{"id":"256dbf62-dfc8-4189-a5cf-cba8e8b7534f","arxiv_id":"2606.18787","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":6.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"A learned per-query radius selector predicts continuous support radii for a frozen LoSF-UDF backbone using off-grid targets from parabolic interpolation of UDF error curves, yielding improved fine-scale reconstruction accuracy.","lead":"The paper proposes training a neural network to predict a suitable support radius for each point when reconstructing surfaces from point clouds using Unsigned Distance Fields. This could help make consumer 3D scanning more accurate for AR/VR and indoor modeling without changing the core reconstruction model.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.3","headline":"Parabolic interpolation of cached UDF error curves may yield unreliable supervision targets","rationale":"The reader's weakest assumption directly identifies the load-bearing step in the training pipeline. No additional internal inconsistency appears from the abstract description of the method; the concern is precisely the unverified quality of the supervision signal.","tokens_in":1578,"tokens_out":286,"duration_ms":13898,"concrete_test":"For a held-out set of query points, densely sample the UDF error curve at 0.01 increments around the reported target radius, locate the empirical minimum, and measure the absolute deviation from the parabolic-interpolated value; if the median deviation exceeds 5% of the radius range or the interpolated value is not the minimum in >15% of cases, the supervision signal is unreliable.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim requires that the learned radius selector, trained on off-grid targets from parabolic interpolation of cached UDF error curves, produces radii that improve fine-scale reconstruction when plugged into the frozen LoSF-UDF backbone. This holds only if the interpolated targets accurately locate the radius minimizing reconstruction error for each query and generalize across heterogeneous local geometry. The abstract supplies no description of how error curves are cached, the sampling density used, or any check that the parabolic fit recovers the true minimum rather than an artifact of the curve shape.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The paper claims that a learned per-query radius selector, trained on off-grid target radii obtained by parabolic interpolation of cached UDF error curves, can be plugged into a frozen LoSF-UDF backbone to improve fine-scale surface reconstruction accuracy from point clouds compared to fixed or heuristic radius selection.","tokens_in":1664,"tokens_out":312,"duration_ms":17852,"significance":"If the interpolated targets prove reliable and the selector generalizes, the approach could address a key limitation of local UDF methods by adapting support radii to heterogeneous geometry, with potential benefits for consumer 3D applications. The frozen-backbone design is a practical strength that avoids retraining the full model.","major_comments":[{"comment":"Abstract: the claim that 'experiments show improved fine-scale reconstruction accuracy' supplies no quantitative results, baselines, dataset details, or ablation studies, preventing verification of the central experimental claim.","section":"Abstract"},{"comment":"Training procedure (described in abstract): the off-grid targets are generated by parabolic interpolation of cached UDF error curves; the manuscript must show that these targets recover the true error-minimizing radius (e.g., via denser sampling validation or error-curve analysis) rather than curve-shape artifacts, as this supervision signal is load-bearing for the learned selector.","section":"Abstract"}],"minor_comments":[{"comment":"LoSF-UDF acronym is used without expansion.","section":"Abstract"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the constructive comments on our manuscript. We address each major comment below and will incorporate revisions to strengthen the presentation of results and validation of the supervision signal.","responses":[{"response":"We agree that the abstract lacks the necessary quantitative details to support the central claim. In the revised manuscript we will expand the abstract to report specific metrics (e.g., Chamfer distance and normal consistency improvements), name the datasets and baselines, and briefly reference the ablation studies that demonstrate the benefit of the learned selector.","revision_made":"yes","referee_comment":"[Abstract] Abstract: the claim that 'experiments show improved fine-scale reconstruction accuracy' supplies no quantitative results, baselines, dataset details, or ablation studies, preventing verification of the central experimental claim."},{"response":"We acknowledge that explicit validation of the parabolic targets is required. The current manuscript does not contain a dedicated denser-sampling comparison or error-curve analysis; we will add such validation (including quantitative deviation statistics and example curves) in a new subsection of the experiments to confirm that the interpolated radii reliably approximate the true minima.","revision_made":"yes","referee_comment":"[Abstract] Training procedure (described in abstract): the off-grid targets are generated by parabolic interpolation of cached UDF error curves; the manuscript must show that these targets recover the true error-minimizing radius (e.g., via denser sampling validation or error-curve analysis) rather than curve-shape artifacts, as this supervision signal is load-bearing for the learned selector."}],"tokens_in":1166,"tokens_out":341,"duration_ms":19107,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The paper's main move is replacing fixed or curvature-heuristic radii in local-patch UDF reconstruction with a learned per-query predictor that outputs a continuous support radius. The selector trains on off-grid targets taken from parabolic interpolation of cached UDF error curves and then slots into a frozen LoSF-UDF backbone.\n\nThis is new relative to the usual one-dimensional heuristics. It directly targets the problem that local geometry varies within a single scan, which fixed radii or simple curvature rules cannot handle well. Keeping the backbone frozen is a practical choice that lowers the barrier to trying the idea.\n\nThe soft spot is the complete absence of numbers. The abstract claims better fine-scale accuracy but gives no baselines, datasets, ablation results, or even the size of the reported gain. Without those, it is impossible to judge whether the improvement is real or worth the added complexity. The stress-test concern also lands: parabolic interpolation of cached error curves can only supply good targets if the curves are densely sampled and the fit actually recovers the true minimum rather than an artifact. The abstract says nothing about sampling density or caching procedure, so that assumption remains unchecked.\n\nThe work is aimed at practitioners doing consumer-grade 3D capture with UDF methods. A referee could usefully examine the experimental setup and test whether the interpolated targets generalize across different local geometries. I would send it for peer review to get those details evaluated rather than desk-reject on the abstract alone.","headline":"Learned continuous radius selector trained on parabolic UDF error targets is the core idea, but the abstract supplies no numbers or method details so the gains and target reliability stay unverified.","tokens_in":2131,"tokens_out":372,"would_cite":false,"duration_ms":16340,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"A neural selector learns a continuous per-query support radius to improve unsigned distance field surface reconstruction from point clouds.","keywords":["point cloud reconstruction","unsigned distance field","radius estimation","surface reconstruction","local patch methods","neural network selector","fine-scale accuracy"],"falsifier":"Measure reconstruction error on a held-out set of point clouds whose true per-point optimal radii are known from dense ground-truth surfaces; if the selector's predictions do not reduce error relative to the best fixed radius, the claim fails.","tokens_in":2465,"feed_emoji":"📐","tokens_out":606,"duration_ms":13596,"temperature":0.7,"pith_summary":"Local UDF methods reconstruct surfaces from point clouds but depend on a support radius that is usually fixed or set by a simple curvature rule, which struggles with varying local geometry. The paper trains a small network to output a continuous radius for each query point, using target values created by parabolic interpolation on cached error curves from a frozen backbone model. This selector is inserted into the existing pipeline without retraining the main UDF network. Experiments indicate higher accuracy on fine details compared with fixed or heuristic radii. The approach keeps the core reconstruction lightweight while adapting the radius to local conditions.","feed_headline":"Neural selector picks per-point radii to sharpen UDF reconstructions","feed_subtitle":"Training on interpolated error curves lets the network output continuous support radii that raise fine detail accuracy in a frozen backbone.","key_machinery":"The learned per-query radius selector: a neural network that takes a query point and local patch and outputs a single continuous radius value used as the support radius for the UDF evaluation.","core_discovery":"The paper shows that a learned per-query radius selector, trained on off-grid target radii from parabolic interpolation of cached UDF error curves, can be plugged into a frozen LoSF-UDF backbone to predict continuous support radii and thereby raise fine-scale reconstruction accuracy on point cloud data.","pith_inferences":["The same interpolation-based supervision could be tested on signed distance fields or other implicit representations.","Integrating the selector with online adaptation might allow radius adjustment during live scanning sessions.","Consumer devices that capture point clouds could ship the selector as a lightweight add-on to reduce manual parameter tuning."],"forward_implications":["Reconstruction accuracy improves on scenes with mixed curvature without changing the backbone model size.","The method supports continuous rather than discrete radius choices at query time.","Freezing the UDF backbone keeps training cost low while still adapting to heterogeneous geometry.","The selector can be swapped into other local-patch UDF pipelines that use a support radius."],"fun_headline_variants":["Learned per-query radius selector for UDF reconstruction","Per-query radii selected from interpolated UDF errors","Continuous radius predictor for frozen LoSF-UDF","Radius selector trained on parabolic UDF error curves"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"That the target radii obtained by parabolic interpolation of UDF error curves supply reliable and generalizable supervision for training the selector.","fun_headline_variants_meta":{"raw":{"variants":["Learned per-query radius selector for UDF reconstruction","Per-query radii selected from interpolated UDF errors","Continuous radius predictor for frozen LoSF-UDF","Radius selector trained on parabolic UDF error curves"]},"model":"grok-4.3","cost_usd":0.005505,"raw_usage":{"total_tokens":2569,"prompt_tokens":519,"num_sources_used":0,"completion_tokens":58,"cost_in_usd_ticks":55049500,"prompt_tokens_details":{"text_tokens":519,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":1992,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":519,"tokens_out":58,"duration_ms":12150,"temperature":1.0,"reasoning_tokens":1992,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-06-26T21:57:26.263372+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"Measure reconstruction error on a held-out set of point clouds whose true per-point optimal radii are known from dense ground-truth surfaces; if the selector's predictions do not reduce error relative to the best fixed radius, the claim fails.","supporting_citations":[],"review_version":1}