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!

A coisotropic embedding theorem for pre-multisymplectic manifolds

We prove a coisotropice embedding theorem à là Gotay for pre-multisymplectic manifolds

Useful Fields:

This article addresses a specialized topic in differential geometry and symplectic geometry, introducing new theoretical advancements through the coisotropic embedding theorem for pre-multisymplectic manifolds. The novelty is noteworthy as it extends existing frameworks, potentially influencing future research on the geometric foundations of physics and symplectic structures. However, its applicability may be limited given its specialized nature.

Pseudoreflections on Prym Varieties

We show that for every g greater or equal than 5, the locus of Prym varieties in the moduli space of principally polarized abelian varieties of dimension g-1 that possess a pseudoreflection of geometr...

Useful Fields:

The article introduces new results concerning the structure of Prym varieties in relation to moduli spaces and pseudoreflections. The focus on dimensions greater than 5 provides novelty and addresses a gap in the understanding of geometric origins in this context. Additionally, it contrasts important properties of Prym and Jacobian varieties, thereby potentially advancing the study of algebraic geometry, particularly in moduli theory. The methodological rigor in explicitly identifying irreducible families adds to its significance.

Verb Mirage: Unveiling and Assessing Verb Concept Hallucinations in Multimodal Large Language Models

Multimodal Large Language Models (MLLMs) have garnered significant attention recently and demonstrate outstanding capabilities in various tasks such as OCR, VQA, captioning, etc\textit{etc}. Ho...

Useful Fields:

This article presents a novel exploration of verb hallucination in MLLMs, addressing a significant gap in current research on hallucinations, primarily focusing on object concepts. Its unique approach and the proposal of a new method ensure its potential impact on both theoretical understanding and practical applications of MLLMs. The robustness of the experimentation and public availability of resources enhances the article's value for future research.

Solving 1D Poisson problem with a Variational Quantum Linear Solver

Different hybrid quantum-classical algorithms have recently been developed as a near-term way to solve linear systems of equations on quantum devices. However, the focus has so far been mostly on the ...

Useful Fields:

The article introduces a novel approach to solving a specific class of linear systems using a variational quantum linear solver (VQLS), which is particularly relevant given the current interest in applying quantum algorithms to practical problems. The methodological rigor displayed in detailing the decomposition and the potential tradeoffs enhances the article's impact. Its combination of theoretical advancement and empirical results on real hardware indicates a strong applicability to quantum computing contexts, while its focus on real-world problems broadens its relevance.

Who Speaks Next? Multi-party AI Discussion Leveraging the Systematics of Turn-taking in Murder Mystery Games

Multi-agent systems utilizing large language models (LLMs) have shown great promise in achieving natural dialogue. However, smooth dialogue control and autonomous decision making among agents still re...

Useful Fields:

This article presents a novel framework that significantly contributes to the field of AI dialogue systems by addressing the challenges of multi-agent conversation dynamics. It utilizes established conversational norms, making the research both innovative and applicable to existing systems. The empirical evaluation, incorporating both automated and human assessments, is robust and lends credibility to the findings, thus showcasing potential for advancement in AI interaction.

Probing the contents of semantic representations from text, behavior, and brain data using the psychNorms metabase

Semantic representations are integral to natural language processing, psycholinguistics, and artificial intelligence. Although often derived from internet text, recent years have seen a rise in the po...

Useful Fields:

The article introduces a systematic evaluation of diverse sources of semantic representation, which is novel and highly relevant to emerging trends in natural language processing and psycholinguistics. Its methodological rigor, particularly the use of representational similarity analysis and a comprehensive metabase, enhances the credibility of findings. The implications for aligning large language models with human semantics make it influential for future AI developments.

Uncertainty-aware retinal layer segmentation in OCT through probabilistic signed distance functions

In this paper, we present a new approach for uncertainty-aware retinal layer segmentation in Optical Coherence Tomography (OCT) scans using probabilistic signed distance functions (SDF). Traditional p...

Useful Fields:

The article presents a novel approach to retinal layer segmentation using probabilistic signed distance functions, addressing key challenges in the field of Optical Coherence Tomography (OCT). The incorporation of uncertainty modeling enhances the robustness of segmentation, which is critical for clinical applications. The methodology's rigorous evaluation against various noise conditions demonstrates both practical applicability and adaptability, making this work likely to influence future research in medical imaging and segmentation methodologies.

Phonon-assisted control of magnonic and electronic band splitting

We demonstrate theoretically the ability to control non-relativistic magnonic and electronic spin splitting by manipulating phonon modes. Using MnF2_2 as a representative material, exhibiting...

Useful Fields:

This article highlights a novel theoretical approach to manipulating magnonic and electronic properties through phonon control, which is a relatively unexplored area. The rigor of the theoretical framework and the choice of MnF$_2$ as a case study contribute significantly to its impact and potential applicability in future research. This work could inspire experimental validations and new applications in spintronics, which amplifies its relevance.

Factorisability of the Harer-Zagier Transform of the HOMFLY-PT polynomial

The Harer-Zagier (HZ) transform maps the HOMFLY-PT polynomial into a rational function. For some special knots and links, the latter has a simple factorised form, both in the numerator and denominator...

Useful Fields:

This article provides a novel insight into the factorisability of the Harer-Zagier transform, connecting known polynomial invariants with geometric interpretations and topological string theory. The robustness of its conjectures, supported by proofs in special cases, strengthens its impact, creating potential pathways for future research in knot theory and related fields.

Trickle groups

A new family of groups, called trickle groups, is presented. These groups generalize right-angled Artin and Coxeter groups, as well as cactus groups. A trickle group is defined by a presentation with ...

Useful Fields:

This article introduces the novel concept of trickle groups, which expands upon established groups such as Artin and Coxeter groups, showcasing potential for broad application and further research. The development of a rewriting system for normal forms adds methodological rigor, while the exploration of parabolic subgroups introduces significant depth. The implications for existing groups suggest a meaningful impact on group theory and broader mathematical contexts.

DEYOLO: Dual-Feature-Enhancement YOLO for Cross-Modality Object Detection

Object detection in poor-illumination environments is a challenging task as objects are usually not clearly visible in RGB images. As infrared images provide additional clear edge information that com...

Useful Fields:

The article presents a novel approach to cross-modality object detection by introducing DEYOLO, which effectively combines RGB and infrared image data. The methodological rigor is evident in the dual-enhancement mechanism and the attention to reducing interference between modalities, which are both innovative aspects that could significantly improve detection accuracy in challenging environments. The extensive experimental validation further substantiates the claims of superior performance over existing state-of-the-art methods, indicating strong applicability in real-world scenarios.

Hahn series and Mahler equations: Algorithmic aspects

Many articles have recently been devoted to Mahler equations, partly because of their links with other branches of mathematics such as automata theory. Hahn series (a generalization of the Puiseux ser...

Useful Fields:

This article addresses a fundamental question in the theory of Mahler equations, exploring the algorithmic aspects of Hahn series, which is a significant advancement in the field. The novelty lies in providing a positive answer to the existence of an algorithm, which can inspire future research in both theoretical mathematics and computational approaches. The rigorous methodology in constructing a computable well-ordered receptacle further enhances its impact by presenting practical implications for solving complex equations. However, while significant, the potential applications might be more niche, primarily focusing on abstract mathematical theories rather than broad interdisciplinary appeal.

Spin and Orbital Rashba effects at the Ni/HfO2_2 interface

We predict the giant ferroelectric control of interfacial properties of Ni/HfO2, namely, (i) the magnetocrystalline anisotropy and (ii) the inverse spin and orbital Rashba effects. The reversible cont...

Useful Fields:

The article presents a novel approach to manipulating interfacial properties at the Ni/HfO2 interface through ferroelectric control, highlighting significant implications for spintronics and low-energy consumption magnetic devices. The combination of ab initio simulations and transport calculations demonstrates methodological rigor, while the discovery of substantial modulation in magnetocrystalline anisotropy and spin-density represents an exciting advancement in the field of multiferroics. Such findings are highly relevant for future applications in memory and logic gate technologies, making this article impactful for both theoretical and practical advancements.

Multifractality and polygonal vortex filaments

In this proceedings article we survey the results in [5] and their motivation, as presented at the 50th Journées EDP 2024. With the aim of quantifying turbulent behaviors of vortex filaments, we study...

Useful Fields:

This article presents novel methodologies to study multifractality in vortex filament dynamics, a significant aspect in turbulence research. Its connection to advanced mathematical concepts such as generalized Riemann functions and Diophantine sets implies a high level of methodological rigor. The focus on polygonal vortex filaments could lead to new insights in both theoretical understanding and practical applications in fluid dynamics.

S3S^3: Synonymous Semantic Space for Improving Zero-Shot Generalization of Vision-Language Models

Recently, many studies have been conducted to enhance the zero-shot generalization ability of vision-language models (e.g., CLIP) by addressing the semantic misalignment between image and text embeddi...

Useful Fields:

The article introduces a novel approach (S^3) that addresses a significant gap in current methodologies related to zero-shot learning in vision-language models, making it a potentially transformational contribution. Its methodological rigor, through extensive experiments across multiple benchmarks, supports the robustness of the findings, enhancing the credibility of the proposed solution. Improved stability in semantic alignment can lead to better real-world applications of vision-language models, influencing future designs and strategies in this area.

Follow the money: a startup-based measure of AI exposure across occupations, industries and regions

The integration of artificial intelligence (AI) into the workplace is advancing rapidly, necessitating robust metrics to evaluate its tangible impact on the labour market. Existing measures of AI occu...

Useful Fields:

The article introduces a novel and practical metric (AISE index) for evaluating AI exposure in the labor market. Its focus on actual adoption by startups, rather than theoretical potential, represents a significant advancement in understanding AI's impact on various occupations. The methodological rigor in linking occupational descriptions with startup applications strengthens its utility, while the insights challenge existing paradigms about AI risk for high-skilled jobs. This work is timely, relevant, and provides actionable guidance for policymakers, making it highly impactful in the field.

HyperGraphOS: A Meta Operating System for Science and Engineering

This paper presents HyperGraphOS, an innovative Operating System designed for the scientific and engineering domains. It combines model based engineering, graph modeling, data containers, and computat...

Useful Fields:

HyperGraphOS showcases significant innovation in operating systems tailored for scientific and engineering applications. Its unique combination of model-based engineering, graph modeling, and data containers enhances flexibility and usability, pushing the boundaries of traditional OS functionality. The methodological rigor is demonstrated through extensive evaluation across diverse domains, which strengthens its applicability and potential for future research. Its architecture and integration with AI further elevate its relevance in modern technological contexts.

Large Language Models for Ingredient Substitution in Food Recipes using Supervised Fine-tuning and Direct Preference Optimization

In this paper, we address the challenge of recipe personalization through ingredient substitution. We make use of Large Language Models (LLMs) to build an ingredient substitution system designed to pr...

Useful Fields:

This paper presents a novel application of Large Language Models (LLMs) in the context of ingredient substitution, which is a relatively unexplored area. The extensive experimentation with various fine-tuning techniques and the inclusion of Direct Preference Optimization add methodological rigor, demonstrating a thoughtful approach to improve the performance of the LLMs. The results are compelling, suggesting practical implications for personalized culinary applications, which could inspire further research in both AI and culinary arts.

Uniform characterisation of an ensemble of main-sequence benchmark stars: effect of Gaia-based data on grid search models

The inference of stellar parameters (such as radius and mass) through asteroseismic forward modelling depends on the number, accuracy, and precision of seismic and atmospheric constraints. ESA's G...

Useful Fields:

The article presents novel insights on the impact of Gaia-based data on the inference of stellar parameters, highlighting methodological advancements that enhance precision in astrophysical modeling. The incorporation of robust statistical analyses and the assessment of various constraints strengthen the work's significance in the field of astrophysics. The study’s findings have broad implications for the accuracy of stellar modeling, which is critical for understanding stellar evolution and population.

Predicting the Curie temperature in substitutionally disordered alloys using a first-principles based model

When exploring new magnetic materials, the effect of alloying plays a crucial role for numerous properties. By altering the alloy composition, it is possible to tailor, e.g., the Curie temperature (&#...

Useful Fields:

The article presents a method grounded in first-principles calculations that significantly enhances the predictability of Curie temperatures in various alloys. The methodological rigor is high, employing density functional theory and considering multiple parameters relevant to magnetism, which adds robustness. The novelty lies in its comprehensive applicability to known and unexplored systems, making it a valuable tool for researchers in magnetism and materials science. Its potential to inspire future experimental work further amplifies its relevance.