EY
Edouard Yvinec
cs.CVcs.CLcs.LGcs.AIArtificial Intelligenceeess.IV
On Valency
published · living versionsW_vu5dc8qp·v1 · currentpublished
EmbeddingGemma 2: Unified and Efficient Multimodal Representations
with Henrique Schechter Vera, Sahil Dua, Jovan Andonov, Sara Smoot +68
1 version
Preprints & journals
22 papers in the corpus · 2020–2026An open vision-language model for diverse medical applications.42839104 · Sellergren, Andrew, Kazemzadeh, Sahar, Mahvar, Fereshteh et al.2026 · 46 citationsNature medicine. 2026;32(10):3695-3704
Gemma 4 Technical Report2607.02770v2 · Gemma Team: Sherif El Abd, Vaibhav Aggarwal, Robin Algayres et al.2026 · 8 citationsarXiv
Decoupled DiLoCo for Resilient Distributed Pre-training2604.21428v1 · Arthur Douillard, Keith Rush, Yani Donchev et al.2026 · 0 citationsarXiv
NUPES: Non-Uniform Post-Training Quantization via Power Exponent Search.40758517 · Yvinec, Edouard, Dapogny, Arnaud, Bailly, Kevin2025 · 2 citationsIEEE transactions on pattern analysis and machine intelligence. 2025;47(11):10012-10021
Gemma 3 Technical Report2503.19786v1 · Gemma Team: Aishwarya Kamath, Johan Ferret, Shreya Pathak et al.2025 · 72 citationsarXiv
SPOT: Text Source Prediction from Originality Score Thresholding2405.20505v1 · Edouard Yvinec, Gabriel Kasser2024 · 0 citationsarXiv
PIPE : Parallelized Inference Through Post-Training Quantization Ensembling of Residual Expansions2311.15806v1 · Edouard Yvinec, Arnaud Dapogny, Kevin Bailly2023 · 0 citationsarXiv
Archtree: on-the-fly tree-structured exploration for latency-aware pruning of deep neural networks2311.10549v1 · R'emi Ouazan Reboul, Edouard Yvinec, Arnaud Dapogny et al.2023 · 0 citationsarXiv
Network Memory Footprint Compression Through Jointly Learnable Codebooks and Mappings2309.17361v1 · Edouard Yvinec, Arnaud Dapogny, Kevin Bailly2023 · 0 citationsarXiv
Gradient-Based Post-Training Quantization: Challenging the Status Quo2308.07662v1 · Edouard Yvinec, Arnaud Dapogny, Kevin Bailly2023 · 0 citationsarXiv
SAfER: Layer-Level Sensitivity Assessment for Efficient and Robust Neural Network Inference2308.04753v2 · Edouard Yvinec, Arnaud Dapogny, Kevin Bailly et al.2023 · 1 citationarXiv
Designing strong baselines for ternary neural network quantization through support and mass equalization2306.17442v1 · Edouard Yvinec, Arnaud Dapogny, Kevin Bailly2023 · 1 citationICIP 2023
Fighting over-fitting with quantization for learning deep neural networks on noisy labels2303.11803v1 · Gauthier Tallec, Edouard Yvinec, Arnaud Dapogny et al.2023 · 6 citationsarXiv
RED++ : Data-Free Pruning of Deep Neural Networks via Input Splitting and Output Merging.35653454 · Yvinec, Edouard, Dapogny, Arnaud, Cord, Matthieu et al.2023 · 22 citationsIEEE transactions on pattern analysis and machine intelligence. 2023;45(3):3664-3676
PowerQuant: Automorphism Search for Non-Uniform Quantization2301.09858v1 · Edouard Yvinec, Arnaud Dapogny, Matthieu Cord et al.2023 · 6 citationsarXiv
SInGE: Sparsity via Integrated Gradients Estimation of Neuron Relevance2207.04089v1 · Edouard Yvinec, Arnaud Dapogny, Matthieu Cord et al.2022 · 4 citationsarXiv
SPIQ: Data-Free Per-Channel Static Input Quantization2203.14642v1 · Edouard Yvinec, Arnaud Dapogny, Matthieu Cord et al.2022 · 17 citationsarXiv
REx: Data-Free Residual Quantization Error Expansion2203.14645v3 · Edouard Yvinec, Arnaud Dapgony, Matthieu Cord et al.2022 · 7 citationsarXiv
To Fold or Not to Fold: a Necessary and Sufficient Condition on Batch-Normalization Layers Folding2203.14646v1 · Edouard Yvinec, Arnaud Dapogny, Kevin Bailly2022 · 11 citationsarXiv
Multi-label Transformer for Action Unit Detection2203.12531v3 · Gauthier Tallec, Edouard Yvinec, Arnaud Dapogny et al.2022 · 7 citationsarXiv
RED : Looking for Redundancies for Data-Free Structured Compression of Deep Neural Networks2105.14797v1 · Edouard Yvinec, Arnaud Dapogny, Matthieu Cord et al.2021 · 16 citationsarXiv
DeeSCo: Deep heterogeneous ensemble with Stochastic Combinatory loss for gaze estimation2004.07098v1 · Edouard Yvinec, Arnaud Dapogny, K'evin Bailly2020 · 2 citationsarXiv
Career total: 30 works. 22 are in this corpus.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.