# Enhancing Generalization of Speech Large Language Models with Multi-Task Behavior Imitation and Speech-Text Interleaving

arXiv: https://arxiv.org/abs/2505.18644 · HTML: https://mlsanity.com/paper/2505.18644/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 30 references that are in the map.
- First posted on arXiv 24 May 2025.

## Builds on (most cited first, 26 of 26)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing](https://mlsanity.com/paper/2110.13900/) (2021)
- [Common Voice: A Massively-Multilingual Speech Corpus](https://mlsanity.com/paper/1912.06670/) (2019)
- [Instruction-Following Evaluation for Large Language Models](https://mlsanity.com/paper/2311.07911/) (2023)
- [LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention](https://mlsanity.com/paper/2303.16199/) (2023)
- [Qwen-Audio: Advancing Universal Audio Understanding via Unified Large-Scale Audio-Language Models](https://mlsanity.com/paper/2311.07919/) (2023)
- [Qwen2-Audio Technical Report](https://mlsanity.com/paper/2407.10759/) (2024)
- [SpeechGPT: Empowering Large Language Models with Intrinsic Cross-Modal Conversational Abilities](https://mlsanity.com/paper/2305.11000/) (2023)
- [Snips Voice Platform: an embedded Spoken Language Understanding system for private-by-design voice interfaces](https://mlsanity.com/paper/1805.10190/) (2018)
- [CosyVoice 2: Scalable Streaming Speech Synthesis with Large Language Models](https://mlsanity.com/paper/2412.10117/) (2024)
- [Mini-Omni: Language Models Can Hear, Talk While Thinking in Streaming](https://mlsanity.com/paper/2408.16725/) (2024)
- [Speech Model Pre-training for End-to-End Spoken Language Understanding](https://mlsanity.com/paper/1904.03670/) (2019)
- [Spirit LM: Interleaved Spoken and Written Language Model](https://mlsanity.com/paper/2402.05755/) (2024)
- [BLSP: Bootstrapping Language-Speech Pre-training via Behavior Alignment of Continuation Writing](https://mlsanity.com/paper/2309.00916/) (2023)
- [GPTEval: A Survey on Assessments of ChatGPT and GPT-4](https://mlsanity.com/paper/2308.12488/) (2023)
- [LLM-based NLG Evaluation: Current Status and Challenges](https://mlsanity.com/paper/2402.01383/) (2024)
- [SpeechVerse: A Large-scale Generalizable Audio Language Model](https://mlsanity.com/paper/2405.08295/) (2024)
- [Multi-Task Learning in Natural Language Processing: An Overview](https://mlsanity.com/paper/2109.09138/) (2021)
- [LLaSM: Large Language and Speech Model](https://mlsanity.com/paper/2308.15930/) (2023)
- [Scaling Speech-Text Pre-training with Synthetic Interleaved Data](https://mlsanity.com/paper/2411.17607/) (2024)
- [Leveraging Chain of Thought towards Empathetic Spoken Dialogue without Corresponding Question-Answering Data](https://mlsanity.com/paper/2501.10937/) (2025)

## Built on by (most cited first, 3 of 3)

- [Closing the Modality Reasoning Gap for Speech Large Language Models](https://mlsanity.com/paper/2601.05543/) (2026)
- [How Auditory Knowledge in LLM Backbones Shapes Audio Language Models: A Holistic Evaluation](https://mlsanity.com/paper/2603.19195/) (2026)
- [Rethinking Speech-LLM Integration for ASR: Effective Joint Speech-Text Training by Interleaving](https://mlsanity.com/paper/2607.01733/) (2026)

## Abstract (arXiv)

Large language models (LLMs) have shown remarkable generalization across tasks, leading to increased interest in integrating speech with LLMs. These speech LLMs (SLLMs) typically use supervised fine-tuning to align speech with text-based LLMs. However, the lack of annotated speech data across a wide range of tasks hinders alignment efficiency, resulting in poor generalization. To address these issues, we propose a novel multi-task 'behavior imitation' method with speech-text interleaving, called MTBI, which relies solely on paired speech and transcripts. By ensuring the LLM decoder generates equivalent responses to paired speech and text, we achieve a more generalized SLLM. Interleaving is used to further enhance alignment efficiency. We introduce a simple benchmark to evaluate prompt and task generalization across different models. Experimental results demonstrate that our MTBI outperforms SOTA SLLMs on both prompt and task generalization, while requiring less supervised speech data.