Writing study: Evidential Neural Radiance Fields

CVPR overviewCorpus atlasBy contribution typeVisual galleryRules & templateWriting patternsPaper studies

← CVPR 2026 oral writing corpus

Ruxiao Duan, Alex Wong · Official CVPR 2026 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: 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

≈ 8.08 page-equivalents

Main figures / tables

8 / 2

Attached appendix figures / tables

0 / 0

Attached appendix span

Separate file recorded below; not in this PDF total

Before references / appendix: ≈ 80.8%References / intervening material: ≈ 19.2%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.

Source-bound correction: Table 1 is raster-visible on physical PDF page 5, below the full-width results table; its caption is absent from extracted text. Added after visual inspection on 2026-09-26; position is estimated from the rendered page, not recovered text coordinates.

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	≈ 1.05 pp.	10.5%
2. Related Work	2	≈ 1 pp.	10%
3. Preliminaries	3	≈ 0.68 pp.	6.8%
4. Method	4	≈ 2.3 pp.	23%
5. Experiments	6	≈ 2.41 pp.	24.1%
6. Conclusion	8	≈ 0.29 pp.	2.9%
7. Acknowledgment	9	≈ 0.04 pp.	0.4%
References	9	≈ 1.92 pp.	19.2%

All detected figure and table locations

Main text

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

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 12 physical pages; 6 detected figures and 5 tables. These may include overlapping main-text material.

Located appendix only: starts on physical p.1; ≈ 11.88 page-equivalents, 6 figure labels and 5 table labels. This interval is distinct from the whole file.

Detected heading	PDF page	Estimated span
Supplementary Material	1	≈ 0.05 pp.
8. Derivations: Point-to-Pixel Propagation of Radiance and Uncertainty	1	≈ 0.92 pp.
9. Derivations: Pixel Radiance Marginal Distribution and Loss Function	2	≈ 2 pp.
10. Further Discussions	4	≈ 8.91 pp.

• Figure 9 · p. 4
• Table 3 · p. 4
• Figure 10 · p. 4
• Table 4 · p. 5
• Table 5 · p. 6
• Table 6 · p. 7
• Table 7 · p. 8
• Figure 11 · p. 9
• Figure 12 · p. 10
• Figure 13 · p. 11
• Figure 14 · p. 12

Read-only audit 2026-09-26: restored actual supplementary section headings from cached PDF layout, excluding table-of-contents entries and retaining physical PDF page coordinates. This measures reading-order spans, not occupied text area.

Argument and organization

1. Probabilistic ladder

   Introduce deterministic and normal radiance models before the evidential extension.

2. Point-to-pixel method and applications

   Derive uncertainty propagation, then evaluate reconstruction, calibration, cost and application-specific uses.

Problem, claim and contribution

The gap is not generic confidence estimation but separating aleatoric and epistemic uncertainty in one rendering pass.

Introduction and Figure 1 · Main paper · Additional source 2

Argument and section order

Point-level modeling, propagation and pixel-level learning are separate subsections, so mathematical levels do not silently switch.

Sections 4.1–4.4, PDF pp. 4–5 · Main paper · Additional source 2

Figures, captions and visual explanation

Paired case panels keep training/test views beside AU/EU maps: reflective observations motivate data uncertainty, while unseen viewpoints illustrate knowledge uncertainty.

Figures 5–6, PDF p. 7 · Main paper · Additional source 2

Tables, comparisons and evidence

Reconstruction/uncertainty metrics, timing and training-sample scaling are reported separately, avoiding a single quality number being used to substantiate all claims.

Tables 1–2 and Figure 4, PDF pp. 5–6 · Main paper · Additional source 2

Appendix and reproducibility

Derivations precede sensitivity and per-scene mean/standard-deviation tables; the discussion explicitly cautions that one scene factor can affect both uncertainty types.

Supplement Sections 8–10, pp. 1–8 · Main paper · Additional source 2

Clarity and writing conventions

Bold result subheadings separate reconstruction, uncertainty quantification and qualitative interpretation; they also acknowledge metrics where alternatives perform better.

Section 5.2, PDF p. 7 · Main paper · Additional source 2

What to reuse—and what not to infer

Transferable lesson: Teach a probabilistic extension through a hierarchy of simpler formulations, and place diagnostic uncertainty maps next to the conditions that generate them.

Boundary: AU/EU maps support interpretation but do not uniquely identify causes. The supplement explicitly discusses overlapping causes; do not turn illustrative cases into a universal causal rule.

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

Separate supplement pp. 1–12 scanned: Sections 8–10 derivations/discussion, per-scene repeated-run tables and qualitative uncertainty maps.

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.

• Official CVPR 2026 CVF proceedings PDF; retrieved 2026-09-26 · 10 pages
• Official CVPR 2026 separate supplementary PDF; retrieved 2026-09-26 · 12 pages