Relative Repairability: A Calibration-Based Diagnostic for High-Sparsity Post-Pruning Allocation

• URL: http://arxiv.org/abs/2605.25508v1
• Date: Mon, 25 May 2026 07:12:38 GMT
• Title: Relative Repairability: A Calibration-Based Diagnostic for High-Sparsity Post-Pruning Allocation
• Abstract summary: Relative Repairability (RR) is a calibration based diagnostic that compares the raw activation distortion caused by layerwise pruning with the residual distortion left after channelwise variance matching repair.<n>RR is most useful near an architecture dependent recoverability transition, where standard structural or magnitude based allocation priors begin to lose reliability but post repair recovery has not yet fully collapsed.<n>These results suggest that high sparsity pruning should allocate not only retained weights, but also repairable damage.
• Score: 12.288779306467768
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: At very high sparsity, neural network pruning does more than decide which weights remain. It also determines where pruning induced damage is placed across the network, and whether that damage can be recovered by a fixed lightweight repair procedure. We study this problem through the lens of repair conditioned sparsity allocation. We introduce Relative Repairability (RR), a calibration based diagnostic that compares the raw activation distortion caused by layerwise pruning with the residual distortion left after channelwise variance matching repair. RR estimates the fraction of local damage that remains after repair, using only unlabeled calibration data. Across ResNet18, ResNet34, and VGG16 BN on CIFAR10 and CIFAR100, we find that RR is not a universally dominant allocation rule. Instead, it is most useful near an architecture dependent recoverability transition, where standard structural or magnitude based allocation priors begin to lose reliability but post repair recovery has not yet fully collapsed. On CIFAR100 ResNet18, a fine grained sweep shows that RR improves over ERK across the central transition band and surpasses LAMP near the upper part of this band. A projection forced ablation further shows that capped ERK can over protect projection layers, shifting excessive sparsity onto regular convolutions and reducing post repair recovery. These results suggest that high sparsity pruning should allocate not only retained weights, but also repairable damage.

Related papers

• ResLRP: The Role of Residual Cancellation in Attribution Instability in Vision Transformers [58.411900827911346]
  We show that the missing ingredient is the treatment of residual connections.<n>We introduce Residual-aware Layer-wise Relevance Propagation (ResLRP)<n>ResLRP is a simple extension of LRP whose propagation rules explicitly account for cancellations in residual branches.
  arXiv  Detail & Related papers  (2026-09-15T13:17:30Z)
• LoRA-RC: Reservoir Computing with Low-Rank Adaptation [0.0]
  LoRA-RC adapts the recurrent matrix through a low-rank correction driven by streaming prediction errors.<n>On a Lorenz system with an abrupt parameter drift, LoRA-RC cuts post-drift prediction error by 56% versus a fixed RC and 51% versus readout-only adaptation.
  arXiv  Detail & Related papers  (2026-09-11T01:19:48Z)
• Let Confidence Change, Not the Prediction: Prediction-Preserving Repair for Post-hoc Calibration [4.300307544233363]
  Post-hoc calibration corrects reported confidence, yet a multiclass calibrator can also change the associated top-1 prediction.<n>We propose Calibrator-Output Repair for Top-1 Decision Preservation (CORD), the first post-fit adapter to impose exact prediction preservation by repairing the full calibrated probability vector.
  arXiv  Detail & Related papers  (2026-09-01T11:03:30Z)
• Universal recovery in approximate quantum error correction [53.73800998389749]
  Universal recovery is the existence of a single recovery map that corrects an entire family of error channels.<n>In exact quantum error correction, linearity guarantees that a code correcting a given error set also corrects every channel whose Kraus operators lie in its linear span.<n>We show that a single recovery map can simultaneously correct every channel controlled by a given error set.
  arXiv  Detail & Related papers  (2026-08-29T00:25:52Z)
• HARP: Hadamard-Preconditioned Adaptive Rotation Processor for Extreme LLM Quantization [48.87937677659571]
  We introduce HARP (Hadamard-preconditioned Adaptive Rotation Processor), a learnable structured two-sided processor that replaces fixed Hadamard mixing.<n>Across 2-4 bit settings on models ranging from 1B to 70B parameters, HARP improves perplexity and zero-shot accuracy over fixed RHT.<n>HarP preserves deployment efficiency, reaching 128 tok/s versus 61 tok/s for FP16.
  arXiv  Detail & Related papers  (2026-05-28T12:24:15Z)
• Guarded Repair for Harm-Aware Post-hoc Replacement of LLM Mathematical Reasoning [0.0]
  GuardedRepair is a best-of-N repair framework that diagnoses cached reasoning traces, selectively triggers repair, and accepts answer-changing candidates.<n>We show that GuardedRepair substantially improves the fixed/broken tradeoff, while also revealing that replacement risk is reduced rather than eliminated.
  arXiv  Detail & Related papers  (2026-05-23T14:51:13Z)
• Clipping Bottleneck: Stabilizing RLVR via Stochastic Recovery of Near-Boundary Signals [83.0127582612634]
  Near-boundary Rescue (NSR) is a minimal, plug-and-play modification that retains out-of-bound tokens to recover lost signals.<n>NSR substantially improves training stability and delivers consistent gains over strong baselines such as DAPO and GSPO.
  arXiv  Detail & Related papers  (2026-05-21T16:45:31Z)
• How Sparsity Allocation Shapes Label-Free Post-Pruning Recoverability [4.9112670395105305]
  Post-pruning repair methods can partially recover collapsed sparse models, but their effectiveness depends on the sparse model left by the upstream pruning allocation.<n>This paper studies how sparsity allocation shapes post-repair recoverability under a fixed activation-statistic repair backend.
  arXiv  Detail & Related papers  (2026-05-21T04:03:24Z)
• Adaptive Signal Resuscitation: Channel-wise Post-Pruning Repair for Sparse Vision Networks [8.5133415329194]
  One-shot magnitude pruning can cause severe accuracy collapse in the high-sparsity regime.<n>We propose Adaptive Signal Resuscitation (ASR) as a training-free channel-wise repair method.<n>ASR estimates a variance-matching correction for each output channel and stabilizes it with a data-driven shrinkage rule.
  arXiv  Detail & Related papers  (2026-05-20T17:19:01Z)
• DegBins: Degradation-Driven Binning for Depth Super-Resolution [57.61783461543096]
  DegBins is a novel DSR framework that leverages degradation-driven binning to adaptively enhance residual modeling.<n>DegBins consistently outperforms existing state-of-the-art methods in terms of accuracy, robustness, and generalization.
  arXiv  Detail & Related papers  (2026-05-10T16:17:11Z)
• Restoration-Aligned Generative Flow Models for Blind Motion Deblurring [57.425021485907756]
  We present DeFlow, a framework that resolves misalignment by reformulating the flow trajectory itself.<n>Under this formulation, the standard flow matching loss naturally takes the form of a residual loss.<n>Experiments on GoPro, HIDE, RealBlur, and RWBI demonstrate that DeFlow achieves strong restoration fidelity and perceptual realism.
  arXiv  Detail & Related papers  (2026-05-09T10:05:51Z)
• Exactness Matters for Physical Rule Enforcement [4.230349220081444]
  It remains unclear when stronger physical rule enforcement becomes reliable and when it becomes a source of distribution shift.<n>We study this question through operator exactness, meaning whether the repair map is the identity on the target manifold.<n> Controlled mismatch, screened cleanup, adaptive gating, and external-backbone checks show that the best approximate-regime operating point can be raw or near-identity.
  arXiv  Detail & Related papers  (2026-05-08T07:59:59Z)
• ANNEAL: Adapting LLM Agents via Governed Symbolic Patch Learning [12.44509691077682]
  We introduce ANNEAL, a neuro-symbolic agent that converts recurring failures into governed symbolic edits of a process knowledge graph.<n>Its core mechanism, Failure-Driven Knowledge Acquisition (FDKA), localizes the responsible operator, synthesizes a typed patch through constrained LLM generation, and validates the proposal.<n>Across four domains and 27 multi-seed runs, ANNEAL is the only evaluated system that commits persistent structural repairs.
  arXiv  Detail & Related papers  (2026-05-04T05:24:03Z)
• Beyond calibration: estimating the grouping loss of modern neural networks [68.8204255655161]
  Proper scoring rule theory shows that given the calibration loss, the missing piece to characterize individual errors is the grouping loss. We show that modern neural network architectures in vision and NLP exhibit grouping loss, notably in distribution shifts settings.
  arXiv  Detail & Related papers  (2022-10-28T07:04:20Z)
• Localized Calibration: Metrics and Recalibration [133.07044916594361]
  We propose a fine-grained calibration metric that spans the gap between fully global and fully individualized calibration. We then introduce a localized recalibration method, LoRe, that improves the LCE better than existing recalibration methods.
  arXiv  Detail & Related papers  (2021-02-22T07:22:12Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.