mlsanityData to 14 Oct 2026

Calibration-Preserving Pruning: Compression as a Reliability Contract

arXiv 2608.23744 ↗First posted 24 Aug 2026cs.LG cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

Split conformal prediction, not the pruning rule, supplies finite-sample marginal coverage once a pruned model is fixed independently of the conformal calibration split. We study the separate efficiency problem: can pruning preserve score geometry well enough to obtain smaller valid prediction sets? Calibration-Preserving Pruning (CPP) augments a base pruning score with nonconformity-gradient saliency and uses disjoint pruning, validation-selection, conformal-calibration, and test splits. Bounded score perturbations imply bounded conformal-quantile shifts and controlled set inflation, but do not make the generic coverage theorem CPP-specific. Final five-seed Qwen2.5-1.5B results at 50% sparsity show the largest gains on large-label tasks. On DBpedia-14, CPP-SparseGPT reduces mean set size from (10.1) to (8.6) while changing accuracy from (0.347) to (0.366); CPP-Wanda reduces (11.2) to (9.0) with an accuracy trade-off from (0.310) to (0.295). Across 15 dataset--sparsity cells, CPP-SparseGPT produces smaller sets in 13 and higher accuracy in 11. Matched controls show that generic supervised gradients explain much of the gain: true-label CPP is not statistically resolved from matched Wanda+SNIP, whereas threshold-aware candidate-label CPP reaches (7.8) mean set size at explicit accuracy and offline-compute costs. RoBERTa-base and Llama-3-8B diagnostics support transfer, but our claims remain limited to reliability-sensitive classification.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 19 of its 30 references that are in the map. First posted on arXiv 24 Aug 2026.

Builds on

19 of 30 references.

References · 19

  1. 2017On Calibration of Modern Neural Networkscited by 4,960
  2. 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
  3. 2016Learning without Forgettingcited by 2,496
  4. 2015Learning both Weights and Connections for Efficient Neural Networkscited by 2,031
  5. 2023SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shotcited by 1,121
  6. 2017PackNet: Adding Multiple Tasks to a Single Network by Iterative Pruningcited by 898
  7. 2021A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantificationcited by 792
  8. 2022DualPrompt: Complementary Prompting for Rehearsal-free Continual Learningcited by 642
  9. 2021Adaptive Conformal Inference Under Distribution Shiftcited by 375
  10. 2023Orthogonal Subspace Learning for Language Model Continual Learningcited by 312
  11. 2021Distribution-Free, Risk-Controlling Prediction Setscited by 288
  12. 2021Learn then Test: Calibrating Predictive Algorithms to Achieve Risk Controlcited by 271
7 more
  1. 2024Continual Learning of Large Language Models: A Comprehensive Surveycited by 234
  2. 2024Language Models with Conformal Factuality Guaranteescited by 131
  3. 2025Spurious Forgetting in Continual Learning of Language Modelscited by 51
  4. 2020Calibrated Prediction with Covariate Shift via Unsupervised Domain Adaptationcited by 43
  5. 2024Prune 'n Predict: Optimizing LLM Decision-making with Conformal Predictioncited by 18
  6. 2022The Open-World Lottery Ticket Hypothesis for OOD Intent Classificationcited by 2
  7. 2026Bayesian Lottery Ticket Hypothesiscited by 2

Built on by

Not cited yet.