KVarN: Variance-Normalized KV-Cache Quantization Mitigates Error Accumulation in Reasoning Tasks
Paper • 2606.03458 • Published • 49
Felix Arnold
plex1
AI & ML interests
None yet
Recent Activity
upvoted a paper 8 months ago
SINQ: Sinkhorn-Normalized Quantization for Calibration-Free
Low-Precision LLM Weights liked a Space about 1 year ago
snowkylin/circuit-transformer-demoOrganizations
None yet