← Search

Matthijs Van keirsbilck

2 accepted papers

2025

Hymba: A Hybrid-head Architecture for Small Language Models

ICLR 2025spotlight

We propose Hymba, a family of small language models featuring a hybrid-head parallel architecture that integrates attention mechanisms and state space models (SSMs) within the same layer, offering parallel and complementary processing of the same inputs. In this hybrid-head module, attention heads p…

Cited by 12SourcePDFScholar
2025

Nemotron-Flash: Towards Latency-Optimal Hybrid Small Language Models

NeurIPS 2025poster

Efficient deployment of small language models (SLMs) is essential for numerous real-world applications with stringent latency constraints.While previous work on SLM design has primarily focused on reducing the number of parameters to achieve parameter-optimal SLMs, parameter efficiency does not nece…

Cited by 0SourceScholar