Same Architecture, Different Capacity: Optimizer-Induced Spectral Scaling Laws
Paper • 2605.21803 • Published • 2
Kye Gomez
kye
AI & ML interests
Neuroscience, Behavior Science, Anti-Matter, Anti-Gravity propulsion,
Recent Activity
upvoted a paper about 1 hour ago
Same Architecture, Different Capacity: Optimizer-Induced Spectral Scaling Laws upvoted a paper about 1 hour ago
Gated DeltaNet-2: Decoupling Erase and Write in Linear Attention upvoted a paper about 1 hour ago
Q-ARVD: Quantizing Autoregressive Video Diffusion ModelsOrganizations
upvoted 5
papers about 1 hour ago
Gated DeltaNet-2: Decoupling Erase and Write in Linear Attention
Paper • 2605.22791 • Published • 16
Q-ARVD: Quantizing Autoregressive Video Diffusion Models
Paper • 2605.21072 • Published • 17
WorldKV: Efficient World Memory with World Retrieval and Compression
Paper • 2605.22718 • Published • 29
LatentOmni: Rethinking Omni-Modal Understanding via Unified Audio-Visual Latent Reasoning
Paper • 2605.22012 • Published • 35
upvoted 6
papers 1 day ago
Toto 2.0: Time Series Forecasting Enters the Scaling Era
Paper • 2605.20119 • Published • 35
Mem-π: Adaptive Memory through Learning When and What to Generate
Paper • 2605.21463 • Published • 4
Stable Audio 3
Paper • 2605.17991 • Published • 11
Generative Recursive Reasoning
Paper • 2605.19376 • Published • 24
A Survey of Large Audio Language Models: Generalization, Trustworthiness, and Outlook
Paper • 2605.20266 • Published • 52
OScaR: The Occam's Razor for Extreme KV Cache Quantization in LLMs and Beyond
Paper • 2605.19660 • Published • 39
upvoted 5
papers 3 days ago
Stage-adaptive Token Selection for Efficient Omni-modal LLMs
Paper • 2605.20035 • Published • 4
Delta Attention Residuals
Paper • 2605.18855 • Published • 6
When Vision Speaks for Sound
Paper • 2605.16403 • Published • 95
AutoResearchClaw: Self-Reinforcing Autonomous Research with Human-AI Collaboration
Paper • 2605.20025 • Published • 112
Anti-Self-Distillation for Reasoning RL via Pointwise Mutual Information
Paper • 2605.11609 • Published • 186
upvoted 4
papers 4 days ago
Actionable World Representation
Paper • 2605.18743 • Published • 4
Agent Bazaar: Enabling Economic Alignment in Multi-Agent Marketplaces
Paper • 2605.17698 • Published • 6
AtlasVA: Self-Evolving Visual Skill Memory for Teacher-Free VLM Agents
Paper • 2605.17933 • Published • 6
From Runnable to Shippable: Multi-Agent Test-Driven Development for Generating Full-Stack Web Applications from Requirements
Paper • 2605.17242 • Published • 11