← 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.
AutocompleteWith cards better245 / 280 · beu-7b93658ae8724a84

A coupled Kolmogorov-Arnold Network and Level-Set framework for evolving interfaces

Mathematics · 2601.09818v2

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

A coupled Kolmogorov-Arnold Network and Level-Set framework for evolving interfaces

Introduction / related work

nterfaces. Consequently, solving moving boundary PDEs generally requires deeper network ar- chitectures, which increases training cost and hampers the interpretability of the solution [14]. In a recent study [15] investigated the use of PINN for phase transition problems, with adaptive weighting scheme and residual balancing and require substantial optimization effort and tuning. 1 arXiv:2601.09818v2 [math-ph] 9 Feb 2026

Another study by Mullins et al. [16] used MLP-type architecture for level-set equation to handle interfacial dynamics and demonstrated comparable performance similar to conventional techniques but may need careful tuning of weights and deep networks may demand very high sampling to extend it to coupled Stefan-type interface problems. In contrast to MLP based architectures, the recently proposed Kolmogorov-Arnold Networks (KANs) [17], grounded in the Kolmogorov-Arnold representation theorem [18, 19], have better interpretability and accuracy with shallower networks. They incorporate nonlinearity directly into learnable basis functions on the network edges rather than relying on fixed node activations, KANs with shallower networks matched comparable acurracy of

↓ 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: With 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 complete the sentence about KANs matching comparable accuracy of deeper MLP architectures. The draft context emphasizes training cost, interpretability, and shallower networks as key advantages of KANs over MLPs for PDE solving. Output A better preserves this argument by adding 'while using substantially fewer parameters'—directly echoing the draft's concern about training cost and the source material's claim that 'much smaller KANs can achieve comparable or better accuracy than much larger MLPs.' Output B's 'across a range of function-fitting and PDE benchmarks' is plausible but slightly broader than the immediate context; the draft is specifically about PDEs (moving boundary PDEs, phase transition, Stefan-type problems), and 'function-fitting' introduces a slight scope drift. Output A's parameter-efficiency point is more tightly integrated with the visible argument about architecture depth and cost.

Evidence

Same model and position · retrieved passages supplied · first output

Preferred by blind review

deeper MLP architectures while using substantially fewer parameters.

Passages supplied to the Evidence version

Card 1 · KAN: Kolmogorov-Arnold Networks

Inspired by the Kolmogorov-Arnold representation theorem, we propose Kolmogorov-Arnold Networks (KANs) as promising alternatives to Multi-Layer Perceptrons (MLPs). While MLPs have fixed activation functions on nodes ("neurons"), KANs have learnable activation functions on edges ("weights"). KANs hav…

Read full passage excerpt

Inspired by the Kolmogorov-Arnold representation theorem, we propose Kolmogorov-Arnold Networks (KANs) as promising alternatives to Multi-Layer Perceptrons (MLPs). While MLPs have fixed activation functions on nodes ("neurons"), KANs have learnable activation functions on edges ("weights"). KANs have no linear weights at all -- every weight parameter is replaced by a univariate function parametrized as a spline. We show that this seemingly simple change makes KANs outperform MLPs in terms of accuracy and interpretability. For accuracy, much smaller KANs can achieve comparable or better accuracy than much larger MLPs in data fitting and PDE solving. Theoretically and empirically, KANs possess faster neural scaling laws than MLPs. For interpretability, KANs can be intuitively visualized and can easily interact with human users.

Card 2 · PO-CKAN:Physics Informed Deep Operator Kolmogorov Arnold Networks with Chunk Rational Structure

We propose PO-CKAN, a physics-informed deep operator framework based on Chunkwise Rational Kolmogorov--Arnold Networks (KANs), for approximating the solution operators of partial differential equations. This framework leverages a Deep Operator Network (DeepONet) architecture that incorporates Chunkw…

Read full passage excerpt

We propose PO-CKAN, a physics-informed deep operator framework based on Chunkwise Rational Kolmogorov--Arnold Networks (KANs), for approximating the solution operators of partial differential equations. This framework leverages a Deep Operator Network (DeepONet) architecture that incorporates Chunkwise Rational Kolmogorov-Arnold Network (CKAN) sub-networks for enhanced function approximation. The principles of Physics-Informed Neural Networks (PINNs) are integrated into the operator learning framework to enforce physical consistency. This design enables the efficient learning of physically consistent spatio-temporal solution operators and allows for rapid prediction for parametric time-dependent PDEs with varying inputs (e.g., parameters, initial/boundary conditions) after training. Validated on challenging benchmark problems, PO-CKAN demonstrates accurate operator learning with results closely matching high-fidelity solutions. PO-CKAN adopts a DeepONet-style branch--trunk architecture …

Card 3 · GINN-KAN: Interpretability pipelining with applications in Physics Informed Neural Networks

Neural networks are powerful function approximators, yet their ``black-box" nature often renders them opaque and difficult to interpret. While many post-hoc explanation methods exist, they typically fail to capture the underlying reasoning processes of the networks. A truly interpretable neural netw…

Read full passage excerpt

Neural networks are powerful function approximators, yet their ``black-box" nature often renders them opaque and difficult to interpret. While many post-hoc explanation methods exist, they typically fail to capture the underlying reasoning processes of the networks. A truly interpretable neural network would be trained similarly to conventional models using techniques such as backpropagation, but additionally provide insights into the learned input-output relationships. In this work, we introduce the concept of interpretability pipelineing, to incorporate multiple interpretability techniques to outperform each individual technique. To this end, we first evaluate several architectures that promise such interpretability, with a particular focus on two recent models selected for their potential to incorporate interpretability into standard neural network architectures while still leveraging backpropagation: the Growing Interpretable Neural Network (GINN) and Kolmogorov Arnold Networks (KA …

Card 4 · A Survey on Kolmogorov-Arnold Network

This review study explores the theoretical foundations, evolution, applications, and future potential of Kolmogorov–Arnold Networks (KAN), a neural network model inspired by the Kolmogorov–Arnold representation theorem. KANs set themselves apart from traditional neural networks by employing learnabl…

Read full passage excerpt

This review study explores the theoretical foundations, evolution, applications, and future potential of Kolmogorov–Arnold Networks (KAN), a neural network model inspired by the Kolmogorov–Arnold representation theorem. KANs set themselves apart from traditional neural networks by employing learnable, spline-parameterized functions rather than fixed activation functions, allowing for flexible and interpretable representations of high-dimensional functions. The review explores Kan’s architectural strengths, including adaptive edge-based activation functions that enhance parameter efficiency and scalability across varied applications such as time series forecasting, computational biomedicine, and graph learning. Key advancements including Temporal-KAN (T-KAN), FastKAN, and Partial Differential Equation (PDE) KAN illustrate KAN’s growing applicability in dynamic environments, significantly improving interpretability, computational efficiency, and adaptability for complex function approxim …

Card 5 · Kolmogorov Arnold Informed neural network: A physics-informed deep learning framework for solving forward and inverse problems based on Kolmogorov Arnold Networks

AI for partial differential equations (PDEs) has garnered significant attention, particularly with the emergence of Physics-informed neural networks (PINNs). The recent advent of Kolmogorov-Arnold Network (KAN) indicates that there is potential to revisit and enhance the previously MLP-based PINNs.…

Read full passage excerpt

AI for partial differential equations (PDEs) has garnered significant attention, particularly with the emergence of Physics-informed neural networks (PINNs). The recent advent of Kolmogorov-Arnold Network (KAN) indicates that there is potential to revisit and enhance the previously MLP-based PINNs. Compared to MLPs, KANs offer interpretability and require fewer parameters. PDEs can be described in various forms, such as strong form, energy form, and inverse form. While mathematically equivalent, these forms are not computationally equivalent, making the exploration of different PDE formulations significant in computational physics. Thus, we propose different PDE forms based on KAN instead of MLP, termed Kolmogorov-Arnold-Informed Neural Network (KINN) for solving forward and inverse problems. We systematically compare MLP and KAN in various numerical examples of PDEs, including multi-scale, singularity, stress concentration, nonlinear hyperelasticity, heterogeneous, and complex geometr …

Model-reported card use

Card 1 · fact

Affected output: deeper MLP architectures while using substantially fewer parameters.

Model explanation: Supplied the concrete comparison that smaller KANs achieve comparable accuracy to much larger MLPs, completing the unfinished clause about shallower networks matching MLP accuracy.

No Evidence

Same model and position · no retrieved passages · first output

Not preferred in this comparison

deeper MLP architectures across a range of function-fitting and PDE benchmarks.

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 Evidence version. Its original rationale appears below.

Read the original model judgment

Both outputs attempt to complete the sentence about KANs matching comparable accuracy of deeper MLP architectures. Output A ('deeper MLP architectures while using substantially fewer parameters') is grounded: the sources explicitly state KANs 'outperform MLPs in terms of accuracy and interpretability' and 'much smaller KANs can achieve comparable or better accuracy than much larger MLPs in data fitting and PDE solving' (source [17]). The claim about 'substantially fewer parameters' is directly supported by 'smaller KANs' vs 'much larger MLPs' and 'require fewer parameters' (source KINN). Output B adds 'across a range of function-fitting and PDE benchmarks' which is an unsupported specificity. The sources mention 'data fitting and PDE solving' and 'various numerical examples of PDEs' but do not establish that KANs were tested 'across a range of function-fitting and PDE benchmarks' as a general finding. This is an extrapolation beyond what the draft and sources establish. The draft's own context mentions 'level-set equation,' 'Stefan-type interface problems,' and 'phase transition problems' — not a systematic range of benchmarks. Output B's claim about benchmark breadth is therefore unsupported.

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.