7 papers
Normalized Rewards for Preference Optimization
Shawn Im, Federico Danieli, Skyler Seto +2
Direct Alignment Algorithms (DAAs) such as DPO have become a common way to post-train and align LLMs with human preferences. However, DAAs have been observed to over-optimize their…
Locking Pretrained Weights via Deep Low-Rank Residual Distillation
Keitaro Sakamoto, Pierre Ablin, Federico Danieli +1
The quality of open-weight language models has dramatically improved in recent years. Sharing weights greatly facilitates model adoption by enabling their use across diverse hardwa…
Theoretical Limits of Language Model Alignment
Lucas Monteiro Paes, Natalie Mackraz, Barry-John Theobald +1
Language model (LM) alignment improves model outputs to reflect human preferences while preserving the capabilities of the base model. The most common alignment approaches are (i)…
GenCtrl -- A Formal Controllability Toolkit for Generative Models
Emily Cheng, Carmen Amo Alonso, Federico Danieli +4
As generative models become ubiquitous, there is a critical need for fine-grained control over the generation process. Yet, while controlled generation methods from prompting to fi…
ParaRNN: Unlocking Parallel Training of Nonlinear RNNs for Large Language Models
Federico Danieli, Pau Rodriguez, Miguel Sarabia +2
Recurrent Neural Networks (RNNs) laid the foundation for sequence modeling, but their intrinsic sequential nature restricts parallel computation, creating a fundamental barrier to…
Understanding Input Selectivity in Mamba: Impact on Approximation Power, Memorization, and Associative Recall Capacity
Ningyuan Huang, Miguel Sarabia, Abhinav Moudgil +3
State-Space Models (SSMs), and particularly Mamba, have recently emerged as a promising alternative to Transformers. Mamba introduces input selectivity to its SSM layer (S6) and in…