5 papers
Improving Recursive Transformers with Mixture of LoRAs
Mohammadmahdi Nouriborji, Morteza Rohanian, Omid Rohanian
Parameter sharing in recursive transformers reduces model size but collapses layer-wise expressivity. We propose Mixture of LoRAs (MoL), a lightweight conditional-computation mecha…
Efficient Interleaved Speech Modeling through Knowledge Distillation
Mohammadmahdi Nouriborji, Morteza Rohanian
Current speech language models exceed the size and latency constraints of many deployment environments. We build compact, expressive speech generation models through layer-aligned…
Optimizing Speech Language Models for Acoustic Consistency
Morteza Rohanian, Michael Krauthammer
We study speech language models that incorporate semantic initialization and planning losses to achieve robust and consistent generation. Our approach initializes speech tokens wit…
Towards Scalable and Cross-Lingual Specialist Language Models for Oncology
Morteza Rohanian, Tarun Mehra, Nicola Miglino +3
Clinical oncology generates vast, unstructured data that often contain inconsistencies, missing information, and ambiguities, making it difficult to extract reliable insights for d…
Uncertainty Modeling in Multimodal Speech Analysis Across the Psychosis Spectrum
Morteza Rohanian, Roya M. Hüppi, Farhad Nooralahzadeh +8
Capturing subtle speech disruptions across the psychosis spectrum is challenging because of the inherent variability in speech patterns. This variability reflects individual differ…