Publications

Learning diverse attacks on large language models for robust red-teaming and safety tuning
Seanie Lee
Minsu Kim
Lynn Cherif
David Dobre
Juho Lee
Sung Ju Hwang
Kenji Kawaguchi
Gauthier Gidel
Yoshua Bengio
Nikolay Malkin
Moksh J. Jain
Red-teaming, or identifying prompts that elicit harmful responses, is a critical step in ensuring the safe and responsible deployment of lar… (see more)ge language models (LLMs). Developing effective protection against many modes of attack prompts requires discovering diverse attacks. Automated red-teaming typically uses reinforcement learning to fine-tune an attacker language model to generate prompts that elicit undesirable responses from a target LLM, as measured, for example, by an auxiliary toxicity classifier. We show that even with explicit regularization to favor novelty and diversity, existing approaches suffer from mode collapse or fail to generate effective attacks. As a flexible and probabilistically principled alternative, we propose to use GFlowNet fine-tuning, followed by a secondary smoothing phase, to train the attacker model to generate diverse and effective attack prompts. We find that the attacks generated by our method are effective against a wide range of target LLMs, both with and without safety tuning, and transfer well between target LLMs. Finally, we demonstrate that models safety-tuned using a dataset of red-teaming prompts generated by our method are robust to attacks from other RL-based red-teaming approaches.
2024-05-28
ArXiv (preprint)
doi.org
arxiv.org
MODL: Multilearner Online Deep Learning
Antonios Valkanas
Boris Oreshkin
Mark Coates
Online deep learning solves the problem of learning from streams of data, reconciling two opposing objectives: learn fast and learn deep. Ex… (see more)isting work focuses almost exclusively on exploring pure deep learning solutions, which are much better suited to handle the"deep"than the"fast"part of the online learning equation. In our work, we propose a different paradigm, based on a hybrid multilearner approach. First, we develop a fast online logistic regression learner. This learner does not rely on backpropagation. Instead, it uses closed form recursive updates of model parameters, handling the fast learning part of the online learning problem. We then analyze the existing online deep learning theory and show that the widespread ODL approach, currently operating at complexity
2024-05-28
ArXiv (preprint)
doi.org
arxiv.org
Structured Learning in Time-dependent Cox Models
Guanbo Wang
Yi Lian
Archer Yang
Robert W. Platt
Rui Wang
Sylvie Perreault
Marc Dorais
Mireille E. Schnitzer
2024-05-28
Statistics in Medicine (published)
doi.org
arxiv.org
The Cost of Arbitrariness for Individuals: Examining the Legal and Technical Challenges of Model Multiplicity
Prakhar Ganesh
Ihsan Ibrahim Daldaban
Ignacio Cofone
Golnoosh Farnadi
Model multiplicity, the phenomenon where multiple models achieve similar performance despite different underlying learned functions, introdu… (see more)ces arbitrariness in model selection. While this arbitrariness may seem inconsequential in expectation, its impact on individuals can be severe. This paper explores various individual concerns stemming from multiplicity, including the effects of arbitrariness beyond final predictions, disparate arbitrariness for individuals belonging to protected groups, and the challenges associated with the arbitrariness of a single algorithmic system creating a monopoly across various contexts. It provides both an empirical examination of these concerns and a comprehensive analysis from the legal standpoint, addressing how these issues are perceived in the anti-discrimination law in Canada. We conclude the discussion with technical challenges in the current landscape of model multiplicity to meet legal requirements and the legal gap between current law and the implications of arbitrariness in model selection, highlighting relevant future research directions for both disciplines.
2024-05-28
ArXiv (preprint)
doi.org
arxiv.org
Towards a Reliable French Speech Recognition Tool for an Automated Diagnosis of Learning Disabilities
Jihene Rezgui
Félix Jobin
Younes Kechout
Chritine Turgeon
Foutse Khomh
Dyslexia, characterized by severe challenges in reading and spelling acquisition, presents a substantial barrier to proficient literacy, res… (see more)ulting in significantly reduced reading speed (2 to 3 times slower) and diminished text comprehension. With a prevalence ranging from 5G to 10% in the population, early intervention by speech and language pathologists (SLPs) can mitigate dyslexia's effects, but the diagnosis bottleneck impedes timely support. To address this, we propose leveraging machine learning tools to expedite the diagnosis process, focusing on automating phonetic transcription, a critical step in dyslexia assessment. We investigated the practicality of two model configurations utilizing Google's speech-to-text API with children speech in evaluation scenarios and compared their results against transcriptions crafted by experts. The first configuration focuses on Google API's speech-to-text while the second integrates Phonemizer, a text-to-phonemes tool based on a dictionary. Results analysis indicate that our Google-Phonemizer model yields reading accuracies comparable to those computed from human-made transcriptions, offering promise for clinical application. These findings underscore the potential of AI-driven solutions to enhance dyslexia diagnosis efficiency, paving the way for improved accessibility to vital SLP services.
2024-05-28
2024 International Conference on Smart Applications, Communications and Networking (SmartNets) (published)
doi.org
Understanding Intrinsic Socioeconomic Biases in Large Language Models
Mina Arzaghi
Florian Carichon
Golnoosh Farnadi
Large Language Models (LLMs) are increasingly integrated into critical decision-making processes, such as loan approvals and visa applicatio… (see more)ns, where inherent biases can lead to discriminatory outcomes. In this paper, we examine the nuanced relationship between demographic attributes and socioeconomic biases in LLMs, a crucial yet understudied area of fairness in LLMs. We introduce a novel dataset of one million English sentences to systematically quantify socioeconomic biases across various demographic groups. Our findings reveal pervasive socioeconomic biases in both established models such as GPT-2 and state-of-the-art models like Llama 2 and Falcon. We demonstrate that these biases are significantly amplified when considering intersectionality, with LLMs exhibiting a remarkable capacity to extract multiple demographic attributes from names and then correlate them with specific socioeconomic biases. This research highlights the urgent necessity for proactive and robust bias mitigation techniques to safeguard against discriminatory outcomes when deploying these powerful models in critical real-world applications.
2024-05-28
ArXiv (preprint)
doi.org
arxiv.org
Advancing Cultural Inclusivity: Optimizing Embedding Spaces for Balanced Music Recommendations
Armin Moradi
Nicola Neophytou
Golnoosh Farnadi
2024-05-27
ArXiv (preprint)
doi.org
arxiv.org
Estimating Expectations without Sampling: Neural Stein Estimation
Mohsin Hasan
Dinghuai Zhang
Cheikh Ahmed
Awa Khouna
Yoshua Bengio
We propose a method for estimating the expected value of a given function …
2024-05-27
approximateinference.org/AABI/2024/Symposium (accepted)
openreview.net
openreview.net
Implicitly Bayesian Prediction Rules in Deep Learning
Bruno Mlodozeniec
Richard E. Turner
David Scott Krueger
The Bayesian approach leads to coherent updates of predictions under new data, which makes adhering to Bayesian principles appealing in deci… (see more)sion-making contexts. Traditionally, integrating Bayesian principles into models like deep neural networks involves setting priors on parameters and approximating posteriors. This is done despite the fact that, typically, priors on parameters reflect any prior beliefs only insofar as they dictate function space behaviour. In this paper, we rethink this approach and consider what properties characterise a prediction rule as being Bayesian. Algorithms meeting such criteria can be deemed implicitly Bayesian — they make the same predictions as some Bayesian model, without explicitly manifesting priors and posteriors. We argue this might be a more fruitful approach towards integrating Bayesian principles into deep learning. In this paper, we propose how to measure how close a general prediction rule is to being implicitly Bayesian, and empirically evaluate multiple prediction strategies using our approach. We also show theoretically that agents relying on non-implicitly Bayesian prediction rules can be easily exploited in adversarial betting settings.
2024-05-27
approximateinference.org/AABI/2024/Symposium_Archival_Track (accepted)
proceedings.mlr.press
openreview.net
An Introduction to Vision-Language Modeling
Florian Bordes
Richard Yuanzhe Pang
Anurag Ajay
Alexander C. Li
Adrien Bardes
Suzanne Petryk
Oscar Mañas
Zhiqiu Lin
Anas Mahmoud
Bargav Jayaraman
Mark Ibrahim
Melissa Hall
Yunyang Xiong
Jonathan Lebensold
Candace Ross
Srihari Jayakumar
Chuan Guo
Diane Bouchacourt
Haider Al-Tahan
Karthik Padthe … (see 21 more)
Vasu Sharma
Huijuan Xu 0001
Xiaoqing Ellen Tan
Megan Richards
Samuel Lavoie
Pietro Astolfi
Reyhane Askari Hemmat
Jun Chen
Kushal Tirumala
Rim Assouel
Mazda Moayeri
Arjang Talattof
Kamalika Chaudhuri
Zechun Liu
Xilun Chen
Quentin Garrido
Karen Ullrich
Aishwarya Agrawal
Kate Saenko
Asli Celikyilmaz
Vikas Chandra
2024-05-27
ArXiv (preprint)
doi.org
arxiv.org
Neuro-GPT: Towards A Foundation Model for EEG
Wenhui Cui
Woojae Jeong
Philipp Thölke
Takfarinas Medani
Karim Jerbi
Anand A. Joshi
Richard M. Leahy
To handle the scarcity and heterogeneity of electroencephalography (EEG) data for Brain-Computer Interface (BCI) tasks, and to harness the p… (see more)ower of large publicly available data sets, we propose Neuro-GPT, a foundation model consisting of an EEG encoder and a GPT model. The foundation model is pre-trained on a large-scale data set using a self-supervised task that learns how to reconstruct masked EEG segments. We then fine-tune the model on a Motor Imagery Classification task to validate its performance in a low-data regime (9 subjects). Our experiments demonstrate that applying a foundation model can significantly improve classification performance compared to a model trained from scratch, which provides evidence for the generalizability of the foundation model and its ability to address challenges of data scarcity and heterogeneity in EEG. The code is publicly available at github.com/wenhui0206/NeuroGPT.
2024-05-27
2024 IEEE International Symposium on Biomedical Imaging (ISBI) (published)
doi.org
arxiv.org
Partial Models for Building Adaptive Model-Based Reinforcement Learning Agents
Safa Alver
Ali Rahimi-Kalahroudi
Doina Precup
In neuroscience, one of the key behavioral tests for determining whether a subject of study exhibits model-based behavior is to study its ad… (see more)aptiveness to local changes in the environment. In reinforcement learning, however, recent studies have shown that modern model-based agents display poor adaptivity to such changes. The main reason for this is that modern agents are typically designed to improve sample efficiency in single task settings and thus do not take into account the challenges that can arise in other settings. In local adaptation settings, one particularly important challenge is in quickly building and maintaining a sufficiently accurate model after a local change. This is challenging for deep model-based agents as their models and replay buffers are monolithic structures lacking distribution shift handling capabilities. In this study, we show that the conceptually simple idea of partial models can allow deep model-based agents to overcome this challenge and thus allow for building locally adaptive model-based agents. By modeling the different parts of the state space through different models, the agent can not only maintain a model that is accurate across the state space, but it can also quickly adapt it in the presence of a local change in the environment. We demonstrate this by showing that the use of partial models in agents such as deep Dyna-Q, PlaNet and Dreamer can allow for them to effectively adapt to the local changes in their environments.
2024-05-27
ArXiv (preprint)
doi.org
arxiv.org