Inventor · disambiguated record
Hongxu Yin
Also filed as: YIN HONGXU · YIN HONGXU DANNY
4 granted patents·28 pending applications·6 citations·filing 2017–2025
63Inventor score
Top patents by PatentIndex Score
32 records- 0187US12406023B1Neural network training methodNVIDIA CORP·Filed 2021·Granted Sep 2, 2025·4 cites·29 claims
- 0269US11521068B2Method and system for neural network synthesisUNIV PRINCETON·Filed 2018·Granted Dec 6, 2022·2 cites·27 claims
- 0369US2025384295A1Techniques for implementing multimodal large language models with mixtures of vision encodersNVIDIA CORP·Filed 2025·Application pending·0 cites
- 0469US2025384268A1Techniques for implementing multimodal large language models with mixtures of vision encodersNVIDIA CORP·Filed 2025·Application pending·0 cites
- 0563US2025342346A1Cross-modality alignment for large language modelsNVIDIA CORP·Filed 2024·Application pending·0 cites
- 0661US2025335769A1Learnable semi-structured sparsity for large language modelsNVIDIA CORP·Filed 2024·Application pending·0 cites
- 0760US2025265467A1Techniques for compressing artificial neural networksNVIDIA CORP·Filed 2024·Application pending·0 cites
- 0859US2025252301A1Fine-tuning a neural network model using a weight-decomposed low-rank adaptationNVIDIA CORP·Filed 2024·Application pending·0 cites
- 0959US2025252303A1Many-in-one elastic neural networkNVIDIA CORP·Filed 2024·Application pending·0 cites
- 1057US12452542B2Hallucinating details for over-exposed pixels in videos using learned reference frame selectionNVIDIA CORP·Filed 2024·Granted Oct 21, 2025·0 cites·20 claims
- 1157US2024119361A1Techniques for heterogeneous continual learning with machine learning model architecture progressionNVIDIA CORP·Filed 2023·Application pending·0 cites
- 1256US2025272959A1Region-aware vision language processorNVIDIA CORP·Filed 2025·Application pending·0 cites
- 1356US2024253217A1Loss-guided diffusion modelsNVIDIA CORP·Filed 2023·Application pending·0 cites
- 1456US2025191369A1Language instructed temporal localization in videosNVIDIA CORP·Filed 2024·Application pending·0 cites
- 1555US2024127067A1Sharpness-aware minimization for robustness in sparse neural networksNVIDIA CORP·Filed 2023·Application pending·0 cites
- 1655US2024096115A1Landmark detection with an iterative neural networkNVIDIA CORP·Filed 2023·Application pending·0 cites
- 1754US2024185034A1Generating global hierarchical self-attentionNVIDIA CORP·Filed 2023·Application pending·0 cites
- 1854US2025078998A1System and method for mental health disorder detection system based on wearable sensors and artificial neural networksUNIV PRINCETON·Filed 2022·Application pending·0 cites
- 1953US2023186077A1Adaptive token depth adjustment in transformer neural networksNVIDIA CORP·Filed 2022·Application pending·0 cites
- 2052US2024119291A1Dynamic neural network model sparsificationNVIDIA CORP·Filed 2023·Application pending·0 cites
- 2150US2023394781A1Global context vision transformerNVIDIA CORP·Filed 2022·Application pending·0 cites
- 2249US2023077258A1Performance-aware size reduction for neural networksNVIDIA CORP·Filed 2021·Application pending·0 cites
- 2348US12393842B2System and method for incremental learning using a grow-and-prune paradigm with neural networksUNIV PRINCETON·Filed 2020·Granted Aug 19, 2025·0 cites·15 claims
- 2447US2022156982A1Calculating data compression parametersNVIDIA CORP·Filed 2020·Application pending·0 cites
- 2546US2022292360A1Pruning neural networksNVIDIA CORP·Filed 2021·Application pending·0 cites
- 2644US2021142177A1Synthesizing data for training one or more neural networksNVIDIA CORP·Filed 2019·Application pending·0 cites
- 2743US2022284283A1Neural network training techniqueNVIDIA CORP·Filed 2021·Application pending·0 cites
- 2841US2020320395A1Methods and devices for optimizing machine learning model compactness and accuracy through hardware latency hysteresis effectALIBABA GROUP HOLDING LTD·Filed 2019·Application pending·0 cites
- 2941US2022284232A1Techniques to identify data used to train one or more neural networksNVIDIA CORP·Filed 2021·Application pending·0 cites
- 3040US2020382286A1System and method for smart, secure, energy-efficient iot sensorsUNIV PRINCETON·Filed 2019·Application pending·0 cites
- 3139US2021133540A1System and method for compact, fast, and accurate lstmsUNIV PRINCETON·Filed 2019·Application pending·0 cites
- 3239US2019374160A1Hierarchical health decision support system and methodUNIV PRINCETON·Filed 2017·Application pending·0 cites
Join the waitlist — get patent alerts
Get an alert when Hongxu Yin files or is granted a new patent.
We store only your email — no account needed. See our privacy policy.
Identity basis: PatentsView inventor disambiguation (2025Q4-odp release). How scoring works →