KV4-RL: Unlocking Faster Asynchronous Rollouts for RLVR with INT4 KV Cache Quantization
Paper • 2610.14448 • Published • 49
SamSmith
SamSmithSS
AI & ML interests
None yet
Recent Activity
upvoted a paper about 20 hours ago
KV4-RL: Unlocking Faster Asynchronous Rollouts for RLVR with INT4 KV Cache Quantization upvoted a paper about 20 hours ago
Long-Horizon RL Needs Muon upvoted a paper about 20 hours ago
OmniXRef: A Unified Framework for Multimodal Multi-Reference Video EditingOrganizations
None yet
upvoted 9
papers about 20 hours ago
Long-Horizon RL Needs Muon
Paper • 2610.20256 • Published • 30
OmniXRef: A Unified Framework for Multimodal Multi-Reference Video Editing
Paper • 2610.22744 • Published • 20
Tree-Aware Posterior Distillation for Long-Horizon Agentic Reinforcement Learning
Paper • 2610.23181 • Published • 42
SpecMRL: Redesign Matryoshka Representation Learning within Parametric Spectrum
Paper • 2610.26824 • Published • 81
Embeddings Are Overpriced? A Frozen 16 MB Vocabulary for Small and Ternary Language Models
Paper • 2610.22428 • Published • 13
BlenderWorld: Benchmarking Visual Agents for 3D Creation from Real-World References
Paper • 2610.23986 • Published • 22
Towards Looped Models Done Right, Part I: Topology, Input Injection, Recurrent-State Design
Paper • 2610.23036 • Published • 12
OPMD: On-Policy Mechanism Distillation Beyond Output Supervision
Paper • 2610.14022 • Published • 11
upvoted a paper 21 days ago
upvoted 2
papers 4 months ago
upvoted a paper 9 months ago