YDYexing Du

Yexing Du

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Yexing Du has a total of 31 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    ProjectEvalPlus: An Agentic Software Engineering Benchmark with Automatic Language Extension and User Simulated Evaluation

    by Kaiyuan Liu, Youcheng Pan, Yexing Du, Lei Zhang, Daojing He, Yang Xiang

    Large Language Models (LLMs) have shown impressive capabilities in Software Engineering (SWE) , yet evaluating their performance on realistic project-level SWE remains challenging. Existing benchmarks focus on isolated snippets, lacking project-level context, language extensibilities, and end-user reality. Thus, we introduce ProjectEvalPlus , a comprehensive user-interaction-simulation benchmark that extends automatic evaluation to multiple programming languages via Automatic Language Extensibility (ALE) , supports multi-agent scenario stages evaluation, and provides detailed feedback on logical and runtime errors. Through unified testing on Python, Java, and JavaScript, we demonstrate that ProjectEvalPlus effectively assesses LLM agents’ capabilities, reveals language-specific biases, and enables iterative cross-language adaptation. Our results highlight both the strengths and limitations of current LLM agents in practical software development, offering actionable insights for future multi-language, project-level software engineering evaluation.


  • Open Access

    SLAM-LLM: A Modular, Open-Source Multimodal Large Language Model Framework and Best Practice for Speech, Language, Audio and Music Processing

    by Ziyang Ma, Guanrou Yang, Wenxi Chen, Zhifu Gao, Yexing Du, Xiquan Li, Zhisheng Zheng, Haina Zhu, Jianheng Zhuo, Zheshu Song, et al.

    The recent surge in open-source Multimodal Large Language Models (MLLM) frameworks, such as LLaVA, provides a convenient kickoff for artificial intelligence developers and researchers. However, most of the MLLM frameworks take vision as the main input modality, and provide limited in-depth support for the modality of speech, audio, and music. This situation hinders the development of audio-language models, and forces researchers to spend a lot of effort on code writing and hyperparameter tuning. We present SLAM-LLM, an open-source deep learning framework designed to train customized MLLMs, focused on speech, language, audio, and music processing. SLAM-LLM provides a modular configuration of different encoders, projectors, LLMs, and parameter-efficient fine-tuning plugins. SLAM-LLM also includes detailed training and inference recipes for mainstream tasks, along with high-performance checkpoints like LLM-based Automatic Speech Recognition (ASR), Automated Audio Captioning (AAC), and Music Captioning (MC). Some of these recipes have already reached or are nearing state-of-the-art performance, and some relevant techniques have also been accepted by academic papers. We hope SLAM-LLM will accelerate iteration, development, data engineering, and model training for researchers. We are committed to continually pushing forward audio-based MLLMs through this open-source framework, and call on the community to contribute to the LLM-based speech, audio and music processing.


  • MCAT: Scaling Many-to-Many Speech-to-Text Translation With MLLMs to 70 Languages

    by Yexing Du, Kaiyuan Liu, Youcheng Pan, Bo Yang, Keqi Deng, Xie Chen, Yang Xiang, Ming Liu, Bing Qin, YaoWei Wang

    Multimodal Large Language Models (MLLMs) have achieved great success in Speech-to-Text Translation (S2TT) tasks. However, current research is constrained by two key challenges: language coverage and efficiency. Most of the popular S2TT datasets are substantially English-centric, which restricts the scaling-up of MLLMs' many-to-many translation capabilities. Moreover, the inference speed of MLLMs degrades dramatically when the speech is converted into long sequences (e.g., 750 tokens). To address these limitations, we propose a Multilingual Cost-effective Accelerated Speech-to-Text Translator (MCAT) framework, which includes two innovations. First, a language scaling method that leverages curriculum learning and a data balancing strategy are introduced to extend the language coverage supported by MLLMs to 70 languages and achieve mutual translation among these languages. Second, an optimized speech adapter module is designed to reduce the length of the speech sequence to only 30 tokens. Extensive experiments were conducted on MLLMs of different scales (9B and 27B). The experimental results demonstrate that MCAT not only surpasses state-of-theart end-to-end models on the FLEURS dataset across 70 × 69 directions but also enhances batch inference efficiency. This is achieved with only $\sim$ 100M trainable parameters and by using only 10 hours of S2TT data per language. Furthermore, we have released MCAT as open-source to promote the development of MLLMs for robust S2TT capabilities.