← 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.
AutocompleteInconclusive243 / 280 · beu-48068e9e0eabee8f

Computational aspects of the Volterra Signature

Mathematics · 2605.18406v1

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

Computational aspects of the Volterra Signature

Introduction / related work

hted monomial expansion (2.15) in Sections 3 and 4. Like in Corollary 2.14, we draw attention to the scalar caseK=kId, where the complexity of the inner summation reduces to the composition of the tensor levelnintokparts. Corollary 2.17.Letxbe a piecewise linear path as introduced in Notation 2.10. Similarly to Corollary 2.14, assume thatK=kI d for a scalar kernelk∈L ∞,1(∆2;R). Then for all 0≤j≤l≤Jandn∈N ≥1 it holds VSig(x;k) (n),tl t0,tj = j∧nX k=1 X 0<i1<···<ik≤j n1+···+nk=n µn1,...,nk i1,...,ik,lv⊗n1 i1 ⊗···⊗v ⊗nk ik ,(2.18) wheren l >0in the inner summation and the weightsµn1,...,nk i1,...,ik+1 are defined by µn1,...,nk i1,...,ik+1 := Z ti1 ti1−1 ··· Z tik tik−1 kY l=1 ˙κnl,rl+1 rl,til drl ∈R,where we setr k+1 =t ik+1,(2.19) with a kernel˙κgiven by ˙κn,τ s,t = Z ∆n−1 s,t k(r1,r 0) n−1Y a=1 k(ra+1,ra) dra,with the conventionr 0 =s, rn =τ.(2.20) Remark2.18.The above form reveals the relation of the Volterra signature to theweighted iterated sum signature[11], although here the weights in general depend on the grading as well. Remark2.19.In the scalar kernel settingK=kI d, the kernel˙κn introduced in (2.20) naturally gives rise to the companion quantity κn,τ s,t := Z t s ˙κn,τ u,t

↓ 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: Inconclusive

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.

Review passes disagreed and the usability assessments conflict, so this pair counts neither as a tie nor as a win.

Read overall AI review rationale

This rationale combines reviews with different display orders.

Order-swapped judges disagreed. Pass 1: Output A correctly identifies that the draft is at a natural completion point. The sentence introduces a companion quantity κ_{s,t}^{n,τ} via an integral equation (2.20), and the mathematical definition is complete with the 'with the convention r^0 = s, r^n = τ' clause. The draft has finished defining this new quantity, and continuing with another sentence would likely be redundant or speculative without seeing what the next intended move is (e.g., properties of κ, or its use in a subsequent theorem). This is a deliberate, appropriate abstention. Output B provides the text 'du,.' which is nonsensical garbage. The 'du' appears to be a fragment of an integral measure (like 'du' in integration), but it's malformed, has a comma and period with no coherent sentence structure, and does not continue the argument in any meaningful way. This is unusable as it would require substantial repair and contradicts the instruction to provide 'one useful sentence' that 'preserves the argument.' Pass 2: The draft ends mid-formula with κ^{n,τ}_{s,t} := ∫_s^t \dot{κ}^{n,τ}_{u,t} du, clearly needing completion of this definition. Output A provides 'du,.' which is nonsensical—'du' partially completes the integral notation but the comma and period are garbled, making it unusable without repair. Output B abstains with 'complete'. While the formula genuinely needs completion (the differential 'du' and likely a closing statement), Output A's continuation is broken and would require substantial repair. An appropriate abstention is preferable to a broken completion. However, re-evaluating: the task asks for 'one useful sentence' and the position clearly needs continuation. Output B's abstention is therefore inappropriate when the draft is incomplete. Output A, despite its garbled form, attempts to provide the needed 'du' differential. Given the strict rules that a deliberate 'complete' is unusable when 'the task clearly needs a next move', and this formula is visibly incomplete, Output B is unusable. Output A, while garbled, contains the correct differential 'du' which is the necessary completion.

Evidence

Same model and position · retrieved passages supplied · first output

du,.

Passages supplied to the Evidence version

Card 1 · The Volterra signature

Modern approaches for learning from non-Markovian time series, such as recurrent neural networks, neural controlled differential equations or transformers, typically rely on implicit memory mechanisms that can be difficult to interpret or to train over long horizons. We propose the \emph{Volterra si…

Read full passage excerpt

Modern approaches for learning from non-Markovian time series, such as recurrent neural networks, neural controlled differential equations or transformers, typically rely on implicit memory mechanisms that can be difficult to interpret or to train over long horizons. We propose the \emph{Volterra signature} $\mathrm{VSig}(x;K)$ as a principled, explicit feature representation for history-dependent systems. By developing the input path $x$ weighted by a temporal kernel $K$ into the tensor algebra, we leverage the associated Volterra--Chen identity to derive rigorous learning-theoretic guarantees. Specifically, we prove an \emph{injectivity} statement (identifiability under augmentation) that leads to a \emph{universal approximation} theorem on the infinite dimensional path space, which in certain cases is achieved by \emph{linear functionals} of $\mathrm{VSig}(x;K)$. Moreover, we demonstrate applicability of the \emph{kernel trick} by showing that the inner product associated with Volte …

Card 2 · Signature SDEs from an affine and polynomial perspective

Signature stochastic differential equations (SDEs) constitute a large class of stochastic processes, here driven by Brownian motions, whose characteristics are linear maps of their own signature, i.e. of iterated integrals of the process with itself, and therefore allow for a generic path dependence…

Read full passage excerpt

Signature stochastic differential equations (SDEs) constitute a large class of stochastic processes, here driven by Brownian motions, whose characteristics are linear maps of their own signature, i.e. of iterated integrals of the process with itself, and therefore allow for a generic path dependence. We show that their prolongation with the corresponding signature is an affine and polynomial process taking values in the set of group-like elements of the extended tensor algebra. By relying on duality theory for affine or polynomial processes, we obtain explicit formulas in terms of converging power series for the Fourier-Laplace transform and the expected value of entire functions of the signature process’ marginals. The coefficients of these power series are solutions of Riccati and linear ordinary differential equations (ODEs) with values in the extended tensor algebra, respectively, whose vector fields can be expressed in terms of the characteristics of the corresponding SDEs. We thu …

Card 3 · The Volterra signature

Modern approaches for learning from non-Markovian time series, such as recurrent neural networks, neural controlled differential equations or transformers, typically rely on implicit memory mechanisms that can be difficult to interpret or to train over long horizons. We propose the \emph{Volterra si…

Read full passage excerpt

Modern approaches for learning from non-Markovian time series, such as recurrent neural networks, neural controlled differential equations or transformers, typically rely on implicit memory mechanisms that can be difficult to interpret or to train over long horizons. We propose the \emph{Volterra signature} $\mathrm{VSig}(x;K)$ as a principled, explicit feature representation for history-dependent systems. By developing the input path $x$ weighted by a temporal kernel $K$ into the tensor algebra, we leverage the associated Volterra--Chen identity to derive rigorous learning-theoretic guarantees. Specifically, we prove an \emph{injectivity} statement (identifiability under augmentation) that leads to a \emph{universal approximation} theorem on the infinite dimensional path space, which in certain cases is achieved by \emph{linear functionals} of $\mathrm{VSig}(x;K)$. Moreover, we demonstrate applicability of the \emph{kernel trick} by showing that the inner product associated with Volte …

Card 4 · Volterra equations driven by rough signals

This article is devoted to the extension of the theory of rough paths in the context of Volterra equations with possibly singular kernels. We begin to describe a class of two parameter functions defined on the simplex called Volterra paths. These paths are used to construct a so-called Volterra-sign…

Read full passage excerpt

This article is devoted to the extension of the theory of rough paths in the context of Volterra equations with possibly singular kernels. We begin to describe a class of two parameter functions defined on the simplex called Volterra paths. These paths are used to construct a so-called Volterra-signature, analogously to the signature used in Lyon’s theory of rough paths. We provide a detailed algebraic and analytic description of this object. Interestingly, the Volterra signature does not have a multiplicative property similar to the classical signature, and we introduce an integral product behaving like a convolution extending the classical tensor product. We show that this convolution product is well defined for a large class of Volterra paths, and we provide an analogue of the extension theorem from the theory of rough paths (which guarantees in particular the existence of a Volterra signature). Moreover the concept of convolution product is essential in the construction of Volterra …

Card 5 · Signature-Based Universal Bilinear Approximations for Nonlinear Systems and Model Order Reduction

This paper deals with non-Lipschitz nonlinear systems. Such systems can be approximated by a linear map of so-called signatures, which play a crucial role in the theory of rough paths and can be interpreted as collections of iterated integrals involving the control process. As a consequence, we iden…

Read full passage excerpt

This paper deals with non-Lipschitz nonlinear systems. Such systems can be approximated by a linear map of so-called signatures, which play a crucial role in the theory of rough paths and can be interpreted as collections of iterated integrals involving the control process. As a consequence, we identify a universal bilinear system, solved by the signature, that can approximate the state or output of the original nonlinear dynamics arbitrarily well. In contrast to other (bi)linearization techniques, the signature approach remains feasible in large-scale settings, as the dimension of the associated bilinear system grows only with the number of inputs. However, the signature model is typically of high order, requiring an optimization process based on model order reduction (MOR). We derive an MOR method for unstable bilinear systems with non-zero initial states and apply it to the signature, yielding a potentially low-dimensional bilinear model. An advantage of our method is that the origi …

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

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.