Skip to content
Computational Pathology Reading Roadmap

Writing study: FedAdamom: Adaptive Momentum for Improved Generalization in Federated Optimization

← CVPR 2026 oral writing corpus

Wenjie Hou, Tianxiang Chen, Feng Wang, Tiantong Wu, Zhiming Zheng, Shaoting Tang, Wei Yang Bryan Lim · Official CVPR 2026 oral record

Figure, table and section ledger

Editorial contribution type: Theory & formal analysis (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
11 physical pages
Before references / appendix
≈ 8.13 page-equivalents
Main figures / tables
3 / 3
Attached appendix figures / tables
0 / 0
Attached appendix span
Separate file recorded below; not in this PDF total
Before references / appendix: ≈ 73.91%References / intervening material: ≈ 26.09%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≈ 1.02 pp.9.27%
2. Related Work2≈ 0.6 pp.5.45%
3. Preliminaries3≈ 0.77 pp.7%
4. Diffusion Dynamics of FL4≈ 1.11 pp.10.09%
5. Adaptive FL Momentum Optimization5≈ 1.1 pp.10%
6. Experimental Analysis6≈ 2.41 pp.21.91%
7. Conclusion8≈ 0.25 pp.2.27%
Acknowledgements9≈ 0.09 pp.0.82%
References9≈ 2.87 pp.26.09%
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 15 physical pages; 1 detected figures and 6 tables. These may include overlapping main-text material.

    Located appendix only: starts on physical p.1; ≈ 14.85 page-equivalents, 1 figure labels and 6 table labels. This interval is distinct from the whole file.

    Detected headingPDF pageEstimated span
    Supplementary Material1≈ 0.06 pp.
    A. Experiment Details1≈ 4.58 pp.
    B. Diffusion Dynamics5≈ 5.95 pp.
    C. Convergence Analysis11≈ 4.27 pp.

    Argument and organization

    1. 3–4. Preliminaries and Diffusion Dynamics

      States assumptions, then compares the dynamics of existing optimizers.

    2. 5–6. Proposed Optimization and Experiments

      Presents the algorithm and analysis before accuracy, rounds, and mechanism tests.

    Problem, claim and contribution

    The opening separates fast optimization from generalization under client heterogeneity rather than treating them as the same benefit.

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

    Argument and section order

    Named assumptions precede the escape-time theorems, followed by the proposed algorithm and a separate convergence analysis.

    §§3–5 pp. 3–6 · Main paper · Additional source 2

    Figures, captions and visual explanation

    Loss-surface comparisons accompany, rather than replace, a table of rounds to target accuracy and accuracy at fixed rounds.

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

    Tables, comparisons and evidence

    A synthetic escape-time test examines a theory-linked prediction, while realistic feature-skew and imbalance experiments test broader behavior.

    Figure 3; Table 3 p. 8 · Main paper · Additional source 2

    Appendix and reproducibility

    Experimental details are grouped before diffusion and convergence proofs, preserving an explicit trail from empirical settings to theoretical support.

    Supplement §§A–C · Main paper · Additional source 2

    Clarity and writing conventions

    Theorems retain quasi-equilibrium and low-temperature assumptions; the writing lesson is to keep these conditions attached when summarizing the result.

    Assumptions 1–3 and Theorems 1–2 p. 4 · Main paper · Additional source 2

    What to reuse—and what not to infer

    Transferable lesson: A theory-motivated optimizer needs both a prediction-specific diagnostic and ordinary task-level comparisons under fixed budgets.

    Boundary: A plotted two-dimensional loss landscape is not a global proof of generalization; the asymptotic analysis depends on stated approximations.

    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 15 supplementary pages for experimental configurations, scale/seed/hyperparameter tests, learning curves, diffusion derivations, and convergence-proof organization; proofs were not independently verified.
    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.