activity
20242026
collaborators

6 papers

cs.LG2026

The risk of KV cache compression

Lukas Haverbeck, Carmen Amo Alonso, Andres Felipe Posada-Moreno +2

Transformer inference on long sequences is expensive because softmax attention repeatedly reads from a large KV cache. The prevalent approach to this bottleneck is KV cache compres…

cs.LG2026

Gated Normalization Removal and Scale Anchoring in Pre-Norm Transformers

Andrei Kanavalau, Carmen Amo Alonso, Sanjay Lall

Normalization layers are standard in transformers, but it is not clear whether their sample-dependent computations are necessary throughout both training and inference. This work d…

cs.LG2026

Design Principles for Sequence Models via Coefficient Dynamics

Jerome Sieber, Antonio Orvieto, Melanie N. Zeilinger +1

Deep sequence models, ranging from Transformers and State Space Models (SSMs) to more recent approaches such as gated linear RNNs, fundamentally compute outputs as linear combinati…

cs.NE2026

Bridging Expressivity and Scalability with Adaptive Unitary SSMs

Arjun Karuvally, Franz Nowak, Anderson T. Keller +3

Recent work has revealed that state space models (SSMs), while efficient for long-sequence processing, are fundamentally limited in their ability to represent formal languages-part…

cs.LG2025

Lambda-Skip Connections: the architectural component that prevents Rank Collapse

Federico Arangath Joseph, Jerome Sieber, Melanie N. Zeilinger +1

Rank collapse, a phenomenon where embedding vectors in sequence models rapidly converge to a uniform token or equilibrium state, has recently gained attention in the deep learning…

cs.LG2024

Understanding the differences in Foundation Models: Attention, State Space Models, and Recurrent Neural Networks

Jerome Sieber, Carmen Amo Alonso, Alexandre Didier +2

Softmax attention is the principle backbone of foundation models for various artificial intelligence applications, yet its quadratic complexity in sequence length can limit its inf…