{"id":"a68c920e-0c59-4d97-a89a-9974cb1d027b","arxiv_id":"1907.10834","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":3.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"Framelet pooling decomposes large medical image learning tasks into smaller ones via filter banks to reduce neural network complexity and computation while preserving details.","lead":"The paper proposes using framelet filter banks to transform high-dimensional medical images into lower-dimensional components for deep learning networks. This aims to cut computational costs while keeping performance similar to full-size training.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.3","headline":"No significant objection identified","rationale":"The reader's weakest assumption correctly isolates the information-preservation step as the key condition. Because the abstract explicitly ties the claim to experimental comparability and no contradictory or underspecified element appears in the argument structure, the concern does not rise to a load-bearing objection that would alter the UNVERDICTED verdict.","tokens_in":1655,"tokens_out":237,"duration_ms":11877,"concrete_test":"Reproduce the reported experiments on the medical imaging datasets using the exact framelet decomposition described; compare test accuracy and training time against the unreduced baseline. If accuracy remains within 2% while wall-clock time drops by the claimed factor, the preservation claim holds for those tasks.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that framelet filter banks transform high-dimensional medical image data into lower-dimensional components while preserving all necessary detailed information, enabling comparable downstream performance at reduced computational cost. The abstract states that experiments confirm comparability to the unreduced baseline. No internal inconsistency, missing assumption, or unsupported step is evident from the provided description of the method and results.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The paper proposes a framelet-pooling aided deep learning network for high-dimensional medical image data. It uses framelet filter banks to decompose large input matrices into lower-dimensional components while aiming to preserve detailed information, thereby reducing neural network complexity and computational costs during training. The central claim is that this approach achieves performance comparable to the standard unreduced method, as supported by various experiments.","tokens_in":1687,"tokens_out":268,"duration_ms":14553,"significance":"If the experimental results hold, the method could provide a practical dimensionality-reduction technique for medical imaging tasks where data size poses computational challenges, potentially enabling more efficient training without major performance loss. The framelet-based decomposition is a novel angle on pooling, but its significance is hard to gauge given the absence of concrete validation details.","major_comments":[{"comment":"Abstract: The assertion that 'various experiments show that our method is comparable to the standard unreduced learning method, while reducing computational burdens' is load-bearing for the central claim yet provides no information on datasets, metrics, baselines, controls, or quantitative results. This omission prevents assessment of whether the framelet decomposition actually preserves task-relevant information without degradation.","section":"Abstract"}],"minor_comments":[],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for their constructive comment. We address it point-by-point below and indicate the revisions we will make.","responses":[{"response":"We agree that the abstract is too terse on these points and that adding concrete details would strengthen the central claim. In the revised version we will expand the abstract to name the medical imaging datasets, report the quantitative metrics and baselines used, and include the key performance numbers showing comparability with reduced computation. These details already appear in the experimental sections of the manuscript; the revision will simply surface them in the abstract as well.","revision_made":"yes","referee_comment":"[Abstract] Abstract: The assertion that 'various experiments show that our method is comparable to the standard unreduced learning method, while reducing computational burdens' is load-bearing for the central claim yet provides no information on datasets, metrics, baselines, controls, or quantitative results. This omission prevents assessment of whether the framelet decomposition actually preserves task-relevant information without degradation."}],"tokens_in":1180,"tokens_out":223,"duration_ms":10369,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The main point is that the paper describes using framelet transforms to break high-dimensional medical image inputs into lower-dimensional components for deep learning, with the goal of cutting network complexity and compute while keeping the necessary details intact through the filter banks. It then splits the overall task into several smaller ones. This is a direct application of existing framelet techniques rather than a new derivation or framework. The description of how the decomposition works is clear enough on paper and could be useful as a practical template for efficiency in medical imaging pipelines where input sizes are large. What stands out is the emphasis on preserving information via the banks instead of simple downsampling. The soft spot is the experimental section. The abstract states that various experiments show performance comparable to the unreduced baseline, yet it supplies zero information on the datasets used, the network architectures, the metrics, the baselines, or any controls for the reduction step. Without those, there is no evidence to evaluate whether the information preservation actually holds or whether hidden costs or artifacts appear downstream. The central assumption that all task-relevant detail survives the transform therefore remains untested in the provided text. This work would mainly interest researchers already focused on computational shortcuts for large medical scans. It deserves a serious referee if the full manuscript contains reproducible experiments with fair comparisons and clear implementation details, because the underlying idea is coherent and the efficiency angle is relevant even if the novelty is limited. I would send it to review to get the results checked rather than desk reject on the abstract alone.","headline":"This applies known framelet filter banks as a pooling step for high-dim medical images but offers no experimental details to check the claims.","tokens_in":2186,"tokens_out":368,"would_cite":false,"duration_ms":19540,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":{"model":"grok-4.3","evidence":[{"relation":"unclear","rs_module":"IndisputableMonolith/Cost/FunctionalEquation.lean; Foundation/AlexanderDuality.lean","rs_theorem":"washburn_uniqueness_aczel; alexander_duality_circle_linking","paper_passage":"By transforming high dimensional data into low dimensional components by filter banks with preserving detailed information... framelet decomposition operator W^(1), W^(2) ... unitary extension principle"}],"headline":"Framelet pooling for medical-image dimensionality reduction has no overlap with RS forcing chain","alignment":"orthogonal","rationale":"The paper's core machinery (framelet packet transforms W^(k) via unitary extension principle, down-sampling by 2^2 per level, parallel training on low-dimensional sub-bands) is a standard wavelet-theoretic dimensionality-reduction technique for U-Net reconstruction tasks. RS derives J-cost, φ-ladder, 8-tick periodicity and D=3 from a single distinction (AbsoluteFloorClosure, AlexanderDuality, Cost/FunctionalEquation). No shared structure, cost function, ratio symmetry or parameter-free constant derivation appears.","tokens_in":51441,"confidence":"high","tokens_out":258,"duration_ms":6297,"cache_read_input_tokens":38528,"cache_creation_input_tokens":0},"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"Framelet pooling decomposes high-dimensional medical images into low-dimensional components via filter banks to cut neural network complexity while preserving details.","keywords":["framelet pooling","deep learning","medical imaging","dimensionality reduction","computational efficiency","filter banks","high-dimensional data","neural network complexity"],"falsifier":"An experiment in which a deep network trained on the framelet-decomposed components shows measurably lower accuracy or higher error than an identical network trained directly on the original high-dimensional medical images for the same diagnostic task.","tokens_in":2540,"feed_emoji":"","tokens_out":592,"duration_ms":16701,"temperature":0.7,"pith_summary":"The paper presents a framelet-pooling method to ease computational demands when training deep networks on large medical image matrices. Filter banks break the high-dimensional inputs into smaller components without discarding necessary details. This decomposition turns one large learning task into several smaller ones, lowering network size and training expense. Experiments indicate the approach yields accuracy comparable to standard unreduced networks. The work targets the curse of dimensionality and related hurdles in medical image analysis.","feed_headline":"Framelet pooling cuts deep learning costs for large medical images","feed_subtitle":"Filter banks break high-dimensional inputs into smaller components while matching unreduced network accuracy.","key_machinery":"Framelet filter banks that decompose high-dimensional inputs into low-dimensional components while preserving detailed information.","core_discovery":"By applying framelet filter banks to transform high-dimensional medical image data into low-dimensional components while preserving detailed information, the framelet-pooling aided network reduces neural network complexity and computational costs by decomposing large-scale tasks into multiple small-scale tasks, achieving performance comparable to the standard unreduced learning method.","pith_inferences":["The same decomposition could apply to other high-dimensional scientific or imaging datasets where direct input size limits training.","Parallel processing of the independent low-dimensional components might further speed up inference.","If the filter banks introduce any domain-specific bias, retraining the downstream network on each component separately could compensate.","The method may allow training on larger image resolutions than would otherwise fit in memory."],"forward_implications":["Network complexity drops because each sub-task operates on smaller inputs.","Training computational costs fall by splitting one large problem into several smaller ones.","Performance stays comparable to methods that use the full unreduced data.","The approach mitigates the curse of dimensionality for medical image matrices.","Generalization issues tied to high input size are addressed through the decomposition."],"fun_headline_variants":["Framelet pooling trims deep learning costs for big medical images","Framelets lower costs for deep learning on high dimensional medical images","Framelet filter banks cut neural network complexity on medical images","Framelets cut deep learning costs by decomposing medical image tasks"],"cache_read_input_tokens":64,"weakest_assumption_plain":"The framelet filter banks preserve all information needed for the downstream task without losses or artifacts that would lower model performance relative to the original high-dimensional inputs.","fun_headline_variants_meta":{"raw":{"variants":["Framelet pooling trims deep learning costs for big medical images","Framelets lower costs for deep learning on high dimensional medical images","Framelet filter banks cut neural network complexity on medical images","Framelets cut deep learning costs by decomposing medical image tasks"]},"model":"grok-4.3","cost_usd":0.012286,"raw_usage":{"total_tokens":5302,"prompt_tokens":560,"num_sources_used":0,"completion_tokens":61,"cost_in_usd_ticks":122862000,"prompt_tokens_details":{"text_tokens":560,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":4681,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":560,"tokens_out":61,"duration_ms":29065,"temperature":1.0,"reasoning_tokens":4681,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-05-24T16:24:14.686472+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"An experiment in which a deep network trained on the framelet-decomposed components shows measurably lower accuracy or higher error than an identical network trained directly on the original high-dimensional medical images for the same diagnostic task.","supporting_citations":[],"review_version":1}