Skip to content
Computational Pathology Reading Roadmap

Writing study: DK-DDIL: Adaptive Knowledge Retention for Dynamic Domain-Incremental Learning in Medical Imaging

← CVPR 2026 oral writing corpus

Yuxi Ma, Sujie Liu, Jing Yang, Jiacheng Wang, Yiping Chen, Baptiste Magnier, Liansheng Wang · Official CVPR 2026 oral record

Figure, table and section ledger

Editorial contribution type: Training, objectives & optimization (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
≈ 8.1 page-equivalents
Main figures / tables
3 / 1
Attached appendix figures / tables
0 / 0
Attached appendix span
Separate file recorded below; not in this PDF total
Before references / appendix: ≈ 81%References / intervening material: ≈ 19%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 headingStarts on PDF pageEstimated spanShare of entire PDF
1. Introduction1≈ 0.98 pp.9.8%
2. Related Work2≈ 0.69 pp.6.9%
3. Method3≈ 3.67 pp.36.7%
4. Experiments6≈ 2.04 pp.20.4%
5. Conclusion and Future works8≈ 0.26 pp.2.6%
Acknowledgments9≈ 0.05 pp.0.5%
References9≈ 1.9 pp.19%
All detected figure and table locations

Main text

Attached appendix

    No recognized caption labels.

    Separate source files—excluded from canonical-PDF totals

    A separate author version may repeat main text and references. Whole-file lengths and counts are not extra appendix material and are never added to the canonical PDF totals. Read the study's source scope for which sections received substantive reading versus layout inspection.

    Additional source 1

    Separate PDF (or ZIP containing it): whole-file layout of 5 physical pages; 2 detected figures and 2 tables. These may include overlapping main-text material.

    Located appendix only: starts on physical p.1; ≈ 4.93 page-equivalents, 2 figure labels and 2 table labels. This interval is distinct from the whole file.

    Detected headingPDF pageEstimated span
    Supplementary Material1≈ 0.15 pp.
    6. Theoretical Motivation of Adaptive Fusion1≈ 0.82 pp.
    7. Algorithmic Overview of DK-DDIL2≈ 0.08 pp.
    8. Influence of Domain Ordering2≈ 0.42 pp.
    9. Results Analysis of DCL2≈ 0.33 pp.
    10. Extended Experimental Settings2≈ 3.12 pp.

    Argument and organization

    1. 3. Method

      Defines dynamic label/domain setting, adaptive module, and inheritance/refinement.

    2. 4. Experiments

      Reports medical and natural-image domains, trainable parameters, and component sensitivity.

    Problem, claim and contribution

    The introduction distinguishes dynamic domains with evolving labels from fixed-label DIL, then connects this distinction to clinical distribution changes.

    §1; §3.1 · Main paper · Additional source 2

    Argument and section order

    DAM and KIR are separately defined and the latter is decomposed into model fusion and domain-contrastive learning.

    §§3.2–3.3 · Main paper · Additional source 2

    Figures, captions and visual explanation

    The architecture uses colored domain streams and changing label symbols, then zooms into rank selection and inheritance/refinement mechanics.

    Figure 1 p. 4 · Main paper · Additional source 2

    Tables, comparisons and evidence

    The main table places trainable-parameter percentage beside average and final accuracy, while a domain-distribution plot makes imbalance visible before results.

    Figure 2 p. 6; Table 1 p. 7 · Main paper · Additional source 2

    Appendix and reproducibility

    Order permutations and replay-setting clarification address continual-learning comparison conditions; pseudocode assembles the separately explained modules.

    Supplement §§7–10; Algorithm 1 · Main paper · Additional source 2

    Clarity and writing conventions

    Average and final accuracy remain distinct throughout; five-run averaging and paired testing are disclosed near the main comparison.

    §4.1–4.2 p. 7 · Main paper · Additional source 2

    What to reuse—and what not to infer

    Transferable lesson: A continual-learning paper should visualize the sequence and evolving label space, not only the network, and report both retention and final performance.

    Boundary: The named skin setting uses public dermoscopic datasets; do not silently relabel it WSI histopathology. Trainable-parameter savings are not measured runtime savings.

    Apply this lesson using the argument and evidence worksheet.

    Inspection record and sources

    Checked
    2026-09-26
    Reading status
    Writing-focused manuscript scan + selected visuals inspected
    Main text
    Manuscript-wide writing scan of main PDF pp. 1–8: framing, section structure, method exposition, figure/table captions, evaluation organization and conclusions; not an exhaustive technical proof audit or reproduction.
    Appendix / supplement
    Scanned five supplementary pages with fusion motivation, domain-order experiments, loss ablations, dataset/metric details, replay clarification, and end-to-end pseudocode.
    Visually inspected PDF pages
    4, 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.