Writing study: GMMamba: Group Masking Mamba for Whole Slide Image Classification

ICCV overviewCorpus atlasBy contribution typeVisual galleryRules & templateWriting patternsPaper studies

← ICCV 2025 oral writing corpus

Tingting Zheng, Hongxun Yao, Kui Jiang, Yi Xiao, Sicheng Zhao · Official ICCV 2025 oral record

Observed writing choices, not acceptance rules. This study examines the inspected manuscript version. It does not establish why the paper was accepted or that its presentation choices should be copied unchanged.

Figure, table and section ledger

Editorial contribution type: Architecture & representation (primary).

Layout extraction, not a required paper template. Counts are distinct caption labels; a multi-panel figure counts once. Section spans are estimated from heading positions, including floats and whitespace. Method and limitations.

Canonical PDF

10 physical pages

Before references / appendix

≈ 8.05 page-equivalents

Main figures / tables

5 / 8

Attached appendix figures / tables

0 / 0

Attached appendix span

No boundary detected; not assumed zero

Before references / appendix: ≈ 80.5%References / intervening material: ≈ 19.5%Estimated layout envelopes, not text/artwork area. Unresolved appendix boundaries remain within the undivided post-body material.

Extraction notes: No attached appendix boundary detected. Zero means no recognized caption, not independently verified absence.

Section-by-section lengths and source pages

The title/abstract/front matter precede the first numbered heading. A span ends at the next detected top-level heading, not at the last sentence; end-page whitespace is included. References and statements are listed separately. These are not measurements of text-only or artwork area.

Top-level heading	Starts on PDF page	Estimated span	Share of entire PDF
1. Introduction	1	≈ 1.42 pp.	14.2%
2. Related Work	2	≈ 1.08 pp.	10.8%
3. Method	3	≈ 1.41 pp.	14.1%
4. Experiments	5	≈ 3.42 pp.	34.2%
5. Conclusion	8	≈ 0.16 pp.	1.6%
Acknowledgements	8	≈ 0.15 pp.	1.5%
References	9	≈ 1.95 pp.	19.5%

All detected figure and table locations

Main text

• Figure 1 · p. 1
• Figure 2 · p. 2
• Figure 3 · p. 3
• Figure 4 · p. 4
• Table 1 · p. 6
• Table 2 · p. 7
• Table 3 · p. 7
• Table 4 · p. 7
• Table 5 · p. 7
• Figure 5 · p. 8
• Table 6 · p. 8
• Table 7 · p. 8
• Table 8 · p. 8

Attached appendix

No recognized caption labels.

Argument and organization

1. 1 Introduction; 2 Related Work

   Separates redundant local processing from weak global aggregation in long WSI instance sequences.

2. 3 Method

   Defines MIL and state-space notation, then location grouping, masking and cross-group super-feature sampling.

3. 4 Experiments; 5 Conclusion

   Reports four classification cohorts, encoder-specific results, module transfer and group/mask sensitivity before a fixed-mask-ratio limitation.

Problem, claim and contribution

Links two named WSI aggregation problems to two named modules, avoiding a generic claim that a new backbone alone solves gigapixel classification.

Sections 1 and 3.2, physical pp.1–2,4 · Main paper

Argument and section order

The method moves from coordinate-based grouping through masked local sequences to global aggregation, while the dataset section separately identifies slides, cases, patch magnification and features.

Sections 3.2–4.2, pp.4–5 · Main paper

Figures, captions and visual explanation

The overview keeps group colors consistent from slide coordinates through two BiMamba passes and masked instances to the final class token.

Figure 3, p.3 · Main paper

Tables, comparisons and evidence

Results separate BRACS feature extractors; ablation tables state group number and mask ratio, retain five-fold dispersion, and distinguish component additions from cross-backbone reuse.

Tables 2–5, p.7 · Main paper

Appendix and reproducibility

No substantive supplement routing was located in the main argument; core implementation and cohort settings appear in Sections 4.1–4.3. Any separately linked supplement remains uninspected.

Sections 4.1–4.3, pp.5–6; main conclusion, p.8 · Main paper

Clarity and writing conventions

The conclusion gives the uniform masking ratio a concrete limitation and proposes adaptive ratios rather than claiming all irrelevant instances are safely removed.

Section 5, p.8 · Main paper

What to reuse—and what not to infer

Transferable lesson: Keep patient/case counts, slide counts, patch scale and feature encoder separate, and state each ablation’s grouping/masking settings in its caption.

Boundary: The text combines cohort-specific fixed split ratios and an official BRACS split with a later five-fold protocol; a writing scan cannot reconcile the exact split implementation or certify leakage absence. Bag classification and attention-selected instances are not clinical validation or proof of tumor localization.

Apply this lesson using the argument and evidence worksheet.

Inspection record and sources

Checked

2026-10-07

Reading status

Writing-focused manuscript scan + selected visuals inspected

Main text

Writing-oriented scan across the main manuscript, physical PDF pp.1–8, covering framing, section progression, method explanations, captions, comparison protocols and concluding scope; reference boundary checked. Selected full-page rasters were inspected. Not exhaustive line-by-line reading, proof verification or experimental reproduction.

Appendix / supplement

Main-paper appendix/supplement references and the reference boundary were checked. No separate supplementary link appears in this dated CVF inventory; this does not establish that no supplementary material exists elsewhere.

Visually inspected PDF pages

3, 7

This is a writing-focused scan of the manuscript and the stated supplementary scope: framing, organization, visual explanation, evidence presentation and wording. It is not a line-by-line proof audit, reproduction of results or exhaustive scientific review.

• Official CVF open-access ICCV 2025 accepted-paper PDF, retrieved 7 October 2026; CVF states identical to accepted version except watermark; not assumed identical to initial submission · 10 pages