Geng Xuelong's Personal Homepage

I am currently pursuing my master's degree at the School of Computer Science, Northwestern Polytechnical University, with a focus on speech recognition and LLM-based speech understanding and dialogue systems. I am particularly interested in exploring how large language models (LLMs) can be applied not only to ASR but also to human-machine dialogue systems that approach real human conversation.

I am always open to discussions and collaborations. Feel free to reach out if you are interested in any of my research topics!

• 1. Unveiling the Potential of LLM-Based ASR on Chinese Open-Source Datasets
  First Author; Publication: IEEE 14th International Conference on Speech and Language Processing (2024) Link to paper
• 2. OSUM: Advancing Open Speech Understanding Models with Limited Resources in Academia
  First Author; Publication: arXiv Preprint arXiv:2501.13306, 2025 Link to paper
• 3. Domain-Specific Prompts for LLM-based ASR: An Empirical Study
  First Author; "To be published"
• 4. Enhancing Non-Core Language Instruction-Following in Speech LLMs via Semi-Implicit Cross-Lingual CoT Reasoning
  Co-author; Publication: ACM International Conference on Multimedia (ACM MM), 2025 Link to paper
• 5. Steering Language Model to Stable Speech Emotion Recognition via Contextual Perception and Chain of Thought
  Co-author; "To be published" Link to paper
• 6. Ideal-LLM: Integrating Dual Encoders and Language-Adapted LLM for Multilingual Speech-to-Text
  Co-author; "To be published" Link to paper
• 7. Selective Invocation for Multilingual ASR: A Cost-effective Approach Adapting to Speech Recognition Difficulty
  Co-author; Publication: Interspeech 2025 Link to paper