no code implementations • 11 May 2024 • Yong Guan, Xiaozhi Wang, Lei Hou, Juanzi Li, Jeff Pan, Jiaoyan Chen, Freddy Lecue
Existing work mainly focuses on directly modeling the entire document, which cannot effectively handle long-range dependencies and information redundancy.
1 code implementation • 3 Apr 2024 • Jiawei Zhang, Chejian Xu, Yu Gai, Freddy Lecue, Dawn Song, Bo Li
This paper introduces KnowHalu, a novel approach for detecting hallucinations in text generated by large language models (LLMs), utilizing step-wise reasoning, multi-formulation query, multi-form knowledge for factual checking, and fusion-based detection mechanism.
no code implementations • 13 Mar 2024 • Shubham Sharma, Sanghamitra Dutta, Emanuele Albini, Freddy Lecue, Daniele Magazzeni, Manuela Veloso
In this paper, we introduce the problem of feature \emph{reselection}, so that features can be selected with respect to secondary model performance characteristics efficiently even after a feature selection process has been done with respect to a primary objective.
1 code implementation • 22 Feb 2024 • Rui Yang, Boming Yang, Sixun Ouyang, Tianwei She, Aosong Feng, Yuang Jiang, Freddy Lecue, Jinghui Lu, Irene Li
We assess LLMs' zero-shot performance in creating domain-specific concept graphs and introduce TutorQA, a new expert-verified NLP-focused benchmark for scientific graph reasoning and QA.
no code implementations • 29 Jan 2024 • Yong Guan, Freddy Lecue, Jiaoyan Chen, Ru Li, Jeff Z. Pan
Specifically, for concept completeness, we present core concepts of a scene based on knowledge graph, ConceptNet, to gauge the completeness of concepts.
1 code implementation • 18 Jan 2024 • Linxin Song, Yan Cui, Ao Luo, Freddy Lecue, Irene Li
Transformer-based models excel in various natural language processing (NLP) tasks, attracting countless efforts to explain their inner workings.
no code implementations • 23 Nov 2023 • Sikha Pentyala, Shubham Sharma, Sanjay Kariyappa, Freddy Lecue, Daniele Magazzeni
We observe that PrivRecourse can provide paths that are private and realistic.
no code implementations • 9 Nov 2023 • Zikai Xiong, Niccolò Dalmasso, Shubham Sharma, Freddy Lecue, Daniele Magazzeni, Vamsi K. Potluru, Tucker Balch, Manuela Veloso
Data distillation and coresets have emerged as popular approaches to generate a smaller representative set of samples for downstream learning tasks to handle large-scale datasets.
no code implementations • 30 Oct 2023 • Luca Longo, Mario Brcic, Federico Cabitza, Jaesik Choi, Roberto Confalonieri, Javier Del Ser, Riccardo Guidotti, Yoichi Hayashi, Francisco Herrera, Andreas Holzinger, Richard Jiang, Hassan Khosravi, Freddy Lecue, Gianclaudio Malgieri, Andrés Páez, Wojciech Samek, Johannes Schneider, Timo Speith, Simone Stumpf
As systems based on opaque Artificial Intelligence (AI) continue to flourish in diverse real-world applications, understanding these black box models has become paramount.
Explainable artificial intelligence Explainable Artificial Intelligence (XAI)
no code implementations • 15 May 2023 • Ola Ahmad, Nicolas Bereux, Loïc Baret, Vahid Hashemi, Freddy Lecue
The result is task-specific causal explanatory graphs that can audit model behavior and express the actual causes underlying its performance.
no code implementations • 13 Apr 2023 • Imtiaz Masud Ziko, Freddy Lecue, Ismail Ben Ayed
We introduce a simple non-linear embedding adaptation layer, which is fine-tuned on top of fixed pre-trained features for one-shot tasks, improving significantly transductive entropy-based inference for low-shot regimes.
no code implementations • 11 Nov 2022 • Danial Dervovic, Nicolas Marchesotti, Freddy Lecue, Daniele Magazzeni
We introduce a family of interpretable machine learning models, with two broad additions: Linearised Additive Models (LAMs) which replace the ubiquitous logistic link function in General Additive Models (GAMs); and SubscaleHedge, an expert advice algorithm for combining base models trained on subsets of features called subscales.
no code implementations • 30 Sep 2022 • Juliette Mattioli, Agnes Delaborde, Souhaiel Khalfaoui, Freddy Lecue, Henri Sohier, Frederic Jurie
This paper reviews the entire engineering process of trustworthy Machine Learning (ML) algorithms designed to equip critical systems with advanced analytics and decision functions.
no code implementations • 9 Aug 2022 • Krunal Kishor Patel, Guy Desaulniers, Andrea Lodi, Freddy Lecue
A NOtice To AirMen (NOTAM) contains important flight route related information.
no code implementations • 14 Mar 2022 • Ola Ahmad, Freddy Lecue
Some methods proposed the adaptation of CNNs to ultra-wide FoV images by learning deformable kernels.
no code implementations • 20 Dec 2021 • Tom Bewley, Freddy Lecue
The potential of reinforcement learning (RL) to deliver aligned and performant agents is partially bottlenecked by the reward engineering problem.
no code implementations • 29 Sep 2021 • Ola Ahmad, Simon Corbeil, Vahid Hashemi, Freddy Lecue
Finally, we believe that our method is orthogonal to logic-based explanation methods and can be leveraged to improve their explanations.
no code implementations • 19 Feb 2021 • Clément Playout, Ola Ahmad, Freddy Lecue, Farida Cheriet
Finally, we provide an in-depth analysis of the effect of the deformable convolutions, bringing elements of discussion on the behavior of CNN models.
1 code implementation • COLING 2020 • Maryam Ziaeefard, Freddy Lecue
We propose a model that captures the interactions between objects in a visual scene and entities in an external knowledge source.
2 code implementations • Findings of the Association for Computational Linguistics 2020 • Fran{\c{c}}ois Gard{\`e}res, Maryam Ziaeefard, Baptiste Abeloos, Freddy Lecue
Given an image and a question in natural language, ConceptBert requires visual elements of the image and a Knowledge Graph (KG) to infer the correct answer.
no code implementations • 29 Sep 2020 • Nicholas Halliwell, Freddy Lecue
Convolutional neural networks (CNNs) are commonly used for image classification.
1 code implementation • 30 Jun 2020 • Jiaoyan Chen, Freddy Lecue, Yuxia Geng, Jeff Z. Pan, Huajun Chen
Zero-shot learning (ZSL) is a popular research problem that aims at predicting for those classes that have never appeared in the training stage by utilizing the inter-class relationship with some side information.
no code implementations • 25 Jun 2020 • Severin Gsponer, Luca Costabello, Chan Le Van, Sumit Pai, Christophe Gueret, Georgiana Ifrim, Freddy Lecue
Sequence classification is the supervised learning task of building models that predict class labels of unseen sequences of symbols.
no code implementations • 13 Aug 2019 • Xochitl Watts, Freddy Lecue
We introduce a novel technique to find global and local explanations for time series data used in binary classification machine learning systems.
no code implementations • 31 May 2019 • Freddy Lecue, Jiaoyan Chen, Jeff Z. Pan, Huajun Chen
We exploit their semantics to augment transfer learning by dealing with when to transfer with semantic measurements and what to transfer with semantic embeddings.
no code implementations • 13 Nov 2018 • Rory Mc Grath, Luca Costabello, Chan Le Van, Paul Sweeney, Farbod Kamiab, Zhao Shen, Freddy Lecue
Our contribution is two-fold: i) we propose positive counterfactuals, i. e. we adapt counterfactual explanations to also explain accepted loan applications, and ii) we propose two weighting strategies to generate more interpretable counterfactuals.
1 code implementation • 22 Jul 2018 • Jiaoyan Chen, Freddy Lecue, Jeff Z. Pan, Ian Horrocks, Huajun Chen
Machine learning explanation can significantly boost machine learning's application in decision making, but the usability of current methods is limited in human-centric explanation, especially for transfer learning, an important machine learning branch that aims at utilizing knowledge from one learning domain (i. e., a pair of dataset and prediction task) to enhance prediction model training in another learning domain.
no code implementations • 27 May 2018 • Freddy Lecue, Jiewen Wu
The main objective of explanations is to transmit knowledge to humans.
no code implementations • 24 Apr 2017 • Freddy Lecue, Jiaoyan Chen, Jeff Pan, Huajun Chen
Data stream learning has been largely studied for extracting knowledge structures from continuous and rapid data records.