mlsanityData to 14 Oct 2026

Pricing the Risk of Runtime Compression: Anytime-Valid Admission and a Served-Output Law for Compressed Serving State

arXiv 2608.15810 ↗First posted 16 Aug 2026cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Runtime compression of serving state trades quality for capacity with no priced guarantee: systems adapt precision on load signals with no soundness statement, and certified approaches budget request-level risk by a union bound over a pre-declared event count. We show the union budget exhausts on every long request in a production serving stack (100% of requests), and replace it with an anytime-valid, physically accounted ledger whose bound holds at every one of 352,333 admission calls on live traffic and which, in a pre-registered held-out confirmatory round, halves the exact-fallback rate at matched risk (0.30 -> 0.14) -- coverage is bought at a price the account states. We then price the remaining distance from the certified witness to what a user experiences: a machine-checked design law (TV <= tanh(a_q w_thr)) turns the served-TV target into a threshold knob, and a three-layer audit of its instantiation -- an operator-norm query envelope measured 1.5x from tight, a measured-ellipsoid replacement for the Cauchy-Schwarz ball that buys nothing (0.89x, held-out sound), and the gate's operating point (~700x) -- localizes the entire 1064x gap to the operating point, a price the law now states rather than an unknown. A priced bound is worth nothing on a request one has not seen, so the third link is the quantifier: exchangeable extrapolation across 80 serving histories replaces binary conformal prediction's vacuous certificates with order-statistic bounds that discriminate (0.41 against 0.51 calibration risk). All probabilistic kernels are Lean 4-checked (228 exported theorems, no sorry); which object deserves this machinery at all is settled empirically in a companion paper that adjudicates -- and rejects -- the natural alternative of certifying routing. What ships is an account: risk you can spend, a gap you can read off a law, and a bound that survives the request you have not seen.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 15 of its 16 references that are in the map. First posted on arXiv 16 Aug 2026.

Builds on

15 of 16 references.

References · 15

  1. 2022GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformerscited by 1,919
  2. 2025MoEQuant: Enhancing Quantization for Mixture-of-Experts Large Language Models via Expert-Balanced Sampling and Affinity Guidancecited by 34
  3. 2025Model-Preserving Adaptive Roundingcited by 23
  4. 2024QERA: an Analytical Framework for Quantization Error Reconstructioncited by 22
  5. 2025EAC-MoE: Expert-Selection Aware Compressor for Mixture-of-Experts Large Language Modelscited by 18
  6. 2025EAQuant: Enhancing Post-Training Quantization for MoE Models via Expert-Aware Optimizationcited by 10
  7. 2026WaterSIC: Information-Theoretically (Near) Optimal Linear Layer Quantizationcited by 7
  8. 2026GEMQ: Global Expert-Level Mixed-Precision Quantization for MoE LLMscited by 7
  9. 2025MorphServe: Efficient and Workload-Aware LLM Serving via Runtime Quantized Layer Swapping and KV Cache Resizingcited by 6
  10. 2026Value-and-Structure Alignment for Routing-Consistent Quantization of Mixture-of-Experts Modelscited by 6
  11. 2025DP-LLM: Runtime Model Adaptation with Dynamic Layer-wise Precision Assignmentcited by 4
  12. 2026Runtime-Certified Bounded-Error Quantized Attentioncited by 4
3 more
  1. 2026WitCert: Sound Runtime Risk Observability and Gating for KV-Cache Quantizationcited by 3
  2. 2026MoBiE: Efficient Inference of Mixture of Binary Experts under Post-Training Quantizationcited by 2
  3. 2026Runtime Observability for Heterogeneous Attention Memorycited by 1

Built on by

Not cited yet.