Publications

Also on Google Scholar. * denotes equal contribution.

Preprints

KV-Kaizen: Learning Context-Adaptive Cache Compression Choices
J. Monteiro, L. Béthune, A. Filippova, S. Laguna, D. Grangier, M. Cuturi
arXiv preprint, 2026 · arXiv

On Trajectory-Aware Training for Masked Diffusion Language Models
M. Madeira, A. Shidani, A. Bizeul, V. Turrisi, L. Béthune, B. Devnani, D. Busbridge, P. Ablin, J. Monteiro
arXiv preprint, 2026 · arXiv

Cartridges++: KV Cache Compression without Off-Context Derailment
S. Laguna, J. Monteiro, M. Cuturi, P. Ablin, E. Gualdoni
arXiv preprint, 2026 · arXiv

KV-Lingo: Learning KV-Cache Translators with Distillation
V. Castin*, K. Sakamoto*, A. Filippova, J. Monteiro, M. Cuturi, P. Ablin
arXiv preprint, 2026 · arXiv

DynaMiCS: Fine-tuning LLMs with Performance Constraints using Dynamic Mixtures
E. Gualdoni, S. Laguna, L. Béthune, J. Monteiro, P. Ablin, M. Cuturi
arXiv preprint, 2026 · arXiv

Nectar: Neural Estimation of Cached-Token Attention via Regression
J. Monteiro, M. Klein, P. Ablin, M. Cuturi
arXiv preprint, 2026 · arXiv

Stochastic KV Routing: Enabling Adaptive Depth-Wise Cache Sharing
A. Filippova, D. Grangier, M. Cuturi, J. Monteiro
arXiv preprint, 2026 · arXiv

LaCy: What Small Language Models Can and Should Learn is Not Just a Question of Loss
S. Ujváry, L. Béthune, P. Ablin, J. Monteiro, M. Cuturi, M. Kirchhof
arXiv preprint, 2026 · arXiv

PairBench: Are Vision-Language Models Reliable at Comparing What They See?
A. Feizi, S. Rajeswar, A. Romero-Soriano, R. Rabbany, V. Zantedeschi, S. Gella, J. Monteiro
arXiv preprint, 2025 · arXiv

Conference and journal papers

2026

The Design Space of Tri-Modal Masked Diffusion Models
L. Béthune, V. Turrisi, B. K. Mlodozeniec, et al. (24 authors, including J. Monteiro)
NeurIPS 2026 (spotlight) · arXiv

Amortizing Maximum Inner Product Search with Learned Support Functions
T. X. Olausson, J. Monteiro, M. Klein, M. Cuturi
ICML 2026 · arXiv · code

Learning Unmasking Policies for Diffusion Language Models
M. Jazbec, T. X. Olausson, L. Béthune, P. Ablin, M. Kirchhof, J. Monteiro, V. Turrisi, J. Ramapuram, M. Cuturi
ICML 2026 (oral) · arXiv

GraphOmni: A Comprehensive and Extensible Benchmark Framework for Large Language Models on Graph-theoretic Tasks
H. Xu, X. Jian, X. Zhao, W. Pang, C. Zhang, S. Wang, Q. Zhang, Z. Dong, J. Monteiro, B. Liu, Q. Sun, T. Yu
ICLR 2026 · arXiv

2025

BiXSE: Improving Dense Retrieval via Probabilistic Graded Relevance Distillation
C. Tsirigotis, V. Adlakha, J. Monteiro, A. Courville, P. Taslakian
COLM 2025 · arXiv

BigDocs: An Open and Permissively-Licensed Dataset for Training Multimodal Models on Document and Code Tasks
J. Rodriguez, X. Jian, S. S. Panigrahi, et al. (43 authors, including J. Monteiro)
ICLR 2025 · arXiv

2024

RepLiQA: A Question-Answering Dataset for Benchmarking LLMs on Unseen Reference Content
J. Monteiro, P. A. Noël, E. Marcotte, S. Rajeswar, V. Zantedeschi, D. Vázquez, N. Chapados, C. Pal, P. Taslakian
NeurIPS 2024 · arXiv · data

Expecting The Unexpected: Towards Broad Out-Of-Distribution Detection
C. Guille-Escuret, P. A. Noël, D. Vázquez, I. Mitliagkas, J. Monteiro
NeurIPS 2024 · arXiv

XC-Cache: Cross-Attending to Cached Context for Efficient LLM Inference
J. Monteiro, E. Marcotte, P. A. Noël, V. Zantedeschi, D. Vázquez, N. Chapados, C. Pal, P. Taslakian
Findings of EMNLP 2024 · arXiv · code

2023

Group Robust Classification Without Any Group Information
C. Tsirigotis, J. Monteiro, P. Rodríguez, D. Vázquez, A. Courville
NeurIPS 2023 · arXiv

CADet: Fully Self-Supervised Out-Of-Distribution Detection With Contrastive Learning
C. Guille-Escuret, P. Rodríguez, D. Vázquez, I. Mitliagkas, J. Monteiro
NeurIPS 2023 · arXiv

StarCoder: may the source be with you!
R. Li, L. Ben Allal, Y. Zi, et al. (67 authors, including J. Monteiro)
Transactions on Machine Learning Research (TMLR), 2023 · arXiv

Constraining Representations Yields Models That Know What They Don’t Know
J. Monteiro, P. Rodríguez, P. A. Noël, I. Laradji, D. Vázquez
ICLR 2023 · arXiv

2022

Monotonicity Regularization: Improved Penalties and Novel Applications to Disentangled Representation Learning and Robust Classification
J. Monteiro, M. O. Ahmed, H. Hajimirsadeghi, G. Mori
UAI 2022 · paper · arXiv

Estimating Distribution Shifts for Predicting Cross-Subject Generalization in Electroencephalography-based Mental Workload Assessment
I. Albuquerque, J. Monteiro, O. Rosanne, T. Falk
Frontiers in Artificial Intelligence, 2022 · paper

2021

Single-shot real-time compressed ultrahigh-speed imaging enabled by a snapshot-to-video autoencoder
J. Monteiro*, X. Liu*, I. Albuquerque, Y. Lai, C. Jiang, S. Zhang, T. Falk, J. Liang
Photonics Research, 2021

2020

An end-to-end approach for the verification problem: learning the right distance
J. Monteiro, I. Albuquerque, J. Alam, R. D. Hjelm, T. Falk
ICML 2020 · arXiv · code

An Ensemble Based Approach for Generalized Detection of Spoofing Attacks to Automatic Speaker Recognizers
J. Monteiro, J. Alam, T. Falk
ICASSP 2020

Multi-task self-supervised learning for Robust Speech Recognition
M. Ravanelli, J. Zhong, S. Pascual, P. Swietojanski, J. Monteiro, J. Trmal, Y. Bengio
ICASSP 2020 · arXiv

2019

Multi-objective training of Generative Adversarial Networks with multiple discriminators
J. Monteiro*, I. Albuquerque*, T. Doan, B. Considine, T. Falk, I. Mitliagkas
ICML 2019 · arXiv

Online Adaptative Curriculum Learning for GANs
T. Doan, J. Monteiro, I. Albuquerque, B. Mazoure, A. Durand, J. Pineau, R. D. Hjelm
AAAI 2019 · arXiv

Workshop papers

Performance Control in Early Exiting to Deploy Large Models at the Same Cost of Smaller Ones
M. Mofakhami, R. Bayat, I. Mitliagkas, J. Monteiro, V. Zantedeschi
Workshop on Efficient Systems for Foundation Models (ES-FoMo II), ICML 2024 · arXiv

Multilingual Code Retrieval Without Paired Data: A New Benchmark and Experiments
J. Monteiro, T. Scholak, V. Mehta, D. Vázquez, C. Pal
Deep Learning for Code workshop, ICLR 2023 · paper

Not Too Close and Not Too Far: Enforcing Monotonicity Requires Penalizing The Right Points
J. Monteiro, M. O. Ahmed, H. Hajimirsadeghi, G. Mori
eXplainable AI approaches for debugging and diagnosis workshop, NeurIPS 2021 · paper

Domain Conditional Predictors for Domain Adaptation
J. Monteiro, X. Gibert, J. Feng, V. Dumoulin, D. S. Lee
Pre-registration workshop, NeurIPS 2020 · arXiv

Generalizing to unseen domains via distribution matching
I. Albuquerque, J. Monteiro, M. Darvishi, T. Falk, I. Mitliagkas
Uncertainty and Robustness in Deep Learning workshop, ICML 2020 · arXiv