can learn performatively stable classifiers with neural networks making predictions about real data that shift according to our proposed procedure.

2022-12-31

AISTATS (publié)