← Back to all development cases
This case was judged by an anonymized AI reviewer and has not been independently reviewed by people; inspect the draft, cards, and both outputs to assess it.
AutocompleteWithout cards better194 / 280 · beu-7169696bef76b5dd

Quantum Gravity from dynamical metric fluctuations

Astrophysics · 2309.10785v1

FLOWING EVIDENCE BENCHMARK

How do we compare writing with and without cards?

At the same writing position, the same model produces one first result with cards and one without. All 280 pairs enter the overall anonymized review, which assesses whether a decision not to continue fits the draft.

Overall: 138 with-card wins, 26 ties, 107 no-card wins, 5 pairs where neither was usable, and 4 inconclusive judgments. Both sides wrote text in 218 pairs; 48 had a deliberate stop and 14 included an invalid output.

SAME DRAFT · WITH OR WITHOUT CARDS

01 · FIXED WRITING POSITION

MANUSCRIPT

Same writing position ▌

Matched autocomplete at the same draft position

02 · TWO MATCHED VERSIONS

SHARED BY BOTH

Draft, writing task, and generation model

A · With cards

Relevant literature cards supplied

B · Without cards

No literature cards supplied

Generation model

Same model and version

A → first output

B → first output

Anonymized AI judge

First outputs are anonymized as X and Y

Continuation review: accuracy, fit to the writing task and usability
Returns: X preferred / tie / Y preferred / both unusable

Review again with the answer order swapped

HOW DOES BLIND JUDGING WORK?

① Anonymize both outputs
The AI judge sees the same draft and both first outputs without knowing which received cards.

② Compare and swap order
The judge applies task-specific criteria in X/Y and then Y/X order.

③ Review disagreements
A third pass resolves disagreements. Incomplete reviews remain in the denominator.

How are positions and cards selected?

We selected 40 writing positions in each of seven fields, for 280 comparisons. The position and relevant literature cards were set before each comparison, so both versions began from the same draft.

The main result retains every frozen position, including places where cards might not add a useful next move. This avoids selecting cases based on their outputs, but 49% describes the whole cohort, not a win rate where evidence clearly had an opportunity to help. This cohort did not record that opportunity label for each position before generation, so those two groups cannot yet be separated reliably.

CARD USE · 140/280

Reported card use in continuations with text

The model reports whether a card affected its continuation. In 140 of 280 tests it produced continuation text and reported card use. The previous count of 150 included 10 cases without continuation text and has been corrected.

COMPARISON · 70 / 7 / 30

Which written version was better?

Of the 107 reported-use cases where both versions produced text, the blinded AI judge preferred the with-card version 70 times, found 7 about the same, and preferred the without-card version 30 times.

How are the two table percentages calculated?

ADOPTION COVERAGE

140reported use with text
÷
280all positions
=
50%

WITH-CARD VERSION PREFERRED

70with cards better
÷
107comparable tests
=
65%

The percentages use different groups: 50% counts continuations with text reporting card use among all 280 tests; 65% covers the 107 such cases with text on both sides. The generating model reports card use, while an anonymized AI judge compares writing quality.

Card use is self-reported by the generator. The anonymized AI judge sees the draft and source snippets, so source-aware preference remains possible; there was no independent human review. The overall result describes these 280 frozen positions, not a win probability for any writing task.

Manuscript writing position

Text excerpt · not a PDF page

Research manuscript · excerpt

Quantum Gravity from dynamical metric fluctuations

1 Introduction

…tail the derivation of the respective flow equations on space-time manifolds with Euclidean and Lorentzian signatures and discuss the diffeomorphism symmetry constraints on the flow as well as the convergence of systematic vertex expansion schemes. We then proceed with a comprehensive review of results of momentum-dependent correlation functions at vanishing cutoff scale, the phase structure of the asymptotically safe Standard Model, and spectral properties of asymptotically safe gravity from direct computations in space-times with Lorentzian signatures such as the graviton spectral function.

↓ Next paragraph was generated here

The highlighted text is near the writing position. This is a layout of the evaluation input, not a PDF screenshot.

A matched comparison at the same writing positionOverall AI review: Without cards better

Both arms use the same model, task and manuscript context. Only the retrieved passages differ. The first recorded output from each arm is shown below.

Read overall AI review rationale

A = without cards · B = with cards

The draft is a complete, self-contained abstract/introduction paragraph that already states what the paper will cover: derivation of flow equations on space-time manifolds with Euclidean and Lorentzian signatures, diffeomorphism symmetry constraints, convergence of vertex expansion schemes, momentum-dependent correlation functions, phase structure of the asymptotically safe Standard Model, and spectral properties of asymptotically safe gravity. This is a classic abstract or introductory paragraph that maps out the paper's contents. Adding an organizational paragraph ('This review is organised as follows...') immediately after would be redundant because the draft already performs the same function—telling the reader what will be discussed. The 'complete' action in Output A correctly recognizes that the text already makes a complete move and continuing would be redundant. Output B's organizational paragraph essentially repeats the same roadmap in a more tedious form without adding substantive content. The draft does not need a continuation at this cursor position; it stands as a complete thought.

Evidence

Same model and position · retrieved passages supplied · first output

Not preferred in this comparison

This review is organised as follows. Section 2 introduces the functional renormalisation group for gravity and the fluctuation approach, Section 3 discusses the derivation of the flow equations in Euclidean and Lorentzian signatures together with diffeomorphism constraints and the convergence of vertex expansions, and Section 4 reviews momentum-dependent correlation functions, the phase structure of the asymptotically safe Standard Model, and spectral properties of asymptotically safe gravity.

Passages supplied to the Evidence version

Card 1 · The Asymptotically Safe Standard Model: From quantum gravity to dynamical chiral symmetry breaking

We present a comprehensive non-perturbative study of the phase structure of the asymptotically safe Standard Model. The physics scales included range from the asymptotically safe trans-Planckian regime in the ultraviolet, the intermediate high-energy regime with electroweak symmetry breaking to stro…

Read full passage excerpt

We present a comprehensive non-perturbative study of the phase structure of the asymptotically safe Standard Model. The physics scales included range from the asymptotically safe trans-Planckian regime in the ultraviolet, the intermediate high-energy regime with electroweak symmetry breaking to strongly correlated QCD in the infrared. All flows are computed with a self-consistent functional renormalisation group approach, using a vertex expansion in the fluctuation fields. In particular, this approach takes care of all physical threshold effects and the respective decoupling of ultraviolet degrees of freedom. Standard Model and gravity couplings and masses are fixed by their experimental low energy values. Importantly, we accommodate for the difference between the top pole mass and its Euclidean analogue. Both, the correct mass determination and the threshold effects have a significant impact on the qualitative properties, and in particular on the stability properties of the specific u …

Card 2 · Lorentzian quantum gravity and the graviton spectral function

We present the first direct and non-perturbative computation of the graviton spectral function in quantum gravity. This is achieved with the help of a novel Lorentzian renormalisation group approach, combined with a spectral representation of correlation functions. We find a positive graviton spectr…

Read full passage excerpt

We present the first direct and non-perturbative computation of the graviton spectral function in quantum gravity. This is achieved with the help of a novel Lorentzian renormalisation group approach, combined with a spectral representation of correlation functions. We find a positive graviton spectral function, showing a massless one-graviton peak and a multi-graviton continuum with an asymptotically safe scaling for large spectral values. We also study the impact of a cosmological constant. Further steps to investigate scattering processes and unitarity in asymptotically safe quantum gravity are indicated.

Card 3 · Foliated asymptotically safe gravity in the fluctuation approach

The gravitational asymptotic safety program envisions a high-energy completion of gravity based on a non-Gaussian renormalization group fixed point. A key step in this program is the transition from Euclidean to Lorentzian signature spacetimes. One way to address this challenge is to formulate the q…

Read full passage excerpt

The gravitational asymptotic safety program envisions a high-energy completion of gravity based on a non-Gaussian renormalization group fixed point. A key step in this program is the transition from Euclidean to Lorentzian signature spacetimes. One way to address this challenge is to formulate the quantum theory based on the Arnowitt-Deser-Misner decomposition of the metric field. This equips the Euclidean spacetime with a preferred direction which may serve as the time-direction in the Lorentzian setting. In this work we use the Wetterich equation in order to compute the renormalization group flow of the graviton two-point function. The resulting beta functions possess a non-Gaussian renormalization group fixed point suitable for rendering the theory asymptotically safe. The phase diagram underlying the flow of the two-point function is governed by the interplay between this non-Gaussian fixed point, the Gaussian fixed point, and an infrared fixed point. The latter ensures that the re …

Card 4 · Neutrino spectral density at electroweak scale temperature

Motivated by the scenario of resonant leptogenesis in which lepton number creation in the electroweak scale is relevant, we investigate the spectral properties and possible collective nature of the standard model neutrinos at electroweak scale temperature ($T$). We adopt the ${R}_{\ensuremath{\xi}}$…

Read full passage excerpt

Motivated by the scenario of resonant leptogenesis in which lepton number creation in the electroweak scale is relevant, we investigate the spectral properties and possible collective nature of the standard model neutrinos at electroweak scale temperature ($T$). We adopt the ${R}_{\ensuremath{\xi}}$ gauge fixing, which includes the unitary gauge as a limiting case, and allows us to study the broken as well as the restored phases of the gauge symmetry in a unified way. We show that the spectral density of the neutrino has a three-peak structure in the low-momentum region due to the scattering with the thermally excited particles (i.e., Landau damping) for $T\ensuremath{\gtrsim}{M}_{\mathrm{W},\mathrm{Z}}(T)$, with ${M}_{\mathrm{W},\mathrm{Z}}(T)$ being the weak-boson masses in the plasma. The three peaks are identified with a novel ultrasoft mode, the usual quasiparticle, and antiplasmino modes. Varying the gauge-fixing parameter $\ensuremath{\xi}$, we show that the three-peak structure …

Card 5 · Lorentzian version of the noncommutative geometry of the standard model of particle physics

A formulation of the noncommutative geometry for the standard model of particle physics with a Lorentzian signature metric is presented. The elimination of the fermion doubling in the Lorentzian case is achieved by a modification of Connes’ internal space geometry [“Gravity coupled with matter and t…

Read full passage excerpt

A formulation of the noncommutative geometry for the standard model of particle physics with a Lorentzian signature metric is presented. The elimination of the fermion doubling in the Lorentzian case is achieved by a modification of Connes’ internal space geometry [“Gravity coupled with matter and the foundation of non-commutative geometry,” Commun. Math. Phys. 182, 155–176 (1996)] so that it has signature 6 (mod 8) rather than 0. The fermionic part of the Connes-Chamseddine spectral action can be formulated, and it is shown that it allows an extension with right-handed neutrinos and the correct mass terms for the seesaw mechanism of neutrino mass generation.

Model-reported card use

The model did not report that any card materially affected this completion.

No Evidence

Same model and position · no retrieved passages · first output

Preferred by blind review

No continuation text was produced.

Model action:complete

Earlier card-use comparison

The earlier diagnostic compared only cases with reported card use and text on both sides; its scope differs from the overall AI review above.

This case does not have both with-card continuation text and a model-reported use record. It remains among all 280 tests but does not enter the reported-use comparison.

The passages are input material; their presence does not mean an output used them or that they support every claim. Draft and source passages are excerpted; no simulated scores or PDF appear here.