Home » Teams » MLIA » Alexandre Rame
logo ISIR
  • Alexandre Rame

  • PhD student
  • Team: MLIA

Publications

  • Mustafa Shukor, Alexandre Rame, Corentin Dancette, Matthieu Cord. Beyond Task Performance: Evaluating and Reducing the Flaws of Large Multimodal Models with In-Context Learning. The Twelfth International Conference on Learning Representations (ICLR), May 2024, Vienna, Austria. ⟨hal-04505149⟩
  • Mustafa Shukor, Corentin Dancette, Alexandre Rame, Matthieu Cord. UnIVAL: Unified Model for Image, Video, Audio and Language Tasks. Transactions on Machine Learning Research Journal, 2023. ⟨hal-04366059⟩
  • Alexandre Rame. Diverse and efficient ensembling of deep networks. Artificial Intelligence [cs.AI]. Sorbonne Université, 2023. English. ⟨NNT : 2023SORUS587⟩. ⟨tel-04485963⟩
  • Alexandre Rame, Matthieu Kirchmeyer, Thibaud Rahier, Alain Rakotomamonjy, Patrick Gallinari, et al.. Diverse Weight Averaging for Out-of-Distribution Generalization. Conference on Neural Information Processing Systems, Nov 2022, New-Orleans, United States. ⟨hal-03891267⟩
  • Alexandre Ramé, Corentin Dancette, Matthieu Cord. Fishr: Invariant Gradient Variances for Out-of-Distribution Generalization. ICML, Jul 2022, Baltimore, United States. ⟨hal-03951454⟩
  • Matthieu Cord, Arthur Douillard, Alexandre Ramé, Guillaume Couairon. DyTox: Transformers for Continual Learning with DYnamic TOken eXpansion. IEEE/CVF Conference on Computer Vision and Pattern Recognition, Jun 2022, New Orleans, United States. ⟨10.1109/CVPR52688.2022.00907⟩. ⟨hal-03997873⟩
  • Rémy Sun, Alexandre Ramé, Clément Masson, Nicolas Thome, Matthieu Cord. Towards efficient feature sharing in MIMO architectures. 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Jun 2022, New Orleans, United States. pp.2696-2700, ⟨10.1109/CVPRW56347.2022.00303⟩. ⟨hal-03951737⟩
  • Alexandre Rame, Corentin Dancette, Matthieu Cord. Fishr: Invariant Gradient Variances for Out-of-Distribution Generalization. 39th International Conference on Machine Learning (ICML 2022), Apr 2022, Baltimore, MD, United States. pp.18347--18377. ⟨hal-03944846⟩