Physics-Conditioned Synthesis of Internal Ice-Layer Thickness for Incomplete Layer Traces

• URL: http://arxiv.org/abs/2604.20783v1
• Date: Wed, 22 Apr 2026 17:10:48 GMT
• Title: Physics-Conditioned Synthesis of Internal Ice-Layer Thickness for Incomplete Layer Traces
• Abstract summary: Internal ice layers imaged by radar provide key evidence of snow accumulation and ice dynamics.<n> radar-derived layer boundary observations are often incomplete, with discontinuous traces and sometimes entirely missing layers.<n>Existing graph-based models for ice stratigraphy assume sufficiently complete layer profiles.
• Score: 1.6114012813668932
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Internal ice layers imaged by radar provide key evidence of snow accumulation and ice dynamics, but radar-derived layer boundary observations are often incomplete, with discontinuous traces and sometimes entirely missing layers, due to limited resolution, sensor noise, and signal loss. Existing graph-based models for ice stratigraphy generally assume sufficiently complete layer profiles and focus on predicting deeper-layer thickness from reliably traced shallow layers. In this work, we address the layer-completion problem itself by synthesizing complete ice-layer thickness annotations from incomplete radar-derived layer traces by conditioning on colocated physical features synchronized from physical climate models. The proposed network combines geometric learning to aggregate within-layer spatial context with a transformer-based temporal module that propagates information across layers to encourage coherent stratigraphy and consistent thickness evolution. To learn from incomplete supervision, we optimize a mask-aware robust regression objective that evaluates errors only at observed thickness values and normalizes by the number of valid entries, enabling stable training under varying sparsity without imputation and steering completions toward physically plausible values. The model preserves observed thickness where available and infers only missing regions, recovering fragmented segments and even fully absent layers while remaining consistent with measured traces. As an additional benefit, the synthesized thickness stacks provide effective pretraining supervision for a downstream deep-layer predictor, improving fine-tuned accuracy over training from scratch on the same fully traced data.

Related papers

• Geometric Collapse: When Vision Models Fail to Verify Physical Causality [52.43819179934414]
  We propose Scrambled Edges, a controlled counterfactual that injects salient edge-like cues.<n>With energy-matched and structure-matched controls, we isolate the effect of unsupported edge evidence from high-frequency energy and edge sparsity.<n>Across CNN/ViT/SSL depth predictors, Scrambled Edges induce up to 3.2x larger deviation from clean predictions than energy-matched noise.
  arXiv  Detail & Related papers  (2026-07-08T00:04:02Z)
• Provably Learning Diffusion Models under the Manifold Hypothesis: Collapse and Refine [60.669081685261965]
  Diffusion models generate high-dimensional data with remarkable quality.<n>How their training efficiently learns the score function remains theoretically unexplained.<n>We instantiate this principle as Score-induced Latent Diffusion (SiLD)
  arXiv  Detail & Related papers  (2026-05-16T16:51:10Z)
• Trust, but Verify: Peeling Low-Bit Transformer Networks for Training Monitoring [20.854288216118423]
  We propose a layer-wise peeling framework for training dynamics, in which each layer is locally optimized against representations of the trained model.<n>Experiments on decoder-only transformer models show that these layer-wise reference bounds can match or even surpass the trained model at various stages.
  arXiv  Detail & Related papers  (2026-05-04T17:30:50Z)
• K-STEMIT: Knowledge-Informed Spatio-Temporal Efficient Multi-Branch Graph Neural Network for Subsurface Stratigraphy Thickness Estimation from Radar Data [1.6114012813668932]
  K-STEMIT is a novel knowledge-informed, efficient, multi-branch-temporal graph neural network.<n>It incorporates physical data synchronized from the Model Atmospheric Regional physical weather model.<n>It consistently achieves the highest accuracy while maintaining near-surface efficiency.
  arXiv  Detail & Related papers  (2026-04-10T21:41:51Z)
• GroundingAnomaly: Spatially-Grounded Diffusion for Few-Shot Anomaly Synthesis [16.74245445257004]
  We propose GroundingAnomaly, a novel few-shot anomaly image generation framework.<n>Our framework introduces a Spatial Conditioning Module that leverages per-pixel semantic maps.<n>A Gated Self-Attention Module is designed to inject conditioning tokens into a frozen U-Net.
  arXiv  Detail & Related papers  (2026-04-09T14:34:50Z)
• NimbusGS: Unified 3D Scene Reconstruction under Hybrid Weather [83.14198482357388]
  We present NimbusGS, a unified framework for reconstructing high-quality 3D scenes from degraded multi-view inputs.<n>Unlike existing methods that target specific weather types, NimbusGS addresses the broader challenge of generalization.
  arXiv  Detail & Related papers  (2026-03-28T10:46:29Z)
• GRIT-LP: Graph Transformer with Long-Range Skip Connection and Partitioned Spatial Graphs for Accurate Ice Layer Thickness Prediction [1.6114012813668932]
  We introduce GRIT-LP, a graph transformer explicitly designed for polar ice-layer thickness estimation from polar radar.<n>GRIT-LP combines an inductive geometric graph learning framework with self-attention mechanism.<n>We demonstrate that GRIT-LP outperforms current state-of-the-art methods with a 24.92% improvement in mean root squared error.
  arXiv  Detail & Related papers  (2025-11-24T03:14:55Z)
• ST-GRIT: Spatio-Temporal Graph Transformer For Internal Ice Layer Thickness Prediction [0.7673339435080445]
  thickness and variability of internal ice layers in radar imagery is crucial for monitoring snow accumulation, assessing ice dynamics and reducing uncertainties in climate models.<n>In this work, we present-temporal graph for ice layer thickness, designed to process radar and capture relationships between shallow and deep ice layers.<n> ST-GRIT consistently outperforms current state-of-the-art methods and other baseline graph neural networks by achieving lower root mean-squared error.
  arXiv  Detail & Related papers  (2025-07-10T03:06:01Z)
• The Forward-Forward Algorithm: Characterizing Training Behavior [0.0]
  The Forward-Forward algorithm is an alternative learning method which consists of two forward passes rather than a forward and backward pass employed by backpropagation.<n>This work explores the dynamics of model and layer accuracy changes in Forward-Forward networks as training progresses in pursuit of a mechanistic understanding of their internal behavior.
  arXiv  Detail & Related papers  (2025-04-15T14:30:18Z)
• DispFormer: A Pretrained Transformer Incorporating Physical Constraints for Dispersion Curve Inversion [56.64622091009756]
  This study introduces DispFormer, a transformer-based neural network for $v_s$ profile inversion from Rayleigh-wave phase and group dispersion curves.<n>DispFormer processes dispersion data independently at each period, allowing it to handle varying lengths without requiring network modifications or strict alignment between training and testing datasets.
  arXiv  Detail & Related papers  (2025-01-08T09:08:24Z)
• SemiDDM-Weather: A Semi-supervised Learning Framework for All-in-one Adverse Weather Removal [57.52777076116241]
  Adverse weather removal aims to restore clear vision under adverse weather conditions. This paper presents a pioneering semi-supervised all-in-one adverse weather removal framework built on the teacher-student network.
  arXiv  Detail & Related papers  (2024-09-29T12:12:22Z)
• SINDER: Repairing the Singular Defects of DINOv2 [61.98878352956125]
  Vision Transformer models trained on large-scale datasets often exhibit artifacts in the patch token they extract. We propose a novel fine-tuning smooth regularization that rectifies structural deficiencies using only a small dataset.
  arXiv  Detail & Related papers  (2024-07-23T20:34:23Z)
• A Generic Shared Attention Mechanism for Various Backbone Neural Networks [53.36677373145012]
  Self-attention modules (SAMs) produce strongly correlated attention maps across different layers. Dense-and-Implicit Attention (DIA) shares SAMs across layers and employs a long short-term memory module. Our simple yet effective DIA can consistently enhance various network backbones.
  arXiv  Detail & Related papers  (2022-10-27T13:24:08Z)
• Self-Supervised Training with Autoencoders for Visual Anomaly Detection [61.62861063776813]
  We focus on a specific use case in anomaly detection where the distribution of normal samples is supported by a lower-dimensional manifold. We adapt a self-supervised learning regime that exploits discriminative information during training but focuses on the submanifold of normal examples. We achieve a new state-of-the-art result on the MVTec AD dataset -- a challenging benchmark for visual anomaly detection in the manufacturing domain.
  arXiv  Detail & Related papers  (2022-06-23T14:16:30Z)

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.