23 papers
Nectar: Neural Estimation of Cached-Token Attention via Regression
João Monteiro, Michal Klein, Pierre Ablin +1
Evaluating softmax attention over a fixed long context requires reading every cached key-value pair for each new query token. For a given context (a book, a manual, a legal corpus)…
Removing Noise, not Finding Gold: Quality Filtering for Large-Scale Pretraining
Thiziri Nait Saada, Louis Bethune, Michal Klein +3
Large-scale models are pretrained on massive web-crawled datasets containing documents of mixed quality, making data filtering essential. A popular method is Classifier-based Quali…
A Unified Perspective on the Dynamics of Deep Transformers
Valérie Castin, Pierre Ablin, José Antonio Carrillo +1
Transformers, which are state-of-the-art in most machine learning tasks, represent the data as sequences of vectors called tokens. This representation is then exploited by the atte…
Learning Unmasking Policies for Diffusion Language Models
Metod Jazbec, Theo X. Olausson, Louis Béthune +6
Diffusion (Large) Language Models (dLLMs) now match the downstream performance of their autoregressive counterparts on many tasks, while holding the promise of being more efficient…
Balanced LoRA: Removing Parameter Invariance to Accelerate Convergence
Valérie Castin, Kimia Nadjahi, Pierre Ablin +1
Low-Rank Adaptation (LoRA) is the most widely adopted method for fine-tuning large language models. Notably, LoRA is inherently overparameterized: multiple pairs of low-rank factor…
LaCy: What Small Language Models Can and Should Learn is Not Just a Question of Loss
Szilvia Ujváry, Louis Béthune, Pierre Ablin +3
Language models have consistently grown to compress more world knowledge into their parameters, but the knowledge that can be pretrained into them is upper-bounded by their paramet…