{"id":"3f078464-b8f6-4989-9941-c7ee55793caf","arxiv_id":"2508.13287","paper_version":4,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":4.0,"correctness_risk":"high","formal_verification":"none","parameter_count":1,"one_line_summary":"The abstract promises pose-free internal volume reconstruction and language-guided segmentation from sparse slices, but no supporting manuscript text is present, so the result is unverifiable.","lead":"The listed paper is supposed to describe InnerGS, a 3D Gaussian method for internal-scene reconstruction and text-guided medical segmentation, but the supplied full text is an unrelated condensed-matter preprint about generalized Brillouin zone fragmentation. Because the submitted text does not support the abstract, the InnerGS claims cannot be evaluated.","discovery_kind":"unclear","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The submitted full text is a different paper, so the abstract's reconstruction and segmentation claims have no supporting derivation or experiments.","rationale":"The reader's verdict of UNVERDICTED is appropriate. My concern is not a scientific objection to InnerGS itself but an evidence-level barrier: the supplied full text is an unrelated condensed-matter paper, so none of the abstract's claims can be checked. This is consistent with the reader's observation that no supporting derivation or experiments are present, though I frame the issue more broadly than the data-sufficiency premise. I do not recommend REJECT because the method may be valid in a corrected submission; the correct action is to withhold judgment until the actual manuscript is available. The unrelated GBZ paper contains its own derivations and photonic simulations, and those may be legitimate on their own terms, but they cannot support InnerGS.","tokens_in":35071,"tokens_out":2836,"duration_ms":30006,"concrete_test":"First, programmatically fetch the full text or PDF for arXiv:2508.13287 and compare its title, abstract, and section headings against the submitted body; if they differ, confirm the mismatch. Second, if the correct InnerGS manuscript is recovered, check whether it includes (a) a derivation of the inner-density Gaussian parameterization, (b) an optimization objective that uses only sliced image data and no camera poses, and (c) quantitative reconstruction or segmentation results on at least one medical or internal-scene benchmark. Absent all three, the abstract-level claims remain unverifiable.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The abstract's central claim—that InnerGS reconstructs smooth internal volumes from sparse sliced data without camera poses and supports text-guided segmentation—is unsupported by the submitted manuscript body. The body, including the supplement, is arXiv:2508.13275, 'Generalized Brillouin Zone Fragmentation' by Meng, Ang, and Lee, a condensed-matter physics paper on non-Hermitian GBZ fragmentation. It contains no 3D Gaussian splatting model, no slice-registration or density-supervision formulation, no segmentation pipeline, no experiments, and no evaluation against any medical or internal-scene dataset. Therefore the key condition for the central claim to hold—that the manuscript actually presents the InnerGS method—fails at the evidence level. The reader's weakest assumption, that sparse slices alone sufficiently constrain pose-free volume reconstruction, is plausible and unaddressed, but it is secondary: even the method's basic existence, its optimization objective, and its reported fidelity are not testable from this document. This is not an argument against the method's validity, but a claim that no verifiable scientific content supports it in the submitted record.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript titled \"InnerGS: Internal Scenes Reconstruction and Segmentation via Factorized 3D Gaussian Splatting\" claims a pose-free 3D Gaussian splatting framework that reconstructs smooth internal structures from sparse sliced data and supports text-guided segmentation of medical scenes. The abstract promises a CUDA implementation and states that the method is plug-and-play and modality-agnostic. However, the submitted full text and supplement are a different paper entirely: arXiv:2508.13275, \"Generalized Brillouin Zone Fragmentation,\" a condensed-matter physics manuscript on non-Hermitian skin effects and GBZ fragmentation. The body contains no equations, algorithmic description, experimental setup, baselines, or numerical results for InnerGS. Thus, in the submitted record, the central claims of the abstract are unsupported by any verifiable scientific content.","tokens_in":35265,"tokens_out":2378,"duration_ms":26102,"significance":"If the abstract's claims were supported, the work could be significant for internal-scene reconstruction and medical segmentation: a pose-free approach to continuous volumetric reconstruction from sparse slices would address an important practical bottleneck, and text-guided segmentation of reconstructed volumes is a useful downstream capability. However, the significance assessment cannot go beyond the abstract, because the submitted manuscript text does not present the method, its optimization objective, its input constraints, or any evaluation. There are no machine-checked proofs, parameter-free derivations, reproducible experiments, or falsifiable predictions for InnerGS in this artifact. The evaluation value of the submission as a scientific record is therefore limited to the abstract's promises, which are not backed by the body.","major_comments":[{"comment":"The submitted manuscript body and supplement are a different paper, arXiv:2508.13275, 'Generalized Brillouin Zone Fragmentation,' which treats non-Hermitian lattice models. There is no derivation of the 'inner 3D Gaussian distribution,' no slice-registration or density-supervision objective, no language-feature integration, and no segmentation pipeline. The abstract's central claim of a 3D Gaussian splatting method for internal-scene reconstruction is therefore not supported by any content in the submitted text.","section":"Full text and Supplement"},{"comment":"The abstract's assertion that the method 'eliminates the need for camera poses' and is 'inherently compatible with any data modalities' is a strong data-sufficiency and architectural claim. The manuscript provides no problem formulation, no identifiability analysis, and no experiments showing that sparse sliced data without poses is sufficient to constrain learned volumetric Gaussians. This load-bearing assumption is stated only in the abstract and is never addressed in the body.","section":"Abstract (pose-free reconstruction claim)"},{"comment":"The submitted artifact contains no quantitative results for InnerGS: no datasets, no baselines, no error bars, and no visual comparisons. Sections and equations in the body belong to the physics paper, such as Eq. (1) for the lattice Hamiltonian and Eq. (5) for the composition IPR. Without any experimental section or evaluation, the claimed high-fidelity reconstruction and text-guided segmentation cannot be verified.","section":"None (experiments are absent)"}],"minor_comments":[{"comment":"The paper's title, abstract, and GitHub link describe InnerGS, while the body heading identifies the text as 'Generalized Brillouin Zone Fragmentation.' This mismatch prevents a reader from extracting the proposed method from the manuscript.","section":"Title/Abstract vs. Body"},{"comment":"The abstract points to https://github.com/Shuxin-Liang/InnerGS for a CUDA implementation, but the manuscript text describes no code architecture, dependencies, or usage instructions. An external repository cannot substitute for the required method description inside the paper.","section":"Abstract, GitHub link"},{"comment":"The supplement references 'Sect. I of [76]' and other internal cross-references from the physics paper, indicating that the content displacement extends into the supplementary material as well.","section":"Supplement"}],"recommendation":"reject","confidential_remarks":"As submitted, the manuscript does not contain the work described in its abstract. The body and supplement are a substantially different paper from a different field. This goes beyond a local technical error; the central claim has no supporting derivation or evaluation in the submitted record. The editor may wish to verify whether a metadata or file-submission error occurred, but in its current form the manuscript cannot be meaningfully reviewed and does not meet the standards for publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague—this submission is not reviewable as InnerGS. The metadata and abstract describe a 3D Gaussian splatting method for internal scene reconstruction and text-guided segmentation. The uploaded full text is an unrelated condensed-matter paper, arXiv:2508.13275, about generalized Brillouin zone fragmentation. There are no equations, no experiments, no baselines for InnerGS anywhere in the record. So the honest summary is that the central claim has no supporting document.\n\nWhat is new, for what it's worth: the abstract sketches a sensible direction—factorized 3D Gaussians for volumetric interiors, no camera poses, plus language embeddings for segmentation. That is a plausible research program in medical imaging, and the GitHub link suggests the authors may have code. But a plausible abstract is not a paper. There is no way to check the method, the training objective, the data, or the claimed \"plug-and-play\" compatibility.\n\nThe reader's scores are, if anything, generous. The circularity concern and the slice-sufficiency assumption are real questions, but they are secondary. The primary issue is that the manuscript body contradicts the title. This is either a mis-upload or a metadata error, and it should be returned to the authors immediately. If the actual InnerGS paper exists, it belongs in a corrected submission; if not, the abstract is an unsupported claim.\n\nI cannot recommend peer review for this record. A referee would have nothing to assess. Desk-reject and ask the authors to resubmit the correct manuscript. For the record, the GBZ physics paper itself is a different matter and may be fine, but it is not what was submitted.","headline":"The submitted record is unreviewable as InnerGS: the abstract describes a 3D Gaussian splatting method, but the uploaded full text is an unrelated condensed-matter physics paper.","tokens_in":35752,"tokens_out":2102,"would_cite":false,"duration_ms":22246,"reading_group":"no","serious_thinker":"no","would_accept_peer_review":false},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"3D Gaussian splatting can, the paper argues, reconstruct object interiors from sparse slices without camera poses and answer text queries about them.","keywords":["3D Gaussian Splatting","internal scene reconstruction","sparse slice data","pose-free reconstruction","text-guided segmentation","medical volume rendering","continuous volumetric density"],"falsifier":"Feed the proposed model a known volume, such as a CT-scanned phantom with internal structures, using only a sparse set of slices and no pose information, then compare the reconstructed density against the ground truth; if the interior accuracy degrades as slice spacing grows, the data-sufficiency premise fails.","tokens_in":34890,"feed_emoji":"🩻","tokens_out":8749,"duration_ms":91133,"temperature":0.7,"pith_summary":"The paper claims that a 3D-Gaussian-based representation can reconstruct an object's continuous interior directly from sparse sliced data, with no camera poses. Its proposed model, InnerGS, treats the inner 3D Gaussian distribution as a volumetric density field rather than as a surface renderer. If true, the same volume could then be segmented by natural-language queries, opening a pose-free route from raw medical slices to queryable 3D anatomy. This matters because current scene reconstruction mostly models external surfaces and usually needs calibrated cameras.","feed_headline":"3D Gaussians rebuild interiors from sparse scans, no camera needed","feed_subtitle":"A pose-free route from sliced medical data to smooth volumetric reconstruction and text-guided segmentation.","key_machinery":"The central mechanism is the 'inner 3D Gaussian distribution': a collection of anisotropic 3D Gaussians placed inside the object and interpreted as a continuous volumetric density, rather than as surface radiance kernels. This factorized Gaussian splatting setup is what lets the model fill the interior from sparse slices, and the same density field is then coupled with language features to support text-guided segmentation. The claim is that this representation, not an external pose solver, does the work of linking sparse 2D input to a dense 3D volume.","core_discovery":"On its own terms, the paper's discovery is that internal, volumetric structure—not just outer surfaces—can be modeled by directly fitting a continuous density through the inner 3D Gaussian distribution. From sparse sliced input, the model is said to reconstruct smooth and detailed interiors without estimating camera poses, and by injecting language features it extends the same representation to text-guided segmentation of medical scenes. The central conceptual move is to reinterpret 3D Gaussians as volume elements that carry density and semantic information inside the object, making the representation inherently compatible with arbitrary data modalities.","pith_inferences":["A direct testable extension is reconstruction from unregistered slice stacks, where no external tracker aligns the slices, since the paper's pose-free claim should survive without alignment information.","If the language-feature coupling works, zero-shot segmentation of anatomy classes absent from training should follow, because text queries can generalise beyond fixed label sets.","The continuous-density interpretation also suggests that quantitative measures such as volumes or cross-sectional areas could be read directly off the Gaussian parameters, turning the representation into a computational model rather than only a renderer."],"forward_implications":["Medical imaging pipelines could skip camera or sensor pose estimation entirely when building 3D volumes from slices.","The reconstructed volume and its text-guided segmentation would come from one representation, so querying anatomy could become a direct operation rather than a separate post-processing step.","Because the method claims modality-agnostic compatibility, the same Gaussian-density machinery could apply to CT, MRI, ultrasound, or industrial cross-sectional scans.","If the density interpretation is right, 3D Gaussian splatting becomes a general interior-scene representation, not just a surface renderer, broadening its use in simulation and planning."],"supporting_citations":[],"fun_headline_variants":["Pose-free 3D Gaussians rebuild interiors from sparse slices","Text-guided interior segmentation from sparse medical slices","Internal scenes from sliced data, no camera poses needed","InnerGS: dense interior reconstruction from sparse data","3D Gaussian volume maps sliced inputs to full interiors"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that sparse sliced data, without camera poses, already contains enough geometric constraint to determine a faithful continuous 3D volume; this premise is asserted in the abstract and is not tested in the submitted body.","fun_headline_variants_meta":{"raw":{"variants":["Pose-free 3D Gaussians rebuild interiors from sparse slices","Text-guided interior segmentation from sparse medical slices","Internal scenes from sliced data, no camera poses needed","InnerGS: dense interior reconstruction from sparse data","3D Gaussian volume maps sliced inputs to full interiors"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000278,"raw_usage":{"total_tokens":1589,"prompt_tokens":815,"completion_tokens":774,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":431,"completion_tokens_details":{"reasoning_tokens":698}},"tokens_in":431,"tokens_out":774,"duration_ms":7537,"temperature":1.0,"reasoning_tokens":698,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T17:15:24.043858+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Feed the proposed model a known volume, such as a CT-scanned phantom with internal structures, using only a sparse set of slices and no pose information, then compare the reconstructed density against the ground truth; if the interior accuracy degrades as slice spacing grows, the data-sufficiency premise fails.","supporting_citations":[],"review_version":1}