This is a experimental project. Feel free to send feedback!

Thesis Tide

Thesis Tide ranks papers based on their relevance to the fields, with the goal of making it easier to find the most relevant papers. It uses AI to analyze the content of papers and rank them!

Music and art: a study in cross-modal interpretation

Our study has investigated the effect of music on the experience of viewing art, investigating the factors which create a sense of connectivity between the two forms. We worked with 138 participants, ...

Useful Fields:

The study presents a novel investigation into the interplay between music and art, focusing on emotional responses and user experience. This interdisciplinary approach adds value to both fields, though the methods could benefit from more rigorous quantitative analysis. The proposal of guidelines and future research directions enhances its applicability and potential impact.

Variable Goal Approach (VGA) Enhancing Pedestrian Dynamics Modeling

In the existing literature, pedestrian dynamics models have successfully captured various complex scenarios such as lane formation, evacuation, bottlenecks, crowd intersections, etc. However, many mod...

Useful Fields:

The Variable Goal Approach represents a substantial advancement in pedestrian dynamics modeling by addressing previous limitations with existing models. Its introduction of human intelligence and stochasticity introduces significant novelty and practical applicability in various pedestrian scenarios, enhancing both the realism and predictive accuracy of simulations. The methodological rigor presented in the study, along with the potential for practical applications in urban planning and safety, contributes to its high relevance score.

Recovery of activation propagation and self-sustained oscillation abilities in stroke brain networks

Healthy brain networks usually show highly efficient information communication and self-sustained oscillation abilities. However, how the brain network structure affects these dynamics after an injury...

Useful Fields:

This article provides novel insights into how brain network dynamics are altered following a stroke, emphasizing the recovery potential over time. Its use of a large dataset and the employment of a reaction-diffusion model contribute to its methodological rigor. Additionally, the findings have significant implications for understanding brain recovery, potentially informing therapeutic strategies and rehabilitation approaches. However, while the study addresses important issues, its focus is somewhat narrow, centered primarily on stroke, which limits broader applicability.

Motion-X++: A Large-Scale Multimodal 3D Whole-body Human Motion Dataset

In this paper, we introduce Motion-X++, a large-scale multimodal 3D expressive whole-body human motion dataset. Existing motion datasets predominantly capture body-only poses, lacking facial expressio...

Useful Fields:

Motion-X++ presents a novel and extensive dataset that addresses the limitations of existing motion datasets by incorporating facial expressions, hand gestures, and fine-grained pose descriptions in a large-scale format. The methodological rigor shown in the development of a scalable annotation pipeline adds significant value, making the dataset an important resource for downstream applications such as animation, robotics, and human-computer interaction. Its multifaceted nature encourages interdisciplinary research and development.

A 1Mb mixed-precision quantized encoder for image classification and patch-based compression

Even if Application-Specific Integrated Circuits (ASIC) have proven to be a relevant choice for integrating inference at the edge, they are often limited in terms of applicability. In this paper, we d...

Useful Fields:

The study presents a novel ASIC neural network encoder that integrates multiple tasks (classification and compression) with a low resource footprint. The mixed-precision quantization, adaptive scaling, and structural pruning introduce innovative methodologies that offer significant advancements in the efficiency and effectiveness of image processing systems at the edge. The practical implications for real-world applications, especially in resource-constrained environments, enhance the relevance of this research.

Solutions to Problems in Amer. Math. Monthly, Math. Magazine, College Math. J., Elemente der Math.,Crux Math., EMS Newsletter Math. Gazette

In this arxiv-post I present my solutions (published or not) to Problems that appeared in Amer. Math. Monthly, Math. Magazine, Elemente der Mathematik and CRUX, that were mostly done in collaboration ...

Useful Fields:

The article presents solutions to mathematical problems published in various journals, indicating collaborative efforts and possibly methodological insights. However, the impact is limited as it primarily compiles existing work rather than introducing new concepts or methodologies. Its utility depends on the originality of the solutions and their applicability to broader mathematical discourse.

Advancing ALS Applications with Large-Scale Pre-training: Dataset Development and Downstream Assessment

The pre-training and fine-tuning paradigm has revolutionized satellite remote sensing applications. However, this approach remains largely underexplored for airborne laser scanning (ALS), an important...

Useful Fields:

The study presents a significant advancement in the utilization of large-scale pre-training for airborne laser scanning (ALS), a field with extensive applications. The construction of a diverse dataset and the innovative geospatial sampling method are both noteworthy contributions that advance methodological rigor in this area. Furthermore, demonstrating clear performance improvements in downstream tasks through a well-optimized pre-training strategy showcases both novelty and applicability, which strengthens its potential impact on future research.

Convexity of Mutual Information along the Fokker-Planck Flow

We conduct a preliminary study of the convexity of mutual information regarded as the function of time along the Fokker-Planck equation and generalize conclusions in the cases of heat flow and OU flow...

Useful Fields:

The article presents a novel approach to studying the convexity of mutual information in the context of Fokker-Planck equations, which is a significant mathematical contribution with implications in statistical mechanics and information theory. The proof of existence and uniqueness of classical solutions adds methodological rigor, enhancing the credibility of the findings. Additionally, the results regarding conditions under which mutual information maintains its properties under evolution are likely to inspire future research in related fields.

Hierarchical Decomposed Dual-domain Deep Learning for Sparse-View CT Reconstruction

Objective: X-ray computed tomography employing sparse projection views has emerged as a contemporary technique to mitigate radiation dose. However, due to the inadequate number of projection views, an...

Useful Fields:

The article presents a novel and theoretically justified approach to a significant problem in medical imaging - sparse-view CT reconstruction. The integration of deep learning with advanced mathematical concepts like deep convolutional framelets and hierarchical decomposition enhances methodological rigor. This work is particularly impactful as it addresses existing limitations in current techniques while providing a clear path for future exploration in the field.

Spin-orbit correlation and spatial distributions for spin-0 hadrons

The spin-orbit correlation in spin-0 hadrons can be investigated through the kinetic energy-momentum tensor form factor F~q(t)\tilde F^q(t). We observe that the latter is also related to a torque ...

Useful Fields:

The article presents a novel approach to understanding spin-orbit correlations in spin-0 hadrons, integrating concepts of chiral stress and kinetic energy-momentum tensors. Its linkage between these aspects and established electromagnetic form factors adds depth and can inspire further exploration in both theoretical and experimental physics. The methodological framework appears rigorous and applicable to current research in particle physics.

Migration of phthalate plasticisers in heritage objects made of poly(vinyl chloride): mechanical and environmental aspects

To clean or not to clean? The solution to this dilemma is related to understanding the plasticiser migration which has a few practical implications for the state of museum artefacts made of plasticise...

Useful Fields:

This article addresses a critical issue in heritage conservation, specifically the migration of plasticisers from PVC, which poses both aesthetic and functional risks to museum objects. The methodological rigor is demonstrated through molecular dynamics simulations and NMR diffusometry, providing a solid foundation for the proposed guidelines. The implications for conservation practice are significant, making this research not only relevant but potentially transformative for how conservators approach the cleaning of heritage items.

Supervised Learning with Evolving Tasks and Performance Guarantees

Multiple supervised learning scenarios are composed by a sequence of classification tasks. For instance, multi-task learning and continual learning aim to learn a sequence of tasks that is either fixe...

Useful Fields:

This article provides a novel approach that addresses a crucial gap in supervised learning by focusing on evolving tasks and their interdependencies. The proposed methodology combines flexibility and rigorous performance guarantees, which enhances its applicability across various contexts. The empirical validation across multiple scenarios reinforces its potential impact within the field. The level of innovation and its relevance to ongoing discussions in machine learning justify a high relevance score.

Effect of particle and substrate wettability on evaporation-driven assembly of colloidal monolayers

Assembled monolayers of colloidal particles are crucial for various applications, including opto-electronics, surface engineering, as well as light harvesting, and catalysis. A common approach for sel...

Useful Fields:

The article presents a novel numerical approach to understand the effects of wettability on colloidal monolayers, which is crucial for various advanced applications. The theoretical models proposed are well-validated through simulations, highlighting the rigor of the methodology. The exploration of how both substrate and particle wettability affect quality and pattern of monolayers fills a significant gap in existing research and has potential implications for practical applications in multiple fields.

Enhanced Quantile Regression with Spiking Neural Networks for Long-Term System Health Prognostics

This paper presents a novel predictive maintenance framework centered on Enhanced Quantile Regression Neural Networks EQRNNs, for anticipating system failures in industrial robotics. We address the ch...

Useful Fields:

The article presents a novel and advanced predictive maintenance framework that employs Enhanced Quantile Regression Neural Networks and Spiking Neural Networks, which is a significant advancement in failure prediction for industrial robotics. The high accuracy rate and substantial operational improvements noted in field tests underline its practical applicability and potential for real-world impact. The integration of these technologies in Industry 4.0 settings makes this research particularly relevant and timely, addressing contemporary challenges in manufacturing.

The impact of the redshift-dependent selection effect of halos on the redshift-space power spectrum

In a wide-area spectroscopic survey of galaxies, it is nearly impossible to obtain a homogeneous sample of galaxies with respect to galaxy properties such as stellar mass and host halo mass across a r...

Useful Fields:

The paper presents a thorough analytical examination of a significant issue (redshift-dependent selection effect) affecting the interpretation of the power spectrum in galaxy clustering analyses. The novelty of addressing biases in cosmological measurements due to selection effects adds value to the existing body of work. Methodologically, the use of N-body simulations and mock catalogs to illustrate their findings enhances credibility and robustness. However, the impact on cosmological parameter estimation appears modest (up to 2% changes), limiting the urgency of the issue compared to other potentially larger biases in cosmology.

End-to-End Deep Learning for Interior Tomography with Low-Dose X-ray CT

Objective: There exist several X-ray computed tomography (CT) scanning strategies to reduce a radiation dose, such as (1) sparse-view CT, (2) low-dose CT, and (3) region-of-interest (ROI) CT (called i...

Useful Fields:

The study introduces a novel end-to-end deep learning approach that addresses critical issues in low-dose X-ray CT, specifically focusing on the coupled artifacts of truncated projections. Its methodological rigor is enhanced by the clear decoupling of problems and the implementation of dual-domain CNNs, representing a significant advancement over traditional methods. This innovation has potential applications in clinical settings, thus underscoring its relevance and impact.

Time-Variant Vector Field Visualization for Magnetic Fields of Neutron Star Simulations

We present a novel visualization application designed to explore the time-dependent development of magnetic fields of neutron stars. The strongest magnetic fields in the universe can be found within n...

Useful Fields:

The article presents a novel visualization application specifically designed for studying time-dependent magnetic fields in neutron stars, a topic of significant interest within astrophysics. The integration of both sparse and dense vector field visualization techniques allows for interactive exploration of complex simulation data, enhancing the understanding of magnetic field behavior in neutron stars. The positive qualitative feedback from domain experts indicates strong applicability and effectiveness, suggesting the potential for substantial impact in visualization methods for astrophysical research. However, the novelty mainly lies in applied visualization rather than a breakthrough in astrophysics itself, which slightly tempers the overall impact.

Existence and multiplicity of solutions for a critical Kirchhoff type elliptic equation with a logarithmic perturbation

In this paper, we are interested in the following critical Kirchhoff type elliptic equation with a logarithmic perturbation \begin{equation}\label{eq0} \begin{cases} -\left(1+b\int_Ω|\nabla{u}|^2\math...

Useful Fields:

The article presents a novel approach to a critical Kirchhoff type elliptic equation with logarithmic perturbations, which is a significant advance in understanding such equations in higher dimensions. The study employs rigorous variational methods to establish both existence and multiplicity of solutions across different dimensions, showcasing methodological rigor. The complexity of the nonlocal and critical features adds to the novelty and potential impact on future research in related fields of mathematical analysis and PDEs.

Comparison of Feature Learning Methods for Metadata Extraction from PDF Scholarly Documents

The availability of metadata for scientific documents is pivotal in propelling scientific knowledge forward and for adhering to the FAIR principles (i.e. Findability, Accessibility, Interoperability, ...

Useful Fields:

The article's focus on improving metadata extraction from scholarly documents is highly relevant as it addresses a significant barrier to research accessibility. The use of multiple advanced methodologies (NLP, CV, multimodal) indicates a robust approach that could yield novel insights. The practical implications for improving adherence to FAIR principles enhance the article's impact.

DriVLM: Domain Adaptation of Vision-Language Models in Autonomous Driving

In recent years, large language models have had a very impressive performance, which largely contributed to the development and application of artificial intelligence, and the parameters and performan...

Useful Fields:

This paper addresses a growing area of research in multimodal large language models (MLLMs) and their applications in autonomous driving, which is a rapidly evolving field. The novelty of using small-scale MLLMs can facilitate access for researchers with limited computational resources, potentially democratizing innovation in this area. The focus on domain adaptation signifies a practical approach to integrating advanced AI techniques in real-world scenarios, although the paper could benefit from a more detailed methodological framework to ensure robustness and replicability.