Writing study: Filtered Conformal Ellipsoids for Graph-Native Time Series

NeurIPS overviewCorpus atlasBy contribution typeVisual galleryRules & templateWriting patternsPaper studies

← NeurIPS 2026 oral writing corpus

Yannick Limmer · Official NeurIPS 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: 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

45 physical pages

Before references / appendix

≈ 10.33 page-equivalents

Main figures / tables

2 / 3

Attached appendix figures / tables

1 / 22

Attached appendix span

≈ 32.91 page-equivalents

Before references / appendix: ≈ 22.96%References / intervening material: ≈ 3.91%Attached appendix: ≈ 73.13%Estimated layout envelopes, not text/artwork area. Unresolved appendix boundaries remain within the undivided post-body material.

Source-bound correction: Top-level heading and reference/appendix boundaries checked against source PDF text. Bibliography titles/author initials, numbered contribution/property bullets, diagram steps, equations and TOC entries are not substantive headings. Section spans are page-coordinate estimates, not ink area or prose-only length. This is an author manuscript mirror, not a verified accepted-version template. Source layout uses physical single-column y/page-height, including equation-heavy pages.

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	≈ 1.37 pp.	3.04%
2 Related work	3	≈ 0.39 pp.	0.87%
3 Preliminaries	3	≈ 1.31 pp.	2.91%
4 Filtered conformal prediction	4	≈ 0.83 pp.	1.84%
5 Theory	5	≈ 2.21 pp.	4.91%
6 Experiments	7	≈ 2.36 pp.	5.24%
7 Discussion and limitations	10	≈ 0.9 pp.	2%
Acknowledgments and Disclosure of Funding	11	≈ 0.23 pp.	0.51%
References	11	≈ 1.77 pp.	3.93%
A Observable-contraction theory	13	≈ 3.61 pp.	8.02%
B Full proofs and derivations	16	≈ 7.73 pp.	17.18%
C Filter specifications, training, and pseudocode	24	≈ 4.15 pp.	9.22%
D Baseline implementations	28	≈ 2.86 pp.	6.36%
E Additional experiments	31	≈ 13.86 pp.	30.8%
F Reproducibility	45	≈ 0.7 pp.	1.56%

All detected figure and table locations

Main text

• Figure 1 · p. 2
• Table 1 · p. 6
• Figure 2 · p. 8
• Table 2 · p. 8
• Table 3 · p. 9

Attached appendix

• Table 4 · p. 26
• Table 5 · p. 29
• Table 6 · p. 30
• Table 7 · p. 31
• Table 8 · p. 32
• Table 9 · p. 35
• Table 10 · p. 35
• Figure 3 · p. 36
• Table 11 · p. 36
• Table 12 · p. 37
• Table 13 · p. 37
• Table 14 · p. 38
• Table 15 · p. 39
• Table 16 · p. 40
• Table 17 · p. 40
• Table 18 · p. 41
• Table 19 · p. 41
• Table 20 · p. 42
• Table 21 · p. 42
• Table 22 · p. 43
• Table 23 · p. 44
• Table 24 · p. 44
• Table 25 · p. 45

Argument and organization

1. Problem and contribution

   Separates learning an ellipsoid's shape from conformal calibration of its radius, avoiding reliance on Gaussian tail probabilities.

2. Argument sequence

   Presents construction, conditional theory and three numbered experimental questions before limitations.

3. Evidence and verification

   The headline table reports coverage next to width, defines the target tolerance and notes native versus shared mean predictors.

Problem, claim and contribution

Separates learning an ellipsoid's shape from conformal calibration of its radius, avoiding reliance on Gaussian tail probabilities.

Introduction; Figure 1 · Main paper

Argument and section order

Presents construction, conditional theory and three numbered experimental questions before limitations.

Sections 4–7 · Main paper

Figures, captions and visual explanation

The five-stage schematic distinguishes frozen filtering, scalar scoring and calibration. Coverage–width panels place a target line beside sharpness.

Figure 1, p. 2; Figure 2, p. 8 · Main paper

Tables, comparisons and evidence

The headline table reports coverage next to width, defines the target tolerance and notes native versus shared mean predictors.

Table 2, p. 8 · Main paper

Appendix and reproducibility

Proofs, algorithms, dataset taxonomy, mean-predictor ledger, per-seed results and assumption diagnostics have explicit sections and tables.

Appendices A–E · Main paper

Clarity and writing conventions

The text repeatedly distinguishes finite-trajectory diagnostics from certification of population assumptions, and retains full-graph non-winning cases.

Table 1; Section 7; Appendix E.15 · Main paper

What to reuse—and what not to infer

Transferable lesson: Place the comparison contract beside the numbers and distinguish empirical checks from proof assumptions.

Boundary: This is not universal full-graph leadership or conditional coverage. The source reports backbone- and scale-dependent limits.

Apply this lesson using the argument and evidence worksheet.

Inspection record and sources

Checked

2026-10-02

Reading status

Writing-focused manuscript scan + selected visuals inspected

Main text

Writing-oriented distributed manuscript scan of Sections 1–7, pp. 1–10; references pp. 11–12: abstract/framing, section sequence, substantive text windows, captions, evidence organization and conclusion; not exhaustive technical verification.

Appendix / supplement

Writing-oriented scan of Appendices A–F, pp. 13–45; proof structure and experiment-ledger scan, not theorem re-derivation; code and experiments were not executed.

Visually inspected PDF pages

2, 8

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.

• Public author arXiv manuscript 2606.17014v2; not verified identical to the accepted or camera-ready OpenReview revision · 45 pages