# Reliable Inference in Edge-Cloud Model Cascades via Conformal Alignment

arXiv: https://arxiv.org/abs/2510.17543 · HTML: https://mlsanity.com/paper/2510.17543/

## Key facts

- Cited by 1 papers in the map (data through 7 Oct 2026).
- Builds on 31 of its 50 references that are in the map.
- First posted on arXiv 20 Oct 2025.

## Builds on (most cited first, 30 of 31)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [Wide Residual Networks](https://mlsanity.com/paper/1605.07146/) (2016)
- [Language Models (Mostly) Know What They Know](https://mlsanity.com/paper/2207.05221/) (2022)
- [A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantification](https://mlsanity.com/paper/2107.07511/) (2021)
- [Conformal Prediction Under Covariate Shift](https://mlsanity.com/paper/1904.06019/) (2019)
- [Conformal prediction beyond exchangeability](https://mlsanity.com/paper/2202.13415/) (2022)
- [Confident Adaptive Language Modeling](https://mlsanity.com/paper/2207.07061/) (2022)
- [Over the Air Deep Learning Based Radio Signal Classification](https://mlsanity.com/paper/1712.04578/) (2017)
- [Trust or Escalate: LLM Judges with Provable Guarantees for Human Agreement](https://mlsanity.com/paper/2407.18370/) (2024)
- [Hypothesis testing with e-values](https://mlsanity.com/paper/2410.23614/) (2024)
- [Selection by Prediction with Conformal p-values](https://mlsanity.com/paper/2210.01408/) (2022)
- [Optimal Inference After Model Selection](https://mlsanity.com/paper/1410.2597/) (2014)
- [TeleQnA: A Benchmark Dataset to Assess Large Language Models Telecommunications Knowledge](https://mlsanity.com/paper/2310.15051/) (2023)
- [When Does Confidence-Based Cascade Deferral Suffice?](https://mlsanity.com/paper/2307.02764/) (2023)
- [Mitigating LLM Hallucinations via Conformal Abstention](https://mlsanity.com/paper/2405.01563/) (2024)
- [Conformal Alignment: Knowing When to Trust Foundation Models with Guarantees](https://mlsanity.com/paper/2405.10301/) (2024)
- [Adaptation with Self-Evaluation to Improve Selective Prediction in LLMs](https://mlsanity.com/paper/2310.11689/) (2023)
- [Probabilistic Conformal Prediction Using Conditional Random Samples](https://mlsanity.com/paper/2206.06584/) (2022)
- [Conformal prediction with local weights: randomization enables local guarantees](https://mlsanity.com/paper/2310.07850/) (2023)
- [Conformal Decision Theory: Safe Autonomous Decisions from Imperfect Predictions](https://mlsanity.com/paper/2310.05921/) (2023)
- [Decision Theoretic Foundations for Conformal Prediction: Optimal Uncertainty Quantification for Risk-Averse Agents](https://mlsanity.com/paper/2502.02561/) (2025)
- [Real-Time Radio Technology and Modulation Classification via an LSTM Auto-Encoder](https://mlsanity.com/paper/2011.08295/) (2020)
- [Gatekeeper: Improving Model Cascades Through Confidence Tuning](https://mlsanity.com/paper/2502.19335/) (2025)
- [A unified framework for bandit multiple testing](https://mlsanity.com/paper/2107.07322/) (2021)
- [Semi-supervised multiple testing](https://mlsanity.com/paper/2106.13501/) (2021)
- [Enhancing Automatic Modulation Recognition through Robust Global Feature Extraction](https://mlsanity.com/paper/2401.01056/) (2024)
- [Calibrating Bayesian Learning via Regularization, Confidence Minimization, and Selective Inference](https://mlsanity.com/paper/2404.11350/) (2024)
- [Conditional Testing based on Localized Conformal p-values](https://mlsanity.com/paper/2409.16829/) (2024)
- [Distilling Calibration via Conformalized Credal Inference](https://mlsanity.com/paper/2501.06066/) (2025)

## Built on by (most cited first, 1 of 1)

- [Calibrate-Then-Delegate: Safety Monitoring with Risk and Budget Guarantees via Model Cascades](https://mlsanity.com/paper/2604.14251/) (2026)

## Abstract (arXiv)

Edge intelligence enables low-latency inference via compact on-device models, but assuring reliability remains challenging. We study edge-cloud cascades that must preserve conditional coverage: whenever the edge returns a prediction set, it should contain the true label with a user-specified probability, as if produced by the cloud model. We formalize conditional coverage with respect to the cloud predictive distribution, and introduce a conformal alignment-based (CAb) cascading mechanism that certifies this property with user control over the risk level. Our method casts escalation from edge to cloud models as a multiple-hypothesis testing (MHT) problem, tailoring conformal alignment (CA) to select which inputs can be safely handled at the edge. The proposed CAb model cascading method yields statistical guarantees on the average fraction of edge decisions that satisfy cloud-level conditional coverage. The procedure applies to arbitrary edge prediction sets, including variants of conformal prediction (CP), and exposes a tunable trade-off among coverage, deferral rate, and set size. Experiments on CIFAR-100 image classification and the TeleQnA question-answering (QA) benchmark show that the proposed CAb cascade maintains the target conditional coverage for edge predictions while substantially reducing offloading to the cloud and incurring modest increases in prediction-set size.