# Scout (narrow): electron dim-4 SME c bounds, Collins 2024–26, 30-day anomaly sweep

**Thread:** `2026-09-13-scout-narrow`  
**Date:** 2026-09-13  
**Author:** Argus subagent (narrow scout)  
**Status:** COMPLETE. Q1 from Dreissen PDF Table 1 + Kostelecký & Russell Data Tables v19 (Tables S2, S3, D6, D20). Q2 = INSPIRE citation audit of recid:646314, 2024–2026. Q3 = 30-day window 2026-08-14 → 2026-09-13.

Evidence classes: Established / Serious speculation / Anomaly / Anecdote / Your own inference.

Master compilation used throughout Q1: V. Alan Kostelecký and Neil Russell, "Data Tables for Lorentz and CPT Violation," arXiv:0801.0287v19 (January 2026 edition of Rev. Mod. Phys. 83, 11 (2011)), 198 pp. Literature cutoff stated in the tables: 31 December 2025. PDF: https://arxiv.org/pdf/0801.0287

Convention note (Established, Data Tables §II and Table P60): the *tilde* matter-sector coefficients in Table S2 have dimensions of **GeV** (`c̃ ∼ m_e c`, with `m_e ≃ 5.11×10^{-4} GeV`). The ion-spectroscopy `c_{MN}` components (Dreissen, Sanner, …) are the **dimensionless** Sun-centered traceless-symmetric tensor, and in those papers they are the *combined* electron–photon coefficients `c'_{μν} = c_{μν} + k_{μν}/2`. Do not mix the two without converting.

---

## 1. Electron dimension-4 SME c-coefficients

### 1.1 The Dreissen `c_XY` number, stated exactly [Established]

**Paper:** Laura S. Dreissen, Chih-Han Yeh, Henning A. Fürst, Kai C. Grensemann, Tanja E. Mehlstäubler, "Improved bounds on Lorentz violation from composite-pulse Ramsey spectroscopy in a trapped ion," *Nature Communications* **13**, 7314 (2022). arXiv:2206.00570 (v4, 24 Jul 2024). DOI: 10.1038/s41467-022-34818-0.

**What was measured:** combined electron–photon tensor `c'_{MN}` (they drop the prime) in the Sun-centered celestial equatorial frame (SCCEF), via rf composite-pulse Ramsey spectroscopy of the `²F_{7/2}` manifold of a single trapped `¹⁷²Yb⁺` ion. 1σ uncertainties.

**Table 1 (verbatim numerical content, correlated components):**

| coefficient | this work (Dreissen 2022) | ¹⁷¹Yb⁺ Sanner et al. | ⁴⁰Ca⁺ Pruttivarasin / Megidish |
|---|---|---|---|
| `c_{X−Y} ≡ c_{XX}−c_{YY}` | `(−5.2 ± 7.8)×10^{-21}` | `(−0.5 ± 1.7)×10^{-20}` | `(6.2 ± 9.2)×10^{-19}` |
| `c_{XY}` | `(4.4 ± 3.9)×10^{-21}` | `(−7.0 ± 8.1)×10^{-21}` | `(2.4 ± 4.8)×10^{-19}` |
| `c_{XZ}` | `(−5.0 ± 9.3)×10^{-21}` | `(0.8 ± 1.3)×10^{-20}` | `(0.8 ± 2.1)×10^{-19}` |
| `c_{YZ}` | `(6.3 ± 8.9)×10^{-21}` | `(1.0 ± 1.3)×10^{-20}` | `(−3.1 ± 2.2)×10^{-19}` |

**The "3.9×10^{-21}" phrase:** it is **the 1σ uncertainty on `c_{XY}`**, not a one-sided bound of `3.9×10^{-21}` and not the central value. Quote (Discussion, p. 6 of the arXiv PDF): "Therefore, the tightest constraint of 3.9×10^{-21} is achieved on `c_{XY}`." Central value is `(4.4 ± 3.9)×10^{-21}` (1.1σ from zero; Lomb–Scargle shows no significant `2ω_⊕` peak). They also write that they "constrain all the coefficients of the `c_{MN}` tensor now at the 10^{-21} level" and improve the previous best (Sanner) by a factor 2.2.

Data Tables entry: Table D6 part 2, Ref. [64] = this paper. Same four numbers, system "Trapped Yb ion."

Sanner comparison paper: C. Sanner et al., *Nature* **567**, 204 (2019), arXiv:1809.10742 (Data Tables [73]). Ca⁺: T. Pruttivarasin et al., *Nature* **517**, 592 (2015), arXiv:1412.2194 ([76]); E. Megidish et al., PRL **122**, 123605 (2019), arXiv:1809.09807 ([75]).

Dreissen does **not** report `c_{TT}`, `c_Q`, `c̃_X`, `c̃_Z` as independent fitted parameters. Those live in other rows / the tilde basis (below). Boost-suppressed `c_{TJ}` components are not constrained at the 10^{-21} level by this run (high-pass at 5 μHz; they fit only `ω_⊕` and `2ω_⊕`).

### 1.2 Data Tables Table S2 — electron tilde c-sector, maximal two-sided sensitivities [Established]

January 2026 edition, Table S2. **Order-of-magnitude, one-coefficient-at-a-time.** Units GeV. Electron column:

| tilde coefficient | electron maximal sensitivity (S2) |
|---|---|
| `c̃_{-}` | `10^{-23} GeV` |
| `c̃_{Q}` | `10^{-20} GeV` |
| `c̃_{X}` | `10^{-23} GeV` |
| `c̃_{Y}` | `10^{-23} GeV` |
| `c̃_{Z}` | `10^{-23} GeV` |
| `c̃_{TX}` | `10^{-20} GeV` |
| `c̃_{TY}` | `10^{-21} GeV` |
| `c̃_{TZ}` | `10^{-21} GeV` |
| `c̃_{TT}` | `10^{-21} GeV` |

These are **not** the numbers to compare a theory prediction against if you have a specific component; use the D6 rows in §1.3. S2 is the depth/breadth map.

### 1.3 Best *specific* electron dim-4 c numbers (theory-comparison set) [Established]

All from Table D6 (electron sector, `d=3,4`) of the January 2026 Data Tables, cross-checked to the source papers named in the tables' bibliography.

**A. Spatial anisotropic (dimensionless `c_{MN}`, combined e–γ, SCCEF) — laboratory, best:**

- `c_{XY} = (4.4 ± 3.9)×10^{-21}` — Dreissen et al., Nat. Comm. 13, 7314 (2022), arXiv:2206.00570. D6 [64].
- `c_{X−Y} = (−5.2 ± 7.8)×10^{-21}` — same.
- `c_{XZ} = (−5.0 ± 9.3)×10^{-21}` — same.
- `c_{YZ} = (6.3 ± 8.9)×10^{-21}` — same.

If a prediction is for the *electron-only* `c_{μν}` rather than `c' = c + k/2`, these still apply as bounds on the combination; a pure-photon interpretation would map them onto `κ̃_{e−}` at the same order (factor 2).

**B. `c_{TT}` (dimensionless, electron) — astrophysical one-sided is the deepest, and it is one-sided:**

- `c_{TT} > −5×10^{-21}` — F. W. Stecker, "Constraining Superluminal Electron and Neutrino Velocities using the 2010 Crab Nebula Flare and the IceCube PeV Neutrino Events," *Astropart. Phys.* **56**, 16 (2014), arXiv:1306.6095. Data Tables D6 [72]\* (asterisk = derived). Stecker's own wording (abstract): vacuum Čerenkov non-occurrence for electrons up to ∼5.1 PeV in the Sept 2010 Crab flare implies **superluminal** electron velocity `δ_e ≲ 5×10^{-21}`; **subluminal** is only `|δ_e| ≲ 8×10^{-17}` from the Crab γ-ray spectrum. D6 maps the superluminal one-sided limit onto `c_{TT} > −5×10^{-21}`. Do not treat this as a two-sided 10^{-21} bound.
- Two-sided astrophysical (weaker): `c_{TT} ∈ (−80 to +4)×10^{-20}` — B. Altschul, *Symmetry* **13**, 688 (2021), arXiv:2104.04587, D6 [66]\*. Same paper: `c_{TX} (−3 to 50)×10^{-19}`, `c_{TY} (−3 to 6)×10^{-19}`, `c_{TZ} (−8 to 140)×10^{-20}`, `c_{XX}+c_{YY}−2c_{ZZ} (−3 to 40)×10^{-19}`.
- Laboratory `c_{TT}` is far weaker: Hohensee et al. Dy spectroscopy, PRL 111, 050401 (2013), arXiv:1303.2747, D6 [79]: `c_{TT} = (−8.8 ± 5.1)×10^{-9}` and `(−14 ± 28)×10^{-9}`. Qin et al. gravitational redshift, PRD 111, 055008 (2025), arXiv:2503.13564, D6 [62]: a ladder of `c_{TT}` results, best laboratory-ish cell in that block `(−4.3 ± 3.5)×10^{-8}` / `(7.4 ± 9.3)×10^{-9}` — still ~12 orders behind Stecker.

**C. Tilde `c̃_Q`, `c̃_{-}`, `c̃_X`, `c̃_Z` (GeV) — conversion from Dreissen plus S2:**

Standard tilde map (Data Tables properties; `m ≡ m_e`):
`c̃_{-} = m(c_{XX}−c_{YY})`, `c̃_Q = m(c_{XX}+c_{YY}−2c_{ZZ})`, `c̃_X = m c_{YZ}`, `c̃_Y = m c_{ZX}`, `c̃_Z = m c_{XY}`, `c̃_{TT} = m c_{TT}`.

Using `m_e = 5.109989×10^{-4} GeV` on Dreissen Table 1 (Your own inference for the arithmetic; the input numbers are Established):

- `c̃_Z = m_e c_{XY} → (2.2 ± 2.0)×10^{-24} GeV`
- `c̃_{-} = m_e c_{X−Y} → (−2.7 ± 4.0)×10^{-24} GeV`
- `c̃_X = m_e c_{YZ} → (3.2 ± 4.5)×10^{-24} GeV`
- `c̃_Y` would use `c_{ZX}=c_{XZ}`: `(−2.6 ± 4.8)×10^{-24} GeV`

S2 rounds these to `10^{-23} GeV` for `c̃_{-}, c̃_X, c̃_Y, c̃_Z`. That is the correct order-of-magnitude for a one-coefficient summary.

`c̃_Q` is **not** set by Dreissen (no handle on `c_{ZZ}` vs. the trace-free quadrupole at this geometry without boost). S2 quotes `10^{-20} GeV` for electron `c̃_Q`. Direct D6 cell that is actually `c̃_Q`: `|c̃_Q| < 2×10^{-8} GeV`, relativistic Li ions, Botermann et al., PRL 113, 120405 (2014), arXiv:1409.7951, D6 [77] — much weaker; the S2 `10^{-20} GeV` is therefore coming from converting a dimensionless combination (Altschul 2021 `c_{XX}+c_{YY}−2c_{ZZ} ∼ 10^{-19}` × `m_e ∼ 5×10^{-23}` is still not `10^{-20}`). **Do not quote `|c̃_Q|<2×10^{-8} GeV` as the best bound; it is a direct but weak measurement. Best dimensionless quadrupole-like combination from astrophysics is Altschul 2021 `c_{XX}+c_{YY}−2c_{ZZ} ∈ (−3 to 40)×10^{-19}` [66]\*.** S2 `10^{-20} GeV` is the compilers' one-at-a-time estimate, not a single-experiment cell.

### 1.4 Photon-sector dimension-4: `κ̃_{e−}` and `κ̃_{o+}` [Established]

From Table S3 (January 2026), maximal two-sided sensitivities, dimensionless:

**Nonbirefringent, parity-even `κ̃_{e−}` (the cavity / "speed-of-light anisotropy" sector):**

| coefficient | S3 sensitivity |
|---|---|
| `(κ̃_{e−})^{XY}` | `10^{-22}` |
| `(κ̃_{e−})^{XZ}` | `10^{-20}` |
| `(κ̃_{e−})^{YZ}` | `10^{-20}` |
| `(κ̃_{e−})^{XX} − (κ̃_{e−})^{YY}` | `10^{-21}` |
| `(κ̃_{e−})^{ZZ}` | `10^{-16}` |

**Nonbirefringent, parity-odd `κ̃_{o+}` (boost-type, much weaker in the lab):**

| coefficient | S3 sensitivity |
|---|---|
| `(κ̃_{o+})^{XY}` | `10^{-14}` |
| `(κ̃_{o+})^{XZ}` | `10^{-14}` |
| `(κ̃_{o+})^{YZ}` | `10^{-14}` |

Also in S3, isotropic: `κ̃_{tr} : 10^{-20}` and `c^{(4)}_{(I)00} = √(4π) κ̃_{tr} : 10^{-19}`. Birefringent `k^{(4)}_{(E)jm}` sit at `10^{-35}` (CMB/spectropolarimetry) — those are **not** `κ̃_{e−}` / `κ̃_{o+}`; they are `κ̃_{e+}` / `κ̃_{o−}`.

Source papers for the photon dim-4 nonbirefringent sector (Data Tables bibliography, the cavity/resonator cluster): S. Herrmann et al., PRL 95, 150401 (2005) [174]; P.L. Stanwix et al., PRL 95, 040404 (2005) [175]; P. Wolf et al., PRD 70, 051902 (2004) [176]; H. Müller et al., PRL 91, 020401 (2003) [177]; J.A. Lipa et al., PRL 90, 060403 (2003) [178]; M. Nagel et al. is the modern 10^{-18} cavity generation (see below). LHC isotropic: D. Amram et al., PRL 132, 211801 (2024), arXiv:2312.11307 [187]: `κ̃_{tr} > −1.06×10^{-13}` (95% CL) from 13 TeV inclusive prompt photons — **not** competitive with S3's `10^{-20}` (the S3 isotropic number is astrophysical / UHECR / vacuum Cherenkov, e.g. Klinkhamer–Risse and Duenkel–Niechciol–Risse [180], [188], [189]).

**Best exact D20 cells for `κ̃_{e−}` / `κ̃_{o+}` (Established):**

The `10^{-22}` figure is **not** a Dreissen translation. It is laser-interferometer analysis of LIGO S5 (2006–2007) data:

V. A. Kostelecký, A. C. Melissinos, and M. Mewes, "Searching for photon-sector Lorentz violation using gravitational-wave detectors," *Phys. Lett. B* **761**, 1 (2016), arXiv:1608.02592. Data Tables D20 [166]. Exact cells:

| coefficient | D20 result [166] |
|---|---|
| `|(κ̃_{e−})^{XY}|` | `< 2.7 × 10^{-22}` |
| `|(κ̃_{e−})^{XX} − (κ̃_{e−})^{YY}|` | `< 5.5 × 10^{-22}` |
| `|(κ̃_{e−})^{XZ}|` | `< 2.1 × 10^{-20}` |
| `|(κ̃_{e−})^{YZ}|` | `< 2.1 × 10^{-20}` |
| `|(κ̃_{o+})^{XY}|` | `< 6.6 × 10^{-15}` |
| `|(κ̃_{o+})^{XZ}|` | `< 5.7 × 10^{-15}` |
| `|(κ̃_{o+})^{YZ}|` | `< 5.2 × 10^{-15}` |

Authors' own claim: "constraints on coefficients for Lorentz violation in the photon sector exceeding current limits by about four orders of magnitude," from preliminary LIGO 2006–2007 data. These are **one-sided inequalities**, not measured central values.

**Best dedicated optical-cavity (rotating Michelson–Morley) numbers**, still the cleanest *laboratory oscillator* result:

M. Nagel, S. R. Parker, E. V. Kovalchuk, P. L. Stanwix, J. G. Hartnett, E. N. Ivanov, A. Peters, M. E. Tobar, "Direct terrestrial test of Lorentz symmetry in electrodynamics to 10^{-18}," *Nature Communications* **6**, 8174 (2015), arXiv:1412.6954. D20 [168], system "Sapphire cavity oscillators." D20 cells:

- `(κ̃_{e−})^{XY} = (−0.7 ± 1.6)×10^{-18}`
- `(κ̃_{e−})^{XZ} = (−5.5 ± 4.0)×10^{-18}`
- `(κ̃_{e−})^{YZ} = (−1.9 ± 3.2)×10^{-18}`
- `(κ̃_{e−})^{XX}−(κ̃_{e−})^{YY} = (−1.5 ± 3.4)×10^{-18}`
- `(κ̃_{o+})^{XY} = (−3.0 ± 3.4)×10^{-14}`
- `(κ̃_{o+})^{XZ} = (0.21 ± 1.7)×10^{-14}`
- `(κ̃_{o+})^{YZ} = (−2.0 ± 1.6)×10^{-14}`

Paper abstract: `Δν/ν = 9.2 ± 10.7 × 10^{-19}` (95% CI). This is the number a cavity theorist compares against; [166] is the number the Data Tables treat as deepest for `(κ̃_{e−})^{XY}`.

Next rotating-cavity generation at `10^{-17}`: T. Zhang et al., *Phys. Lett. A* **416**, 127666 (2021), D20 [165], `(κ̃_{e−})^{XY} = (−2.3 ± 5.4)×10^{-17}`.

Isotropic `\kappã_{tr}`: D20 best astrophysical one-sided cells are Duenkel, Niechciol, Risse, PRD **107**, 083004 (2023), arXiv:2303.05849 [188]\* `\kappã_{tr} < 3×10^{-20}` and PRD **104**, 015010 (2021), arXiv:2106.01012 [189]\* `−(\kappã_{tr} − 4/3 c^e_{00}) < 6×10^{-21}`. Collider (not competitive): Amram et al. PRL **132**, 211801 (2024), arXiv:2312.11307 [187] `\kappã_{tr} > −1.06×10^{-13}` (95% CL).

---

## 2. Collins–Pérez–Sudarsky–Urrutia naturalness, 2024–2026

**Original (not in-window, for orientation):** J. Collins, A. Perez, D. Sudarsky, L. Urrutia, H. Vucetich, "Lorentz invariance and quantum gravity: an additional fine-tuning problem?", *Phys. Rev. Lett.* **93**, 191301 (2004), arXiv:gr-qc/0403053. INSPIRE recid:646314, 342 citations as of this fetch.

**Headline [Established, null result]:** there is **no 2024–2026 paper that recalculates, rebuts, or replaces the Collins loop argument, and no new SUSY-protection paper** in the Groot Nibbelink / Pospelov / Bolokhov line. INSPIRE query `refersto:recid:646314 AND date>2023` returned **36 citing works**. I inspected titles/abstracts. Almost all are drive-by citations (bumblebee metrics, VSR phenomenology, holographic models, theses). Closest items that actually *touch* the physics, not the argument:

1. **G. R. Bengochea, G. Leon, A. Perez**, "Is Planckian discreteness observable in cosmology?", arXiv:2405.12534 (21 May 2024; v2 28 Apr 2025), *Universe* **11**, 139 (2025). DOI 10.3390/universe11050139. Perez is an original Collins coauthor. Subject is Planck-scale inflation / discreteness → CMB, **not** a rerun of the Yukawa-loop percolation. **Class:** Serious speculation (QG cosmology). **Not a rebuttal.**

2. **G. R. Bengochea, G. Leon, A. Perez**, "Emergence of cosmic structure from Planckian discreteness", arXiv:2506.15413 (18 Jun 2025; v2 19 Jan 2026). Extends (1) to quasi-de Sitter; scale-invariant scalar spectrum from Planckian discreteness without a trans-Planckian QFT assumption. Again **not** a loop-naturalness calculation. **Class:** Serious speculation.

3. **D. S. Ageev, Yu. Ageeva**, "One-Loop Renormalization of Anisotropic Two-Scalar Quantum Field Theories", arXiv:2512.19670 (22 Dec 2025; v2 31 Jan 2026), INR-TH-2025-024. One-loop dim-reg / MS-bar for two scalars with the most general two-derivative Lorentz-violating quadratic form. Computes UV poles and beta functions; anisotropy restricts the Wilson–Fisher fixed point. **This is the only 2024–26 paper in the citation list that actually does a one-loop LV renormalization.** It does **not** claim to refute or confirm Collins' O(g²) dim-4 percolation (different model, dim-reg, two scalars). **Class:** Established (calculation) / not a Collins verdict.

4. **G. R. Perez Teruel**, "Geometric Constraints on Quantum Gravity-Inspired Dispersion Relations", arXiv:2512.00933 (30 Nov 2025; v2 10 Jul 2026), IJGMMP. Geometric stability of LQG / causal-set / κ-Poincaré MDRs; cites Collins. **Not a naturalness rebuttal.** **Class:** Serious speculation.

5. **A. Maiezza, J. C. Vasquez**, "Quantum field theory on multifractal spacetime: Varying dimension and ultraviolet completeness", arXiv:2504.06797 (9 Apr 2025). Cites Collins; UV-completeness via varying dimension, not SUSY. **Class:** Serious speculation.

**SUSY-protection counterargument, 2024–2026:** **no new paper found.** The live references remain Groot Nibbelink & Pospelov, PRL **94**, 081601 (2005), arXiv:hep-ph/0404271, and Bolokhov, Groot Nibbelink, Pospelov, PRD **72**, 015013 (2005), arXiv:hep-ph/0505029. Polchinski's 2012 comment (arXiv:1106.6346) is still the last sharp framing I have ("SUSY is the one known example"). Nothing in the 36-paper citation list is a 2024–26 update of that protection, nor a claim that SUSY protection has failed.

**Where I did not look:** full-text of every drive-by citation; papers that discuss LV naturalness **without** citing gr-qc/0403053 (possible, but the Collins paper is the canonical citation; missing it and still "revisiting the argument" is unlikely). No X/Twitter rumours filed.

---

## 3. 30-day anomaly sweep (≈ 2026-08-14 to 2026-09-13)

Window stated. Items outside it are flagged. No Telescope Array paper and no CSL/objective-collapse *experimental* paper landed inside the window (CSL theory preprint arXiv:2604.21705 is April). Five-to-ten receipts:

1. **Pierre Auger Collaboration**, "Bounds on Lorentz invariance violation from muon fluctuations at the Pierre Auger Observatory," arXiv:2602.14720 (v1 16 Feb 2026; v3 4 Jun 2026, accepted PRL). https://arxiv.org/abs/2602.14720 . In-window as the PRL appearance (Argus 2026-09-12 sweep recorded *Phys. Rev. Lett.* **137**, 111001, published 8 Sep 2026, https://journals.aps.org/prl/abstract/10.1103/t5k4-2m32 — I did not re-open the PRL PDF this run). Abstract: first use of shower-to-shower **muon-number fluctuations**; hadronic-sector LIV bounds, composition-independent, claimed strongest to date. **Class:** Established (null/bound). I am **not** quoting a numerical `η` from the 09-12 file because I did not extract it from this paper's PDF.

2. **N. Martynenko et al.**, "Constraining Lorentz invariance violation from the depth of air-shower maximum," arXiv:2608.05106 (5 Aug 2026), INR-TH-2026-007. https://arxiv.org/abs/2608.05106 . Toy analysis of subluminal photon-sector LIV → suppressed Bethe–Heitler → Xmax shift, using published Auger FD Xmax. Authors' number: `M_{LIV} > 1.5×10^{21} GeV` (95% CL), **and they say not to read it as a detector-level experimental limit** (simplified response, composition-dependent). Edge of window (−9 d). **Class:** Serious speculation (method) on established shower data.

3. **R. E. Stewart et al.**, "Vacuum birefringence and the polarized X-ray emission from a radio magnetar," *Nature* **656**, 590–594 (2026), arXiv:2509.19446 (v5 17 Aug 2026). DOI 10.1038/s41586-026-10859-z. Magnetar 1E 1547.0−5408; IXPE+NICER+Parkes; PD ~65% at 2 keV, ~80% in some phases. Authors: QED vacuum birefringence in the magnetosphere. **This is Standard-Model QED, not SME photon birefringence.** **Class:** Anomaly → approaching Established (polarimetry real; QED interpretation best available, plasma alternatives not fully killed). Do not file as spacetime discreteness.

4. **M. Schreck et al.**, arXiv:2608.01118 (2 Aug 2026), GW dim-6 Lorentz-violating gravity. Nonbirefringent coeffs from GW170817/GRB 170817A timing; birefringent from absence of mode splitting in GW150914. Edge of window (−12 d). **Class:** Established (null/bound on old events). I did not extract the mm / 10 μm numbers from the PDF this run, so I am not quoting them.

5. **No new Telescope Array result** in 2026-08-14 → 2026-09-13 that I could find on arXiv (search: Telescope Array + Aug/Sep 2026 + Lorentz/hotspot). **Class:** Established (null of coverage). The TA hotspot remains the old story.

6. **No new LHAASO LIV bound** inside the 30 days. Last collaboration LIV paper on the books is still LHAASO on GRB 221009A, arXiv:2402.06009 (Feb 2024), outside window. Cygnus X-3 super-PeVatron (arXiv:2512.16638 / NSR 2026) is astrophysical, not LIV; press 2 Aug 2026 sits just outside the 30-day cut.

7. **No objective-collapse / CSL experimental paper** in the 30-day window. Closest theory item is outside: "Testing Spontaneous Collapse Models with Coulomb Mediated Squeezing," arXiv:2604.21705 (April 2026). **Class:** Established (null of coverage).

8. **INSPIRE 36-paper Collins citation list** contains one in-window item that is not on-topic: A. Colléaux, "Rational regular black holes in non-polynomial gravity," arXiv:2608.17158 (17 Aug 2026). Cites Collins; not a LIV measurement. **Class:** Serious speculation (theory), not an anomaly.

**Sweep verdict:** no new claimed physics anomaly that looks like a simulation signature in this 30 days. Live items are a hadronic-sector LIV *null* (Auger muons), a method paper on Xmax, and a QED vacuum-birefringence claim. Carpet-3 / GRB 221009A UHE photon remains the outstanding *older* anomaly; nothing new on it this window.

---

## Sources fetched this run

- Kostelecký & Russell, arXiv:0801.0287v19 PDF, Tables S2, S3, D6, D20, bibliography [64][66][72][73][165]–[168][187]–[189].
- Dreissen et al., arXiv:2206.00570 PDF, Tables 1–2 and Discussion quote on `3.9×10^{-21}`.
- Stecker arXiv:1306.6095 abs; Nagel arXiv:1412.6954 abs; Kostelecký–Melissinos–Mewes arXiv:1608.02592 abs.
- INSPIRE API `refersto:recid:646314 and date>2023` (36 hits).
- Collins original gr-qc/0403053; Bengochea/Leon/Perez 2405.12534, 2506.15413; Ageev 2512.19670; Perez Teruel 2512.00933.
- Auger arXiv:2602.14720; Martynenko arXiv:2608.05106; Stewart arXiv:2509.19446.