LLM Digest

Story

arxiv_cs_cl · Nov 9, 2026 · paper

Source brief

How Far Does Second-Order Linear Response Predict Quantization Mismatch? Pre-Registered Measurements on Five Language Models

arxiv.orgNov 9, 2026
original source linked

In brief

Serving systems choose a weight format for each model and deployment. A low-cost screen predicts the resulting KL divergence from the unquantized model's Fisher information using $\tfrac12Δ^\top FΔ$, the expansion und...

Read the original at arxiv.org →Open in live feed Read that day’s brief

How Far Does Second-Order Linear Response Predict Quantization Mismatch? Pre-Registered Measurements on Five Language Models

Earlier in this thread 4 items

How Much Training-Inference Mismatch Can On-Policy Distillation Tolerate?

Coordinated Mixed-Precision KV Cache Quantization for Efficient LLM-based Multi-agent Systems

Qwen 4 Release Date: What We Know So Far

Zero Mismatch, Simple Async RL: REINFORCE Zero, DeepSeek-V4.1 and Kimi K3.1