Yassir Fathullah

Gemma 4 Technical Report

Gemma Team, Sherif El Abd, Vaibhav Aggarwal, Robin Algayres et autres

We introduce Gemma 4, a new generation of open-weight, natively multimodal language models in the Gemma model family. Designed to advance compute efficiency and reasoning, the Gemma 4 model suite features dense and Mixture-of-Experts architectures, ranging from 2.3B to 31B parameters. Alongside …

Efficient Uncertainty Estimation and Sequence Modelling

Yassir Fathullah

Transformer-based autoregressive sequence models have revolutionised natural language processing and speech processing, achieving state-of-the-art performance on a wide range of tasks. However, their deployment in real-world scenarios, especially safety-critical applications like autonomous systems or medical diagnosis, necessitates not only high accuracy but …

Generalised Probabilistic Modelling and Improved Uncertainty Estimation in Comparative LLM-as-a-judge

Yassir Fathullah, Mark Gales

This paper explores generalised probabilistic modelling and uncertainty estimation in comparative LLM-as-a-judge frameworks. We show that existing Product-of-Experts methods are specific cases of a broader framework, enabling diverse modelling options. Furthermore, we propose improved uncertainty estimates for individual comparisons, enabling more efficient …

Cross-Lingual Transfer Learning for Speech Translation

Rao Ma, Mengjie Qian, Yassir Fathullah, Siyuan Tang et autres

Rao Ma, Mengjie Qian, Yassir Fathullah, Siyuan Tang, Mark Gales, Kate Knill. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 2: Short Papers). 2025.

gb (code pays fourni par la source)

Cross-Lingual Transfer Learning for Speech Translation

Rao Ma, Mengjie Qian, Yassir Fathullah, Siyuan Tang et autres

There has been increasing interest in building multilingual foundation models for NLP and speech research. This paper examines how to expand the speech translation capability of these models with restricted data. Whisper, a speech foundation model with strong performance on speech recognition …

Cross-Lingual Transfer Learning for Speech Translation

Rao Ma, Mengjie Qian, Yassir Fathullah, Siyuan Tang et autres

There has been increasing interest in building multilingual foundation models for NLP and speech research. This paper examines how to expand the speech translation capability of these models with restricted data. Whisper, a speech foundation model with strong performance on speech recognition …

Efficient LLM Comparative Assessment: a Product of Experts Framework for Pairwise Comparisons

Adian Liusie, Vatsal Raina, Yassir Fathullah, Mark Gales

LLM-as-a-judge approaches are a practical and effective way of assessing a range of text tasks. However, when using pairwise comparisons to rank a set of candidates, the computational cost scales quadratically with the number of candidates, which has practical limitations. This paper …

Efficient Sample-Specific Encoder Perturbations

Yassir Fathullah, Mark Gales

Encoder-decoder foundation models have displayed state-of-the-art performance on a range of autoregressive sequence tasks. This paper proposes a simple and lightweight modification to such systems to control the behaviour according to a specific attribute of interest. This paper proposes a novel inference-efficient …

Teacher-Student Training for Debiasing: General Permutation Debiasing for Large Language Models

Adian Liusie, Yassir Fathullah, Mark Gales

Large Language Models (LLMs) have demonstrated impressive zero-shot capabilities and versatility in NLP tasks, however they sometimes fail to maintain crucial invariances for specific tasks. One example is permutation sensitivity, where LLMs' outputs may significantly vary depending on the order of the …

End-to-End Speech Recognition Contextualization with Large Language Models

Egor Lakomkin, Chunyang Wu, Yassir Fathullah, Ozlem Kalinli et autres

In recent years, Large Language Models (LLMs) have garnered significant attention from the research community due to their exceptional performance and generalization capabilities. In this paper, we introduce a novel method for contextualizing speech recognition models incorporating LLMs. Our approach casts speech …

TODM: Train Once Deploy Many Efficient Supernet-Based RNN-T Compression For On-Device ASR Models

Yuan Shangguan, Haichuan Yang, Danni Li, Chunyang Wu et autres

Automatic Speech Recognition (ASR) models need to be optimized for specific hardware before they can be deployed on devices. This can be done by tuning the model’s hyperparameters or exploring variations in its architecture. Re-training and re-validating models after making these changes …

Prompting Large Language Models with Speech Recognition Abilities

Yassir Fathullah, Chunyang Wu, Egor Lakomkin, Junteng Jia et autres

Large language models (LLMs) have proven themselves highly flexible, able to solve a wide range of generative tasks, such as abstractive summarization and open-ended question answering. In this paper we extend the capabilities of LLM by directly attaching a small audio encoder …

gb (code pays fourni par la source)