FlowingFlowing
← Back to all development cases
Development case · blind model judgment pending independent human review; not a formal Benchmark conclusion.
AutocompleteNo-Evidence version preferred43 / 162 · ad792f168753bda7

Quantum Computational Complexity -- From Quantum Information to Black Holes and Back

Astrophysics · 2110.14672v1

FLOWING EVIDENCE BENCHMARK

How do we tell whether Evidence helps?

At the same writing position, with the same model and task, how does supplying retrieved paper passages change the first output? We compare matched versions and retain ties, unusable outputs and incomplete reviews.

SAME DRAFT · EVIDENCE ON OR OFF

01 · FIXED WRITING POSITION

MANUSCRIPT

Same writing position ▌

Matched autocomplete at the same draft position

02 · TWO MATCHED INPUTS

SHARED BY BOTH

Manuscript context, model, task and prompt

A · With Evidence

Retrieved paper passages supplied

B · Without Evidence

No retrieved passages supplied

LLM

Same model and version

A → first output

B → first output

Blind judge agent

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

Order check: X / Y → Y / X

HOW DOES BLIND JUDGING WORK?

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

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 autocomplete positions stratified?

Before seeing generation outcomes, we check whether a retrieved passage contains a specific proposition that directly supports the next writing move. Those positions appear in the left opportunity group; the rest are ordinary positions on the right. We select a balanced sample from admitted papers in each field. The 50/50 split is experimental, not a measure of how often either type occurs in writing.

AUTOCOMPLETE · 110

Biology, statistics and astrophysics

55 positions on each side; statistics uses the ten-paper rerun.

AUTOCOMPLETE · 52

Psychology and climate science

26 positions on each side; psychology includes nine papers and climate science four.

How are the table percentages calculated?

Across five fields there are 81 source-opportunity positions. The original blind review preferred Evidence in 51; a task check moved one empty Evidence continuation to both unusable, leaving 50 in public counts. In the ordinary group, another pair of empty outputs moved from no winner to both unusable. Original verdicts remain visible on case pages.

50Evidence version preferred
÷
81All positions in this group
=
62%Evidence preference in this group

Source contribution is a separate review: 16 of 21 Evidence wins entered into source review directly used retrieved papers; another 29 wins await review.

These are development-stage model judgments pending independent human review. They are not formal Benchmark conclusions and do not, on their own, establish causality.

Manuscript writing position

Text excerpt · not a PDF page

Research manuscript · excerpt

Quantum Computational Complexity -- From Quantum Information to Black Holes and Back

1 Introduction

… complexity for generic quantum systems; in particular, we focus on Gaussian states in QFT, both pure and mixed, and on certain classes of CFT states. We then present the conjectured relation to gravitational quantities within the holographic correspondence and discuss several examples in which different versions of the conjectures have been tested. We highlight the relation between complexity, chaos and scrambling in chaotic systems. We conclude with a discussion of open problems and future directions. This article was written for the special issue of EPJ-C Frontiers in Holographic Duality .

… hole. The discovery of the AdS/CFT correspondence – the observation that certain gauge theories are equivalent (or “dual”) to gravitational theories in one higher dimension (see e.g., [ 3 , 4 ] ) – enabled putting the relation between gravity and information on firm ground. Specifically, it permitted Ryu and Takayanagi (RT) to formulate a proposal [ 5 ] (later proven by [ 6 ] ) that relates the entanglement entropy – a quantity characterizing quantum correlations between two regions in conformal field theory (CFT) – and areas of minimal surfaces in asymptotically anti-de Sitter (AdS) spaces.

↓ 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 positionJudgment: No-Evidence version preferred

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.

Evidence

Same model and position · retrieved passages supplied · first output

Not preferred in this comparison

The product produced no insertable continuation text.

The model generated prose, but the product admission rule blocked it as needing evidence. That raw prose has not been independently reviewed here.

Passages supplied to the Evidence version

Entanglement entropy of spherical domains in anti-de Sitter space ↗

It was proposed by Ryu and Takayanagi that the entanglement entropy in conformal field theory (CFT) is related through AdS/CFT correspondence to the area of a minimal surface in the bulk. We apply this holographic geometrical method of calculation of the entanglement entropy to study the vacuum case…

Read full passage excerpt

It was proposed by Ryu and Takayanagi that the entanglement entropy in conformal field theory (CFT) is related through AdS/CFT correspondence to the area of a minimal surface in the bulk. We apply this holographic geometrical method of calculation of the entanglement entropy to study the vacuum case of CFT which is holographically dual to empty anti-de Sitter (AdS) spacetime. We present all possible minimal surfaces spanned on one or two spherical boundaries at AdS infinity. We give exact analytical expressions for regularized areas of these surfaces and identify finite renormalized quantities.

No Evidence

Same model and position · no retrieved passages · first output

Preferred by blind review

The RT proposal provided the first precise dictionary entry between a quantum-information quantity and a geometric object, and it motivated the search for a comparable gravitational dual for quantum computational complexity.

Judgment and task check

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

Read the original model judgment

In this rationale, Output A = Evidence; Output B = No Evidence.

Output A is empty with action 'needs_evidence', but the task requires continuing the manuscript with one coherent academic paragraph. The blank follows a discussion of AdS/CFT and the RT proposal, and the sources provide factual context about RT. An empty response fails the task requirement. Output B provides a coherent continuation: it connects the RT proposal (just discussed in the draft) to the search for a gravitational dual for quantum computational complexity (the paper's stated topic per the first paragraph). The claim that RT 'provided the first precise dictionary entry between a quantum-information quantity and a geometric object' is supported by the draft's description of RT relating 'entanglement entropy... and areas of minimal surfaces.' The claim that it 'motivated the search for a comparable gravitational dual for quantum computational complexity' is a natural bridge to the paper's focus on complexity, grounded in the draft's opening paragraph about 'conjectured relation to gravitational quantities within the holographic correspondence.' No unsupported factual claims are introduced.

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.