Research
A paper is listed when one of its authors is affiliated with the lab. Data: OpenAlex.
ASVspoof 5: Evaluation of Spoofing, Deepfake, and Adversarial Attack Detection Using Crowdsourced SpeechMicrosoft · IEEE Transactions on Audio Speech and Language Processing4 citationsJan 1SLAM-LLM: A Modular, Open-Source Multimodal Large Language Model Framework and Best Practice for Speech, Language, Audio and Music ProcessingAlibaba · IEEE Journal of Selected Topics in Signal Processing4 citationsJan 1PhyGDPO: Physics-Aware Groupwise Direct Preference Optimization for Physically Consistent Text-to-Video GenerationMeta · Lecture notes in computer science4 citationsJan 1ComfyUI-R1: Exploring Reasoning Models for Workflow GenerationAlibaba · Findings of the Association for Computational Linguistics: ACL (Findings of ACL)4 citationsJan 1Distribution Matching Distillation Meets Reinforcement LearningAlibaba · Lecture notes in computer science4 citationsJan 1Not All Tokens Matter: Towards Efficient LLM Reasoning via Token Significance in Reinforcement LearningAlibaba, Microsoft · Proceedings of the conference - Association for Computational Linguistics. Meeting4 citationsJan 1LLP: LLM-Based Product Pricing in E-commerceAlibaba · Conference on Empirical Methods in Natural Language Processing (EMNLP)4 citationsJan 1V-MAGE: A Game Evaluation Framework for Assessing Vision-Centric Capabilities in Multimodal Large Language ModelsMicrosoft · Findings of the Association for Computational Linguistics: ACL (Findings of ACL)4 citationsJan 1Live Avatar: Streaming Real-Time Audio-Driven Avatar Generation with Infinite LengthAlibaba · Lecture notes in computer science4 citationsJan 1GUI-360° : A Comprehensive Dataset and Benchmark for Desktop Productivity Computer-Using AgentsMicrosoft · Conference on Empirical Methods in Natural Language Processing (EMNLP)4 citationsJan 1Comprehensive Language–Image Pre-training for 3D Medical Image UnderstandingMicrosoft · Lecture notes in computer science4 citationsJan 1Jais 2: A Family of Arabic-Centric Open Large Language ModelsCerebras Systems · Conference on Empirical Methods in Natural Language Processing (EMNLP)3 citationsJan 1DeepPlanning: Benchmarking Long-Horizon Agentic Planning with Verifiable ConstraintsAlibaba · Proceedings of the conference - Association for Computational Linguistics. Meeting3 citationsJan 1Benchmarking and Enabling Efficient Chinese Medical Retrieval via Asymmetric EncodersiFlytek · Proceedings of the conference - Association for Computational Linguistics. Meeting3 citationsJan 1Unified Thinker: A General Reasoning Core for Image GenerationAlibaba · Proceedings of the conference - Association for Computational Linguistics. Meeting3 citationsJan 1OASIS: Online Sample Selection for Continual Instruction TuningAlibaba · Proceedings of the conference - Association for Computational Linguistics. Meeting3 citationsJan 1Outcome Accuracy is Not Enough: Aligning the Reasoning Process of Reward ModelsAlibaba · Proceedings of the conference - Association for Computational Linguistics. Meeting3 citationsJan 1SpeechLLM-as-Judges: Towards General and Interpretable Speech Quality EvaluationMicrosoft · Proceedings of the conference - Association for Computational Linguistics. Meeting3 citationsJan 1STReasoner: Empowering LLMs for Spatio-Temporal Reasoning in Time Series via Spatial-Aware Reinforcement LearningMicrosoft · Proceedings of the conference - Association for Computational Linguistics. Meeting3 citationsJan 1HDA-SELD: Hierarchical Cross-Modal Distillation With Multi-Level Data Augmentation for Low-Resource Audio-Visual Sound Event Localization and DetectioniFlytek · IEEE Transactions on Audio Speech and Language Processing3 citationsJan 1Multilingual Amnesia: On the Transferability of Unlearning in Multilingual LLMsAlphabet · Conference of the European Chapter of the Association for Computational Linguistics3 citationsJan 1An End-to-End Optimized Lensless System for Privacy-Preserving Face VerificationNVIDIA, SenseTime · IEEE Transactions on Image Processing3 citationsJan 1Edit in 2D, Verify in 3D: Reinforcement Learning for Multi-view Consistent Scene EditingAlibaba · Lecture notes in computer science3 citationsJan 1Enhancing Weakly Supervised Multimodal Video Anomaly Detection Through Text GuidanceAlibaba · IEEE Transactions on Multimedia3 citationsJan 1L-CLIPScore: A Lightweight Embedding-Based Captioning Metric for Evaluating and TrainingAlibaba · IEEE Transactions on Multimedia3 citationsJan 1