-
Taming LLMs by Scaling Learning Rates with Gradient Grouping
Paper • 2506.01049 • Published • 38 -
Switch EMA: A Free Lunch for Better Flatness and Sharpness
Paper • 2402.09240 • Published • 5 -
Unveiling the Backbone-Optimizer Coupling Bias in Visual Representation Learning
Paper • 2410.06373 • Published • 36 -
OpenMixup: Open Mixup Toolbox and Benchmark for Visual Representation Learning
Paper • 2209.04851 • Published • 3
Juanxi Tian
Juanxi
AI & ML interests
MLLM｜Optimization ｜Visual Generation
Recent Activity
updated
a collection
about 5 hours ago
Gen AI
commented on
a paper
about 9 hours ago
Envision: Benchmarking Unified Understanding & Generation for Causal World Process Insights
authored
a paper
about 10 hours ago
Envision: Benchmarking Unified Understanding & Generation for Causal World Process Insights
Organizations
-
Envision: Benchmarking Unified Understanding & Generation for Causal World Process Insights
Paper • 2512.01816 • Published • 80 -
MergeVQ: A Unified Framework for Visual Generation and Representation with Disentangled Token Merging and Quantization
Paper • 2504.00999 • Published • 95 -
WideRange4D: Enabling High-Quality 4D Reconstruction with Wide-Range Movements and Scenes
Paper • 2503.13435 • Published • 18 -
Trans4D: Realistic Geometry-Aware Transition for Compositional Text-to-4D Synthesis
Paper • 2410.07155 • Published • 11
-
Taming LLMs by Scaling Learning Rates with Gradient Grouping
Paper • 2506.01049 • Published • 38 -
Switch EMA: A Free Lunch for Better Flatness and Sharpness
Paper • 2402.09240 • Published • 5 -
Unveiling the Backbone-Optimizer Coupling Bias in Visual Representation Learning
Paper • 2410.06373 • Published • 36 -
OpenMixup: Open Mixup Toolbox and Benchmark for Visual Representation Learning
Paper • 2209.04851 • Published • 3
-
Envision: Benchmarking Unified Understanding & Generation for Causal World Process Insights
Paper • 2512.01816 • Published • 80 -
MergeVQ: A Unified Framework for Visual Generation and Representation with Disentangled Token Merging and Quantization
Paper • 2504.00999 • Published • 95 -
WideRange4D: Enabling High-Quality 4D Reconstruction with Wide-Range Movements and Scenes
Paper • 2503.13435 • Published • 18 -
Trans4D: Realistic Geometry-Aware Transition for Compositional Text-to-4D Synthesis
Paper • 2410.07155 • Published • 11
models
0
None public yet