mlsanityData to 7 Oct 2026

Reliable Inference in Edge-Cloud Model Cascades via Conformal Alignment

arXiv 2510.17543 ↗First posted 20 Oct 2025cs.LG eess.SP stat.MLcited by 1

Not on the map. Why?

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.

Key facts. Cited by 1 paper 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

31 of 50 references.

References · 31

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  2. 2017On Calibration of Modern Neural Networkscited by 4,860
  3. 2016Wide Residual Networkscited by 3,057
  4. 2022Language Models (Mostly) Know What They Knowcited by 1,783
  5. 2021A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantificationcited by 777
  6. 2019Conformal Prediction Under Covariate Shiftcited by 511
  7. 2022Conformal prediction beyond exchangeabilitycited by 348
  8. 2022Confident Adaptive Language Modelingcited by 297
  9. 2017Over the Air Deep Learning Based Radio Signal Classificationcited by 121
  10. 2024Trust or Escalate: LLM Judges with Provable Guarantees for Human Agreementcited by 99
  11. 2024Hypothesis testing with e-valuescited by 70
  12. 2022Selection by Prediction with Conformal p-valuescited by 66
19 more
  1. 2014Optimal Inference After Model Selectioncited by 63
  2. 2023TeleQnA: A Benchmark Dataset to Assess Large Language Models Telecommunications Knowledgecited by 62
  3. 2023When Does Confidence-Based Cascade Deferral Suffice?cited by 58
  4. 2024Mitigating LLM Hallucinations via Conformal Abstentioncited by 56
  5. 2024Conformal Alignment: Knowing When to Trust Foundation Models with Guaranteescited by 50
  6. 2023Adaptation with Self-Evaluation to Improve Selective Prediction in LLMscited by 49
  7. 2022Probabilistic Conformal Prediction Using Conditional Random Samplescited by 47
  8. 2023Conformal prediction with local weights: randomization enables local guaranteescited by 47
  9. 2023Conformal Decision Theory: Safe Autonomous Decisions from Imperfect Predictionscited by 46
  10. 2025Decision Theoretic Foundations for Conformal Prediction: Optimal Uncertainty Quantification for Risk-Averse Agentscited by 37
  11. 2020Real-Time Radio Technology and Modulation Classification via an LSTM Auto-Encodercited by 19
  12. 2025Gatekeeper: Improving Model Cascades Through Confidence Tuningcited by 10
  13. 2021A unified framework for bandit multiple testingcited by 8
  14. 2021Semi-supervised multiple testingcited by 7
  15. 2024Enhancing Automatic Modulation Recognition through Robust Global Feature Extractioncited by 7
  16. 2024Calibrating Bayesian Learning via Regularization, Confidence Minimization, and Selective Inferencecited by 6
  17. 2024Conditional Testing based on Localized Conformal p-valuescited by 4
  18. 2025Distilling Calibration via Conformalized Credal Inferencecited by 2
  19. 2025Reliable Wireless Indoor Localization via Cross-Validated Prediction-Powered Calibrationcited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Calibrate-Then-Delegate: Safety Monitoring with Risk and Budget Guarantees via Model Cascadescited by 3