publications
publications by categories in reversed chronological order. generated by jekyll-scholar.
2025
- NAACL SubmissionCoRAG: Collaborative Retrieval-Augmented GenerationIn 2025 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL), Apr 2025
- NAACL SubmissionDecoding Dark Matter: Specialized Sparse Autoencoders for Interpreting Rare Concepts in Foundation ModelsIn 2025 Annual Conference of the North American Chapter of the Association for Computational Linguistics (NAACL), Apr 2025
2024
- CVPRAdversarial Text to Continuous Image GenerationIn The IEEE/CVF Conference on Computer Vision and Pattern Recognition 2024, Apr 2024
- EMNLPGrass: Compute Efficient Low-Memory LLM Training with Structured Sparse GradientsIn The 2024 Conference on Empirical Methods in Natural Language Processing (EMNLP) , Apr 2024
- NeurIPS ENLSPInducing Elasticity in Foundation Models: Post-Training Techniques for Adaptable InferenceIn The 4th Workshop on Efficient Natural Language and Speech Processing, Dec 2024
- NeurIPS ATTRIBDecoding Dark Matter: Specialized Sparse Autoencoders for Interpreting Rare Concepts in LLMsIn Second NeurIPS Workshop on Attributing Model Behavior at Scale (ATTRIB 2024), Dec 2024
- EMNLP CustomNLP4ULess is Fed More: Sparsity Reduces Feature Distortion in Federated LearningIn Workshop on Customizable NLP (CustomNLP4U) at EMNLP 2024, Nov 2024
- ICML ES-FoMoGrass: Compute Efficient Low-Memory LLM Training with Structured Sparse GradientsIn Efficient Systems for Foundation Models, Workshop at the International Conference on Machine Learning (ICML), Nov 2024
- ICLR Private MLFed Up with Complexity: Simplifying Many-Task Federated Learning with NTKFedAvgIn Privacy Regulation and Protection in Machine Learning, Workshop at ICLR, Nov 2024
- ICLR Private MLCache Me If You Can: The Case For Retrieval Augmentation in Federated LearningIn Privacy Regulation and Protection in Machine Learning, Workshop at ICLR, Nov 2024
- EACL MOOMINLess is Fed More: Sparsity Reduces Feature Distortion in Federated LearningIn The First Workshop on Modular and Open Multilingual NLP, EACL, Mar 2024
- PreprintNeurIPS 2023 Competition: Privacy Preserving Federated Learning Document VQAIn , Mar 2024
2023
- PAKDDWeb-scale semantic product search with large language modelsIn Pacific-Asia Conference on Knowledge Discovery and Data Mining, Mar 2023
- ACL OralReAugKD: Retrieval-Augmented Knowledge Distillation for Pre-trained Language ModelsIn Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics ACL, Mar 2023
- ICML ES-FoMoReverse Distillation: Training Billion Parameter Models For CTR PredictionIn ICML Workshop on Efficient Systems for Foundation Models (ES-FoMo), Mar 2023
- PreprintAn In-depth Look at Gemini’s Language AbilitiesarXiv preprint arXiv: 2312.11444, Mar 2023
- KDDTutorial on Training Large-scale Foundation Models on Emerging AI ChipsIn Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Mar 2023
2022
- WWWDCAF-BERT: A Distilled Cachable Adaptable Factorized Model For Improved Ads CTR PredictionIn WWW ’22: Proceedings of the Web Conference, Mar 2022...
- PreprintHyperCGAN: Text-to-Image Synthesis with HyperNet-Modulated Conditional Generative Adversarial Networkshttps://openreview.net/forum?id=z-5BjnU3-OQ, Mar 2022
2021
- AAAI OralSymbolic Music Generation with Transformer-GANsIn Proceedings of the AAAI Conference on Artificial Intelligence, Mar 2021
- EMBCGated Transformer for Decoding Human Brain EEG Signals43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), Mar 2021
- NeurIPS ENLSPCTR-BERT: Cost-effective knowledge distillation for billion-parameter teacher modelsIn NeurIPS Workshop on Efficient Natural Language and Speech Processing (ENLSP) (Oral Spotlight), Mar 2021
2020
- NeurIPS ML4CDTransformer-GAN: Symbolic music generation using a learned lossIn 4th NeurIPS Workshop on Machine Learning for Creativity and Design, Mar 2020
- ISMIR NLP4MusASymbolic Music Generation with Transformer-GANsIn 1st ISMIR Workshop on NLP for Music and Audio (NLP4MusA) (Oral spotlight), Mar 2020
- Patent3D convolutional neural networks for television advertisement detection. US Patent 10,706,286Jul 2020US Patent 10,706,286
- PatentText independent speaker-verification on a media operating system using deep learning on raw waveforms. US Patent 10,699,715Jun 2020US Patent 10,699,715
- Amazon CVCDesigning event representations for symbolic musicIn Amazon Computer Vision Conference (CVC), Jun 2020