Skip to the Good Part: Representation Structure & Inference-Time Layer Skipping in Diffusion vs. Autoregressive LLMs
Paper
• 2603.07475 • Published
• 2
Mingxun Cai
mingxun97
AI & ML interests
Video Genration, 3D Generation.
Recent Activity
upvoted a paper about 15 hours ago
Scaling Agentic Capabilities, Not Context: Efficient Reinforcement Finetuning for Large Toolspaces upvoted a paper about 15 hours ago
CARE-Edit: Condition-Aware Routing of Experts for Contextual Image Editing Organizations
None yet
upvoted 6 papers about 15 hours ago
Scaling Agentic Capabilities, Not Context: Efficient Reinforcement Finetuning for Large Toolspaces
Paper
• 2603.06713 • Published
• 12
CARE-Edit: Condition-Aware Routing of Experts for Contextual Image Editing
Paper
• 2603.08589 • Published
• 29
BabyVision: Visual Reasoning Beyond Language
Paper
• 2601.06521 • Published
• 200
Watching, Reasoning, and Searching: A Video Deep Research Benchmark on Open Web for Agentic Video Reasoning
Paper
• 2601.06943 • Published
• 215
mHC: Manifold-Constrained Hyper-Connections
Paper
• 2512.24880 • Published
• 317
upvoted 14 papers 7 months ago
Sapiens: Foundation for Human Vision Models
Paper
• 2408.12569 • Published
• 94
Programming Every Example: Lifting Pre-training Data Quality like Experts at Scale
Paper
• 2409.17115 • Published
• 64
FrugalNeRF: Fast Convergence for Few-shot Novel View Synthesis without Learned Priors
Paper
• 2410.16271 • Published
• 84
rStar-Math: Small LLMs Can Master Math Reasoning with Self-Evolved Deep Thinking
Paper
• 2501.04519 • Published
• 288
Gaussian Splatting with Discretized SDF for Relightable Assets
Paper
• 2507.15629 • Published
• 23
CSD-VAR: Content-Style Decomposition in Visual Autoregressive Models
Paper
• 2507.13984 • Published
• 26
Ultra3D: Efficient and High-Fidelity 3D Generation with Part Attention
Paper
• 2507.17745 • Published
• 36
SeC: Advancing Complex Video Object Segmentation via Progressive Concept Construction
Paper
• 2507.15852 • Published
• 38
NoHumansRequired: Autonomous High-Quality Image Editing Triplet Mining
Paper
• 2507.14119 • Published
• 60
Yume: An Interactive World Generation Model
Paper
• 2507.17744 • Published
• 91
Rep-MTL: Unleashing the Power of Representation-level Task Saliency for Multi-Task Learning
Paper
• 2507.21049 • Published
• 41
MangaNinja: Line Art Colorization with Precise Reference Following
Paper
• 2501.08332 • Published
• 62
GameFactory: Creating New Games with Generative Interactive Videos
Paper
• 2501.08325 • Published
• 67
DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
Paper
• 2501.12948 • Published
• 441