Multi-Task Instruction Tuning via Data Scheduling for Low-Resource Arabic SpeechLLMs

2026-01-18 · Hunzalah Hassan Bhatti, Firoj Alam, Shammur Absar Chowdhury arxiv

Audio large language models (LLMs) enable unified speech understanding and generation, but adapting them to linguistically complex and dialect-rich settings such as Arabic-English remains challenging. We present a controlled study of multi-task instruction tuning for an Arabic-centric audio LLM across generative tasks, including automatic speech recognition (ASR) and speech and text summarization, as well as discriminative tasks, including dialect identification (DID) and speech emotion recognition (SER), in a resource-constrained setting. To support end-to-end Arabic speech summarization, we introduce AraMega-SSum, the first Arabic speech summarization dataset designed for training and benchmarking Arabic-centric audio LLMs. We compare four training strategies: (i) Uniform Mixing (UM), (ii) Task-Progressive Curriculum (TPC), (iii) Aligner-Based Diverse Sampling (ADS) for training-time batch construction, and (iv) a two-stage TPC->ADS strategy. Our results reveal a clear efficiency-robustness trade-off. TPC achieves the strongest performance on generative tasks, including ASR and summarization. ADS improves paralinguistic tasks but reduces generative stability when used alone. The two-stage TPC->ADS strategy provides the best overall balance, achieving the strongest DID and SER performance while outperforming large proprietary models such as Gemini-2.5-Pro on discriminative tasks. We will publicly release AraMega-SSum together with all experimental resources to support future research in Arabic speech understanding.

📄 PDF Abstract BibTeX arXiv:2601.12494

Code (0)

등록된 구현이 없습니다.

Tasks

Speech Emotion Recognition Speech Recognition Text Summarization

Teaching According to Talents! Instruction Tuning LLMs with Competence-Aware Curriculum Learning

2025-09-17 · Yangning Li, Tingwei Lu, Yinghui Li, Yankai Chen 외 arxiv

Efficient instruction tuning aims to enhance the ultimate performance of large language models (LLMs) trained on a given instruction dataset. Curriculum learning as a typical data organization strategy has shown prelimin…

MAFIG: Multi-agent Driven Formal Instruction Generation Framework

2026-04-13 · Shixing Zhao, Zheng Si, Pengpeng Ouyang, Zhengqing Hu 외 arxiv

Emergency situations in scheduling systems often trigger local functional failures that undermine system stability and even cause system collapse. Existing methods primarily rely on robust scheduling or reactive scheduli…

Learning Scheduling Algorithms for Data Processing Clusters

2018-10-03 · Hongzi Mao, Malte Schwarzkopf, Shaileshh Bojja Venkatakrishnan, Zili Meng 외

Efficiently scheduling data processing jobs on distributed compute clusters requires complex algorithms. Current systems, however, use simple generalized heuristics and ignore workload characteristics, since developing a…

Reinforcement Learning Reinforcement Learning (RL)Scheduling

Open-Ended Instruction Realization with LLM-Enabled Multi-Planner Scheduling in Autonomous Vehicles

2026-04-09 · Jiawei Liu, Xun Gong, Fen Fang, Muli Yang 외 arxiv

Most Human-Machine Interaction (HMI) research overlooks the maneuvering needs of passengers in autonomous driving (AD). Natural language offers an intuitive interface, yet translating passenger open-ended instructions in…

Autonomous Vehicles Autonomous Driving

Chain-of-Instructions: Compositional Instruction Tuning on Large Language Models

2024-02-18 · Shirley Anugrah Hayati, Taehee Jung, Tristan Bodding-Long, Sudipta Kar 외

Fine-tuning large language models (LLMs) with a collection of large and diverse instructions has improved the model's generalization to different tasks, even for unseen tasks. However, most existing instruction datasets …