STEPQuant: When and Where Errors Matter in Delta-Rule Recurrent State Quantization
Paper • 2609.38169 • Published • 105
Haokun
FelixMessi
AI & ML interests
None yet
Recent Activity
authored a paper about 17 hours ago
STEPQuant: When and Where Errors Matter in Delta-Rule Recurrent State Quantization upvoted a paper 1 day ago
SGF+: Decoupling Gradient Flows for Autoregressive Video GenerationOrganizations
None yet
upvoted 4
papers 1 day ago
SGF+: Decoupling Gradient Flows for Autoregressive Video Generation
Paper • 2610.10429 • Published • 53
Long-WAM: Scaling the Context of World-Action Models
Paper • 2610.10528 • Published • 94
TRACE: Rollout-Guided Quantization-Aware Training for FP4 Reinforcement Learning of MoE Language Models
Paper • 2610.07767 • Published • 97