Publications

Multilingual Hallucination Gaps
Cléa Chataigner
Afaf Taïk
Golnoosh Farnadi
Large language models (LLMs) are increasingly used as alternatives to traditional searchengines given their capacity to generate text that r… (voir plus)esembles human language. However, thisshift is concerning, as LLMs often generate hallucinations—misleading or false informationthat appears highly credible. In this study, we explore the phenomenon of hallucinationsacross multiple languages in free-form text generation, focusing on what we call multilingualhallucination gaps. These gaps reflect differences in the frequency of hallucinated answersdepending on the prompt and language used. To quantify such hallucinations, we used theFActScore metric and extended its framework to a multilingual setting. We conductedexperiments using LLMs from the LLaMA, Qwen, and Aya families, generating biographiesin 19 languages and comparing the results to Wikipedia pages. Our results reveal varia-tions in hallucination rates, especially between high- and low-resource languages, raisingimportant questions about LLM multilingual performance and the challenges in evaluatinghallucinations in multilingual free-form text generation.
2023-12-31
Automation, Fabrication and Manufacturing Engineering (publié)
proceedings.mlr.press
proceedings.mlr.press
Multi-reservoir ESN-based prediction strategy for dynamic multi-objective optimization
Cuili Yang
Danlei Wang
Jian Tang
JunFei Qiao
Wen Yu
2023-12-31
Information Sciences (publié)
doi.org
MVP: Minimal Viable Phrase for Long Text Understanding.
Louis Clouatre
Amal Zouaq
A. Chandar
2023-12-31
International Conference on Language Resources and Evaluation (publié)
dblp.uni-trier.de
Do not trust what you trust: Miscalibration in Semi-supervised Learning
Shambhavi Mishra
Balamurali Murugesan
Ismail Ben Ayed
Marco Pedersoli
Jose Dolz
State-of-the-art semi-supervised learning (SSL) approaches rely on highly confident predictions to serve as pseudo-labels that guide the tra… (voir plus)ining on unlabeled samples. An inherent drawback of this strategy stems from the quality of the uncertainty estimates, as pseudo-labels are filtered only based on their degree of uncertainty, regardless of the correctness of their predictions. Thus, assessing and enhancing the uncertainty of network predictions is of paramount importance in the pseudo-labeling process. In this work, we empirically demonstrate that SSL methods based on pseudo-labels are significantly miscalibrated, and formally demonstrate the minimization of the min-entropy, a lower bound of the Shannon entropy, as a potential cause for miscalibration. To alleviate this issue, we integrate a simple penalty term, which enforces the logit distances of the predictions on unlabeled samples to remain low, preventing the network predictions to become overconfident. Comprehensive experiments on a variety of SSL image classification benchmarks demonstrate that the proposed solution systematically improves the calibration performance of relevant SSL models, while also enhancing their discriminative power, being an appealing addition to tackle SSL tasks.
2023-12-31
Trans. Mach. Learn. Res. (publié)
doi.org
arxiv.org
Online Measurement of Dioxin Emission in Solid Waste Incineration Using Fuzzy Broad Learning
Heng Xia
Jian Tang
Wen Yu
JunFei Qiao
Dioxin (DXN) is a persistent organic pollutant produced from municipal solid waste incineration (MSWI) processes. It is a crucial environmen… (voir plus)tal indicator to minimize emission concentration by using optimization control, but it is difficult to monitor in real time. Aiming at online soft-sensing of DXN emission, a novel fuzzy tree broad learning system (FTBLS) is proposed, which includes offline training and online measurement. In the offline training part, weighted k-means is presented to construct a typical sample pool for reduced learning costs of offline and online phases. Moreover, the novel FTBLS, which contains a feature mapping layer, enhance layer, and increment layer, by replacing the fuzzy decision tree with neurons applied to construct the offline model. In the online measurement part, recursive principal component analysis is used to monitor the time-varying characteristic of the MSWI process. To measure DXN emission, offline FTBLS is reused for normal samples; for drift samples, fast incremental learning is used for online updates. A DXN data from the actual MSWI process is employed to prove the usefulness of FTBLS, where the RMSE of training and testing data are 0.0099 and 0.0216, respectively. This result shows that FTBLS can effectively realize DXN online prediction.
2023-12-31
IEEE Transactions on Industrial Informatics (publié)
doi.org
Open-Source Conversational AI with SpeechBrain 1.0
Mirco Ravanelli
Titouan Parcollet
Adel Moumen
Sylvain de Langen
Cem Subakan
Peter Plantinga
Yingzhi Wang
Pooneh Mousavi
Luca Della Libera
Artem Ploujnikov
Francesco Paissan
Davide Borra
Salah Zaiem
Zeyu Zhao
Shucong Zhang
Georgios Karakasidis
Sung-Lin Yeh
Pierre Champion
Aku Rouhe
Rudolf Braun … (voir 13 de plus)
Florian Mai
Juan Zuluaga-Gomez
Seyed Mahed Mousavi
Andreas Nautsch
Ha Nguyen
Xuechen Liu
Sangeet Sagar
Jarod Duret
Salima Mdhaffar
Gaëlle Laperrière
Mickael Rouvier
Renato De Mori
Yannick Estève
SpeechBrain is an open-source Conversational AI toolkit based on PyTorch, focused particularly on speech processing tasks such as speech rec… (voir plus)ognition, speech enhancement, speaker recognition, text-to-speech, and much more. It promotes transparency and replicability by releasing both the pre-trained models and the complete "recipes" of code and algorithms required for training them. This paper presents SpeechBrain 1.0, a significant milestone in the evolution of the toolkit, which now has over 200 recipes for speech, audio, and language processing tasks, and more than 100 models available on Hugging Face. SpeechBrain 1.0 introduces new technologies to support diverse learning modalities, Large Language Model (LLM) integration, and advanced decoding strategies, along with novel models, tasks, and modalities. It also includes a new benchmark repository, offering researchers a unified platform for evaluating models across diverse tasks.
2023-12-31
Journal of Machine Learning Research (publié)
doi.org
arxiv.org
Operational Research: Methods and Applications
Fotios Petropoulos
Gilbert Laporte
Emel Aktas
Sibel A. Alumur
Claudia Archetti
Hayriye Ayhan
Maria Battarra
Julia A. Bennell
Jean-Marie Bourjolly
John E. Boylan
Michele Breton
David Canca
Laurent Charlin
Bo Chen
Cihan Tugrul Cicek
Louis Anthony Cox, Jr
Christine S.M. Currie
Erik Demeulemeester
Li Ding
Stephen M. Disney … (voir 62 de plus)
Matthias Ehrgott
Martin J. Eppler
Gunes Erdogan
Bernard Fortz
L. Alberto Franco
Jens Frische
Salvatore Greco
Amanda J. Gregory
Raimo P. Hamalainen
Willy Herroelen
Mike Hewitt
Jan Holmstrom
John N. Hooker
Tugce Isik
Jill Johnes
Bahar Y. Kara
Ozlem Karsu
Katherine Kent
Charlotte Kohler
Martin Kunc
Yong-Hong Kuo
Judit Lienert
Adam N. Letchford
Janny Leung
Dong Li
Haitao Li
Ivana Ljubic
Andrea Lodi
Sebastian Lozano
Virginie Lurkin
Silvano Martello
Ian G. McHale
Gerald Midgley
John D.W. Morecroft
Akshay Mutha
Ceyda Oguz
Sanja Petrovic
Ulrich Pferschy
Harilaos N. Psaraftis
Sam Rose
Lauri Saarinen
Said Salhi
Jing-Sheng Song
Dimitrios Sotiros
Kathryn E. Stecke
Arne K. Strauss
Istenc Tarhan
Clemens Thielen
Paolo Toth
Greet Vanden Berghe
Christos Vasilakis
Vikrant Vaze
Daniele Vigo
Kai Virtanen
Xun Wang
Rafał Weron
Leroy White
Tom Van Woensel
Mike Yearworth
E. Alper Yıldırım
Georges Zaccour
Xuying Zhao
Throughout its history, Operational Research has evolved to include a variety of methods, models and algorithms that have been applied to a … (voir plus)diverse and wide range of contexts. This encyclopedic article consists of two main sections: methods and applications. The first aims to summarise the up-to-date knowledge and provide an overview of the state-of-the-art methods and key developments in the various subdomains of the field. The second offers a wide-ranging list of areas where Operational Research has been applied. The article is meant to be read in a nonlinear fashion. It should be used as a point of reference or first-port-of-call for a diverse pool of readers: academics, researchers, students, and practitioners. The entries within the methods and applications sections are presented in alphabetical order. The authors dedicate this paper to the 2023 Turkey/Syria earthquake victims. We sincerely hope that advances in OR will play a role towards minimising the pain and suffering caused by this and future catastrophes.
2023-12-31
J. Oper. Res. Soc. (publié)
doi.org
arxiv.org
Optimal Robot Formations: Balancing Range-Based Observability and User-Defined Configurations
Syed Shabbir Ahmed
Mohammed Ayman Shalaby
Jerome Le Ny
James Richard Forbes
This paper introduces a set of customizable and novel cost functions that enable the user to easily specify desirable robot formations, such… (voir plus) as a ``high-coverage'' infrastructure-inspection formation, while maintaining high relative pose estimation accuracy. The overall cost function balances the need for the robots to be close together for good ranging-based relative localization accuracy and the need for the robots to achieve specific tasks, such as minimizing the time taken to inspect a given area. The formations found by minimizing the aggregated cost function are evaluated in a coverage path planning task in simulation and experiment, where the robots localize themselves and unknown landmarks using a simultaneous localization and mapping algorithm based on the extended Kalman filter. Compared to an optimal formation that maximizes ranging-based relative localization accuracy, these formations significantly reduce the time to cover a given area with minimal impact on relative pose estimation accuracy.
2023-12-31
IEEE/RSJ International Conference on Intelligent Robots and Systems (publié)
doi.org
arxiv.org
Parameter-Efficient Transfer Learning of Audio Spectrogram Transformers
Umberto Cappellazzo
Daniele Falavigna
Alessio Brutti
Mirco Ravanaelli
The common modus operandi of fine-tuning large pre-trained Transformer models entails the adaptation of all their parameters (i.e., full fin… (voir plus)e-tuning). While achieving striking results on multiple tasks, this approach becomes unfeasible as the model size and the number of downstream tasks increase. In natural language processing and computer vision, parameter-efficient approaches like prompt-tuning and adapters have emerged as solid alternatives by fine-tuning only a small number of extra parameters, without sacrificing performance accuracy. For audio classification tasks, the Audio Spectrogram Transformer model shows impressive results. However, surprisingly, how to efficiently adapt it to several downstream tasks has not been tackled before. In this paper, we bridge this gap and present a detailed investigation of common parameter-efficient methods, revealing that adapters and LoRA consistently outperform the other methods across four benchmarks. Whereas adapters prove to be more efficient in few-shot learning settings, LoRA turns out to scale better as we increase the number of learnable parameters. We finally carry out ablation studies to find the best configuration for adapters and LoRA.
2023-12-31
MLSP (publié)
doi.org
arxiv.org
Performance reserves in brain-imaging-based phenotype prediction
Marc-Andre Schulz
Danilo Bzdok
Stefan Haufe
John-Dylan Haynes
Kerstin Ritter
This study examines the impact of sample size on predicting cognitive and mental health phenotypes from brain imaging via machine learning. … (voir plus)Our analysis shows a 3- to 9-fold improvement in prediction performance when sample size increases from 1,000 to 1 M participants. However, despite this increase, the data suggest that prediction accuracy remains worryingly low and far from fully exploiting the predictive potential of brain imaging data. Additionally, we find that integrating multiple imaging modalities boosts prediction accuracy, often equivalent to doubling the sample size. Interestingly, the most informative imaging modality often varied with increasing sample size, emphasizing the need to consider multiple modalities. Despite significant performance reserves for phenotype prediction, achieving substantial improvements may necessitate prohibitively large sample sizes, thus casting doubt on the practical or clinical utility of machine learning in some areas of neuroimaging.
2023-12-31
Cell Reports (publié)
doi.org
Phoneme Discretized Saliency Maps for Explainable Detection of AI-Generated Voice
Shubham Gupta
Mirco Ravanelli
Pascal Germain
Cem Subakan
In this paper, we propose Phoneme Discretized Saliency Maps (PDSM), a discretization algorithm for saliency maps that takes advantage of pho… (voir plus)neme boundaries for explainable detection of AI-generated voice. We experimentally show with two different Text-to-Speech systems (i.e., Tacotron2 and Fastspeech2) that the proposed algorithm produces saliency maps that result in more faithful explanations compared to standard posthoc explanation methods. Moreover, by associating the saliency maps to the phoneme representations, this methodology generates explanations that tend to be more understandable than standard saliency maps on magnitude spectrograms.
2023-12-31
Conference of the International Speech Communication Association (publié)
doi.org
arxiv.org
PID Accelerated Temporal Difference Algorithms
Mark Bedaywi
Amin Rakhsha
Amir-massoud Farahmand
Long-horizon tasks, which have a large discount factor, pose a challenge for most conventional reinforcement learning (RL) algorithms. Algor… (voir plus)ithms such as Value Iteration and Temporal Difference (TD) learning have a slow convergence rate and become inefficient in these tasks. When the transition distributions are given, PID VI was recently introduced to accelerate the convergence of Value Iteration using ideas from control theory. Inspired by this, we introduce PID TD Learning and PID Q-Learning algorithms for the RL setting, in which only samples from the environment are available. We give a theoretical analysis of the convergence of PID TD Learning and its acceleration compared to the conventional TD Learning. We also introduce a method for adapting PID gains in the presence of noise and empirically verify its effectiveness.
2023-12-31
RLJ (publié)
doi.org
arxiv.org