KVarN: Variance-Normalized KV-Cache Quantization Mitigates Error Accumulation in Reasoning Tasks Paper • 2606.03458 • Published 4 days ago • 49
SINQ: Sinkhorn-Normalized Quantization for Calibration-Free Low-Precision LLM Weights Paper • 2509.22944 • Published Sep 26, 2025 • 80