Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
Can LLMS Beat Humans in Debating? A Dynamic Multi-Agent Framework for Competitive DebateAlibaba · IEEE International Conference on Acoustics Speech and Signal Processing2 citationsApr 21Benchmarking Gaslighting Attacks against Speech Large Language ModelsAlibaba · IEEE International Conference on Acoustics Speech and Signal Processing1 citationApr 21Marco-Voice: A Unified Framework for Expressive Speech Synthesis with Voice CloningAlibaba · IEEE International Conference on Acoustics Speech and Signal Processing1 citationApr 21Think-Augmented Function Calling: Improving LLM Parameter Accuracy through Embedded ReasoningAlibaba · IEEE International Conference on Acoustics Speech and Signal Processing1 citationApr 21FGGM: Fisher-Guided Gradient Masking for Continual LearningAlibaba · IEEE International Conference on Acoustics Speech and Signal Processing1 citationApr 21FadeMem: Biologically-Inspired Forgetting for Efficient Agent MemoryAlibaba · IEEE International Conference on Acoustics Speech and Signal Processing1 citationApr 21FW-VTON: Flattening-and-Warping for Person-to-Person Virtual Try-OnAlibaba · IEEE International Conference on Acoustics Speech and Signal Processing1 citationApr 21pTSE-T: Presentation Target Speaker Extraction Using Unaligned Text CuesAlibaba · IEEE International Conference on Acoustics Speech and Signal Processing1 citationApr 21How Do Answer Tokens Read Reasoning Traces? Self-Reading Patterns in Thinking LLMs for Quantitative ReasoningAlibaba · arXiv (Cornell University)Apr 21TEMPO: Scaling Test-time Training for Large Reasoning ModelsAlibaba · arXiv (Cornell University)Apr 21LoopCTR: Unlocking the Loop Scaling Power for Click-Through Rate PredictionAlibaba · arXiv (Cornell University)Apr 21Thinking Before Matching: A Reinforcement Reasoning Paradigm Towards General Person Re-IdentificationAlibaba · arXiv (Cornell University)Apr 21CulturALL: Benchmarking Multilingual and Multicultural Competence of LLMs on Grounded TasksAlibaba · arXiv (Cornell University)Apr 21UAF: A Unified Audio Front-end LLM for Full-Duplex Speech InteractionAlibaba · arXiv (Cornell University)Apr 21How Do Answer Tokens Read Reasoning Traces? Self-Reading Patterns in Thinking LLMs for Quantitative ReasoningAlibaba · arXiv (Cornell University)Apr 21E2E-AEC: Implementing An End-To-End Neural Network Learning Approach for Acoustic Echo CancellationAlibaba · IEEE International Conference on Acoustics Speech and Signal ProcessingApr 21SynaSpot: A Lightweight, Streaming Multi-modal Framework for Keyword Spotting with Audio-Text SynergyAlibaba · IEEE International Conference on Acoustics Speech and Signal ProcessingApr 21Ranking-Aware Reinforcement Learning for Ordinal RankingAlibaba · IEEE International Conference on Acoustics Speech and Signal ProcessingApr 21Dynamic Fusion for Large Language Models CompressionAlibaba · IEEE International Conference on Acoustics Speech and Signal ProcessingApr 21FlowSE-GRPO: Training Flow Matching Speech Enhancement via Online Reinforcement LearningAlibaba · IEEE International Conference on Acoustics Speech and Signal ProcessingApr 21Mambaformer: State-Space Augmented Self-Attention with Downup Sampling for Monaural Speech EnhancementAlibaba · IEEE International Conference on Acoustics Speech and Signal ProcessingApr 21Fractal Generative DistillationAlibaba · IEEE International Conference on Acoustics Speech and Signal ProcessingApr 21Entropy-Guided Data-Efficient Training for Multimodal Reasoning Reward ModelsAlibaba · IEEE International Conference on Acoustics Speech and Signal ProcessingApr 21LongSpeech: A Scalable Benchmark for Transcription, Translation and Understanding in Long SpeechAlibaba · IEEE International Conference on Acoustics Speech and Signal ProcessingApr 21DDPT: Distillation and Dynamic Prompt Tuning for Improving Complex Reasoning in Large Language ModelsAlibaba · IEEE International Conference on Acoustics Speech and Signal ProcessingApr 21