# CS3-Bench: Evaluating and Enhancing Speech-to-Speech LLMs for Mandarin-English Code-Switching

arXiv: https://arxiv.org/abs/2510.07881 · HTML: https://mlsanity.com/paper/2510.07881/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in MLLM training and visual instruction tuning, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 17 of its 20 references that are in the map.
- First posted on arXiv 9 Oct 2025.

## Placed in

- Multimodal LLMs / vision-language models / [MLLM training and visual instruction tuning](https://mlsanity.com/mllm/mllm-training-and-visual-instruction-tuning/)

## Builds on (most cited first, 17 of 17)

- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [Qwen2.5-Omni Technical Report](https://mlsanity.com/paper/2503.20215/) (2025)
- [Seed-TTS: A Family of High-Quality Versatile Speech Generation Models](https://mlsanity.com/paper/2406.02430/) (2024)
- [CosyVoice 2: Scalable Streaming Speech Synthesis with Large Language Models](https://mlsanity.com/paper/2412.10117/) (2024)
- [GLM-4-Voice: Towards Intelligent and Human-Like End-to-End Spoken Chatbot](https://mlsanity.com/paper/2412.02612/) (2024)
- [Kimi-Audio Technical Report](https://mlsanity.com/paper/2504.18425/) (2025)
- [Mini-Omni: Language Models Can Hear, Talk While Thinking in Streaming](https://mlsanity.com/paper/2408.16725/) (2024)
- [Baichuan-Omni-1.5 Technical Report](https://mlsanity.com/paper/2501.15368/) (2025)
- [WavChat: A Survey of Spoken Dialogue Models](https://mlsanity.com/paper/2411.13577/) (2024)
- [SLAM-Omni: Timbre-Controllable Voice Interaction System with Single-Stage Training](https://mlsanity.com/paper/2412.15649/) (2024)
- [CoVoST: A Diverse Multilingual Speech-To-Text Translation Corpus](https://mlsanity.com/paper/2002.01320/) (2020)
- [VITA-Audio: Fast Interleaved Cross-Modal Token Generation for Efficient Large Speech-Language Model](https://mlsanity.com/paper/2505.03739/) (2025)
- [URO-Bench: Towards Comprehensive Evaluation for End-to-End Spoken Dialogue Models](https://mlsanity.com/paper/2502.17810/) (2025)
- [VocalBench: Benchmarking the Vocal Conversational Abilities for Speech Interaction Models](https://mlsanity.com/paper/2505.15727/) (2025)
- [VocalNet: Speech LLM with Multi-Token Prediction for Faster and High-Quality Generation](https://mlsanity.com/paper/2504.04060/) (2025)

## Built on by (most cited first, 2 of 2)

- [VocalBench-zh: Decomposing and Benchmarking the Speech Conversational Abilities in Mandarin Context](https://mlsanity.com/paper/2511.08230/) (2025)
- [Dual-Form ASR: Semantics-Aware Inverse Text Normalization for Chinese Speech Recognition](https://mlsanity.com/paper/2609.02901/) (2026)

## Abstract (arXiv)

The advancement of multimodal large language models has accelerated the development of speech-to-speech interaction systems. While natural monolingual interaction has been achieved, we find existing models exhibit deficiencies in language alignment. In our proposed Code-Switching Speech-to-Speech Benchmark (CS3-Bench), experiments on 7 mainstream models demonstrate a relative performance drop of up to 66% in knowledge-intensive question answering and varying degrees of misunderstanding in open-ended conversations. Starting from a model with severe performance deterioration, we propose both data constructions and training approaches to improve the language alignment capabilities, specifically employing Chain of Recognition (CoR) to enhance understanding and Keyword Highlighting (KH) to guide generation. Our approach improves the knowledge accuracy from 25.14% to 46.13%, with open-ended understanding rate from 64.5% to 86.5%, and significantly reduces pronunciation errors in the secondary language. CS3-Bench is available at https://huggingface.co/datasets/VocalNet/CS3-Bench.