Lune

Profiles2020-2025

Joseph E Gonzalez

UC Berkeley

Also published as Joseph Gonzalez

24Papers
4Last author
97Unique co-authors
0First author
0Solo

First and last author come from byline position, which some venues assign alphabetically.

Recent papers24

  1. SimpleStrat: Diversifying Language Model Generation with StratificationAuthor 7 of 7NeurIPS 2025 · 17 citations
  2. MoE-Lightning: High-Throughput MoE Inference on Memory-constrained GPUsAuthor 7 of 8ASPLOS 2025 · 15 citations
  3. LLM-Assisted Code Cleaning For Training Accurate Code GeneratorsAuthor 4 of 6ICLR 2024 · 49 citations
  4. Stylus: Automatic Adapter Selection for Diffusion ModelsAuthor 5 of 11NeurIPS 2024 · 27 citations
  5. RALF: Accuracy-Aware Scheduling for Feature Store MaintenanceAuthor 7 of 10VLDB 2024 · 6 citations
  6. Describing Differences in Image Sets with Natural LanguageAuthor 8 of 9CVPR 2024
  7. Decomposing Complex Queries for Tip-of-the-tongue RetrievalAuthor 3 of 4EMNLP Findings 2023 · 8 citations
  8. Using Language to Extend to Unseen DomainsAuthor 7 of 9ICLR 2023 · 3 citations
  9. Data Efficient Language-Supervised Zero-Shot Recognition with Optimal Transport DistillationAuthor 6 of 6ICLR 2022 · 57 citations
  10. Flexible Rule-Based Decomposition and Metadata Independence in Modin: A Parallel Dataframe SystemAuthor 5 of 10VLDB 2022 · 22 citations
  11. NBDT: Neural-Backed Decision TreeAuthor 8 of 9ICLR 2021 · 56 citations
  12. Visual Transformers: Where Do Transformers Really Belong in Vision Models?Author 8 of 10ICCV 2021 · 54 citations
  13. Taxonomizing local versus global structure in neural network loss landscapesAuthor 5 of 7NeurIPS 2021 · 51 citations
  14. Contrastive Code Representation LearningAuthor 5 of 6EMNLP 2021 · 11 citations
  15. Learning Space Partitions for Path PlanningAuthor 7 of 10NeurIPS 2021 · 11 citations
  16. Hindsight Logging for Model TrainingAuthor 6 of 8VLDB 2021 · 10 citations
  17. Grounded Graph Decoding improves Compositional Generalization in Question AnsweringAuthor 4 of 6EMNLP Findings 2021 · 6 citations
  18. Remembering for the Right Reasons: Explanations Reduce Catastrophic ForgettingAuthor 5 of 8ICLR 2021 · 3 citations
  19. FBNetV3: Joint Architecture-Recipe Search Using Predictor PretrainingAuthor 11 of 11CVPR 2021
  20. Towards Scalable Dataframe SystemsAuthor 7 of 11VLDB 2020 · 109 citations
  21. A Statistical Framework for Low-bitwidth Training of Deep Neural NetworksAuthor 5 of 5NeurIPS 2020 · 75 citations
  22. Boundary thickness and robustness in learning modelsAuthor 6 of 8NeurIPS 2020 · 53 citations
  23. Cloudburst: Stateful Functions-as-a-ServiceAuthor 5 of 11VLDB 2020
  24. FBNetV2: Differentiable Neural Architecture Search for Spatial and Channel DimensionsAuthor 12 of 13CVPR 2020
Dusk over the sea between two cliffs drawn in fine vertical lines