FlowingFlowing
← 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 better234 / 280 · beu-39f2e6df40a78994

Neutrino Geoscience: Review, survey, future prospects

Climate and environmental science · 2209.13746v1

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

Neutrino Geoscience: Review, survey, future prospects

8 Future prospects

Such an instrument is capable of accurately determining the mantle geoneutrino flux, which in turned can be used to unravel the geological signals measured at land-based experiments. Given this, we would have the capability to interpret the local lithospheric signal from all of the existing and future detectors.

…geoneutrino flux measurements from particle physics experiments with flux predictions derived from integrating data from geology, geophysics, and geochemistry. The particle physics community are focusing on improving their measurement precision and detector technology. Efforts in geosciences are focused on improving our models for the abundance and distribution of the heat producing elements. Improvements in crustal studies are of greatest need. Global and regional geochemical studies, including studies of crustal terrains and xenoliths can help to constrain the composition of the deep crust.

↓ 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 = with cards · B = without cards

Both outputs provide coherent academic paragraphs, but B fits the draft's argumentative trajectory significantly better. The draft moves from detector capability → geoneutrino flux measurements → particle physics precision improvements → geoscience model improvements → crustal studies as 'greatest need.' Output B directly continues this logic: it addresses why crustal uncertainty matters (separating mantle from crustal contributions), ties back to the 'land-based detectors' mentioned in the draft's opening, and reinforces the stated priority that 'improvements in crustal studies are of greatest need.' Output A, while factually accurate and well-supported by sources, shifts to a different topic—advantages of neutrinos over conventional radiometric methods—which is more introductory/background material. The draft has already established what geoneutrinos are and why they matter; it doesn't need to revisit the 'deep-penetrating properties' point at this position. A's content would fit better earlier in the manuscript. B's tighter thematic coherence with the preceding paragraph—especially its echo of 'land-based' detectors and focus on the crust/mantle separation problem—makes it the more useful continuation.

Evidence

Same model and position · retrieved passages supplied · first output

Not preferred in this comparison

A dedicated geoneutrino detector for geosciences would exploit the deep-penetrating properties of neutrinos, which make this radiation in the decay of the heat producing elements ideally suited for investigating the deep interior of the Earth relative to conventional radiometric methods employing alpha-, beta- and gamma rays.

Passages supplied to the Evidence version

Card 1 · A reference Earth model for the heat producing elements and associated geoneutrino flux

The recent geoneutrino experimental results from KamLAND and Borexino detectors reveal the usefulness of analyzing the Earth geoneutrino flux, as it provides a constraint on the strength of the radiogenic heat power and this, in turn, provides a test of compositional models of the bulk silicate Eart…

Read full passage excerpt

The recent geoneutrino experimental results from KamLAND and Borexino detectors reveal the usefulness of analyzing the Earth geoneutrino flux, as it provides a constraint on the strength of the radiogenic heat power and this, in turn, provides a test of compositional models of the bulk silicate Earth (BSE). This flux is dependent on the amount and distribution of heat producing elements (HPEs: U, Th and K) in the Earth interior. We have developed a geophysically-based, three-dimensional global reference model for the abundances and distributions of HPEs in the BSE. The structure and composition of the outermost portion of the Earth, the crust and underlying lithospheric mantle, is detailed in the reference model, this portion of the Earth has the greatest influence on the geoneutrino fluxes. The reference model combines three existing geophysical models of the global crust and yields an average crustal thickness of 34.4+-4.1 km in the continents and 8.0+-2.7 km in the oceans. In situ s …

Card 2 · Reference Models for Lithospheric Geoneutrino Signal

Abstract Debate continues on the amount and distribution of radioactive heat producing elements (i.e., U, Th, and K) in the Earth, with estimates for mantle heat production varying by an order of magnitude. Constraints on the bulk‐silicate Earth's (BSE) radiogenic power also places constraints on ov…

Read full passage excerpt

Abstract Debate continues on the amount and distribution of radioactive heat producing elements (i.e., U, Th, and K) in the Earth, with estimates for mantle heat production varying by an order of magnitude. Constraints on the bulk‐silicate Earth's (BSE) radiogenic power also places constraints on overall BSE composition. Geoneutrino detection is a direct measure of the Earth's decay rate of Th and U. The geoneutrino signal has contributions from the local ( 40%) and global ( 35%) continental lithosphere and the underlying inaccessible mantle ( 25%). Geophysical models are combined with geochemical data sets to predict the geoneutrino signal at current and future geoneutrino detectors. We propagated uncertainties, both chemical and physical, through Monte Carlo methods. Estimated total signal uncertainties are on the order of 20%, proportionally with geophysical and geochemical inputs contributing 30% and 70%, respectively. We find that estimated signals, calculated using CRUST2.0, CRUS …

Card 3 · Comprehensive geoneutrino analysis with Borexino

This paper presents a comprehensive geoneutrino measurement using the Borexino detector, located at Laboratori Nazionali del Gran Sasso (LNGS) in Italy.The analysis is the result of 3262.74 days of data between December 2007 and April 2019.The paper describes improved analysis techniques and optimiz…

Read full passage excerpt

This paper presents a comprehensive geoneutrino measurement using the Borexino detector, located at Laboratori Nazionali del Gran Sasso (LNGS) in Italy.The analysis is the result of 3262.74 days of data between December 2007 and April 2019.The paper describes improved analysis techniques and optimized data selection, which includes enlarged fiducial volume and sophisticated cosmogenic veto.The reported exposure of ð1.29 Æ 0.05Þ × 10 32 protons × year represents an increase by a factor of two over a previous Borexino analysis reported in 2015.By observing 52.6 þ9.4 -8.6 ðstatÞ þ2.7 -2.1 ðsysÞ geoneutrinos (68% interval) from 238 U and 232 Th, a geoneutrino signal of 47.0 þ8.4 -7.7 ðstatÞ þ2.4 -1.9 ðsysÞ TNU with þ18.3 -17.2 % total precision was obtained.This result assumes the same Th/U mass ratio as found in chondritic CI meteorites but compatible results were found when contributions from 238 U and 232 Th were both fit as free parameters.Antineutrino background from reactors is fit u …

Card 4 · Considerations For a Dedicated Geoneutrino Detector For Geosciences

A combination of several sources including: radiogenic heating, processes of mantle and core formation and differentiation, delayed radiogenic heating, earthquakes, and tidal friction account for the surface heat flux in the Earth. Radiogenic heating is of much interest in various fields of geoscien…

Read full passage excerpt

A combination of several sources including: radiogenic heating, processes of mantle and core formation and differentiation, delayed radiogenic heating, earthquakes, and tidal friction account for the surface heat flux in the Earth. Radiogenic heating is of much interest in various fields of geosciences. Inferences from recent experiments with reactor antineutrinos and solar neutrinos showed that the age of geoneutrinos is at hand for constraining radiogenic heat. Because of the deep penetrating properties of the neutrinos this type of radiation in the decay of the heat producing elements (HPE) is ideally suited for the investigation of the deep interiors of the Earth compared to conventional radiometric methods for HPE employing alpha-, beta- and gamma rays. This presentation will address the considerations for a dedicated geoneutrino detector to be set up for investigating the interior regions all the way to the center of the Earth.

Model-reported card use

Card 4 · fact

Affected output: A dedicated geoneutrino detector for geosciences would exploit the deep-penetrating properties of neutrinos, which make this radiation in the decay of the heat producing elements ideally suited for investigating the deep interior of the Earth relative to conventional radiometric methods employing alpha-, beta- and gamma rays.

Model explanation: The row supplied the concrete rationale for a dedicated geoscience detector—deep-penetrating neutrino radiation from HPE decay versus conventional radiometric methods—which the paragraph otherwise lacked.

No Evidence

Same model and position · no retrieved passages · first output

Preferred by blind review

Beyond refining compositional models, the value of geoneutrino measurements ultimately depends on the precision with which the mantle and crustal contributions can be separated, since the lithospheric signal dominates the total flux at most land-based sites. Reducing this crustal uncertainty therefore remains the most direct route to resolving the mantle component.

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.

The blinded model review preferred the no-Evidence version. Its original rationale appears below.

Read the original model judgment

Both outputs provide coherent academic continuations, but A better serves the manuscript's argumentative flow. The draft establishes that geoneutrino flux measurements can separate mantle from lithospheric signals, then notes that 'Improvements in crustal studies are of greatest need.' Output A directly builds on this by explaining WHY crustal uncertainty reduction matters—because lithospheric signal dominates and separating mantle/crustal contributions is essential for resolving the mantle component. This creates logical continuity with both the preceding sentence about crustal studies being 'of greatest need' and the broader goal of interpreting 'the local lithospheric signal from all of the existing and future detectors' stated in the first paragraph. Output B, while factually grounded in Source 4 about dedicated detectors and neutrino penetration properties, introduces a somewhat tangential point about detector advantages over conventional radiometric methods. This shifts focus away from the draft's central thread about integrating particle physics measurements with geoscience models and the specific priority of crustal improvements. The draft already assumes geoneutrino detection is valuable; it doesn't need justification for why neutrinos are useful. A maintains topical coherence; B reads more like a detour into detector physics that doesn't advance the paper's stated purpose.

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.