# Aslema at NADI 2026: Data Augmentation for Intent Recognition and Slot Filling

arXiv: https://arxiv.org/abs/2608.18689 · HTML: https://mlsanity.com/paper/2608.18689/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 37 references that are in the map.
- First posted on arXiv 19 Aug 2026.

## Builds on (most cited first, 23 of 23)

- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [Self-Instruct: Aligning Language Models with Self-Generated Instructions](https://mlsanity.com/paper/2212.10560/) (2022)
- [SpeechGPT: Empowering Large Language Models with Intrinsic Cross-Modal Conversational Abilities](https://mlsanity.com/paper/2305.11000/) (2023)
- [SWIFT:A Scalable lightWeight Infrastructure for Fine-Tuning](https://mlsanity.com/paper/2408.05517/) (2024)
- [AIR-Bench: Benchmarking Large Audio-Language Models via Generative Comprehension](https://mlsanity.com/paper/2402.07729/) (2024)
- [VoiceBench: Benchmarking LLM-Based Voice Assistants](https://mlsanity.com/paper/2410.17196/) (2024)
- [SLURP: A Spoken Language Understanding Resource Package](https://mlsanity.com/paper/2011.13205/) (2020)
- [MASSIVE: A 1M-Example Multilingual Natural Language Understanding Dataset with 51 Typologically-Diverse Languages](https://mlsanity.com/paper/2204.08582/) (2022)
- [AudioBench: A Universal Benchmark for Audio Large Language Models](https://mlsanity.com/paper/2406.16020/) (2024)
- [NADI 2024: The Fifth Nuanced Arabic Dialect Identification Shared Task](https://mlsanity.com/paper/2407.04910/) (2024)
- [Speech-MASSIVE: A Multilingual Speech Dataset for SLU and Beyond](https://mlsanity.com/paper/2408.03900/) (2024)
- [Casablanca: Data and Models for Multidialectal Arabic Speech Recognition](https://mlsanity.com/paper/2410.04527/) (2024)
- [VoxCPM2 Technical Report](https://mlsanity.com/paper/2606.06928/) (2026)
- [The Landscape of Arabic Large Language Models (ALLMs): A New Era for Arabic Language Technology](https://mlsanity.com/paper/2506.01340/) (2025)
- [Instruction Data Generation and Unsupervised Adaptation for Speech Language Models](https://mlsanity.com/paper/2406.12946/) (2024)
- [SpokenNativQA: Multilingual Everyday Spoken Queries for LLMs](https://mlsanity.com/paper/2505.19163/) (2025)
- [Munsit at NADI 2025 Shared Task 2: Pushing the Boundaries of Multidialectal Arabic ASR with Weakly Supervised Pretraining and Continual Supervised Fine-tuning](https://mlsanity.com/paper/2508.08912/) (2025)
- [Beyond MCQ: An Open-Ended Arabic Cultural QA Benchmark with Dialect Variants](https://mlsanity.com/paper/2510.24328/) (2025)
- [TEDxTN: A Three-way Speech Translation Corpus for Code-Switched Tunisian Arabic - English](https://mlsanity.com/paper/2511.10780/) (2025)
- [Multi-Task Instruction Tuning via Data Scheduling for Low-Resource Arabic SpeechLLMs](https://mlsanity.com/paper/2601.12494/) (2026)
- [MENASpeechBank: A Reference Voice Bank with Persona-Conditioned Multi-Turn Conversations for AudioLLMs](https://mlsanity.com/paper/2602.07036/) (2026)
- [Multi-turn Conversational AI from Text to Multimodal Interaction: Data, Models, Evaluation, and Open Challenges](https://mlsanity.com/paper/2608.17605/) (2026)

## Built on by (most cited first, 1 of 1)

- [NADI 2026: The Second Multidialectal Arabic Speech Processing Shared Task](https://mlsanity.com/paper/2609.27086/) (2026)

## Abstract (arXiv)

We present Aslema, our system for NADI 2026 Shared Task 5, which consists of two subtasks: intent recognition and slot filling. We evaluate four omni LLMs in a zero-shot setting and compare them with fine-tuned models. Our results show that fine-tuning consistently outperforms zero-shot inference. We further explore synthetic data augmentation by using an LLM to generate culturally grounded Tunisian Derja utterances, followed by voice cloning to generate synthetic speech. Incorporating this synthetic data improves performance on both tasks. Our final submitted system, based on Qwen3-Omni-30B and trained with a mixture of original and synthetic data, achieves 86.8% intent accuracy and 34.7 WER on the devtest split. On the official test set it ranks 1st in slot filling (59.5 CoER) and 4th among 8 teams in intent recognition (66.1% accuracy). We release our experimental scripts and will soon share the synthetic dataset to support further research in this area.