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!

TeamCraft: A Benchmark for Multi-Modal Multi-Agent Systems in Minecraft

Collaboration is a cornerstone of society. In the real world, human teammates make use of multi-sensory data to tackle challenging tasks in ever-changing environments. It is essential for embodied age...

Useful Fields:

The article presents a novel benchmark, TeamCraft, that addresses the critical gap in evaluating multi-modal multi-agent systems in complex, dynamic environments like Minecraft. The use of a widely recognized platform along with extensive task variants for training and evaluation marks significant contributions to both the benchmarking of AI models and the understanding of their limitations. The robustness of the methodology and the potential applications highlight its importance for future research and development in this area.

Preprocessing is All You Need: Boosting the Performance of Log Parsers With a General Preprocessing Framework

Log parsing has been a long-studied area in software engineering due to its importance in identifying dynamic variables and constructing log templates. Prior work has proposed many statistic-based log...

Useful Fields:

The article introduces a novel approach to log parsing by emphasizing the importance of preprocessing, which has been traditionally undervalued in this area. The methodological rigor displayed through the development of a general preprocessing framework and its empirical validation presents it as a significant contribution. Its enhanced performance metrics for existing parsers demonstrate potential practical applications that can inspire future research and development in log parsing methodologies.

Non-linear Transport in Non-centrosymmetric Systems: From Fundamentals to Applications

Ohm's law has been a cornerstone of electronics since its experimental discovery. This law establishes that in a conductive system, the voltage is directly proportional to the current. Even when t...

Useful Fields:

This article is highly relevant due to its exploration of a significant shift in traditional understanding of transport phenomena in conductive systems, particularly highlighting non-linear effects in non-centrosymmetric materials. The investigation into the connection between system symmetry and non-linear behavior adds novelty and depth to the existing literature. The potential applications in spintronics and energy harvesting further emphasize its applicability and significance.

From classical techniques to convolution-based models: A review of object detection algorithms

Object detection is a fundamental task in computer vision and image understanding, with the goal of identifying and localizing objects of interest within an image while assigning them corresponding cl...

Useful Fields:

The article provides a thorough review of the evolution of object detection from traditional methods to deep learning approaches, making it relevant for academics and practitioners in computer vision. Its critical comparison of models highlights both strengths and limitations, contributing to the discourse on future research directions.

Uncertainty Quantification for Transformer Models for Dark-Pattern Detection

The opaque nature of transformer-based models, particularly in applications susceptible to unethical practices such as dark-patterns in user interfaces, requires models that integrate uncertainty quan...

Useful Fields:

This article assesses a significant gap in the use of transformer models, focusing on the increasingly relevant problem of unethical design practices (dark patterns) in user interfaces. The integration of uncertainty quantification adds both methodological rigor and practical applicability to the field, enhancing trust and explainability in machine learning applications. The study not only proposes innovative techniques but also validates them comprehensively, which contributes to advancing the understanding of responsible AI practices.

Constructing projective modules

We discuss elements of a social history of the theory of projective modules over commutative rings. We attempt to study the question: how did the theory of projective modules become one of "mains...

Useful Fields:

This article offers a unique historical perspective on the development and rise of projective modules, which is often neglected in more technical mathematical discussions. Its emphasis on social and cultural factors in mathematics adds novelty and interconnectivity to the study of projective modules, making it a significant contribution. The rigorous analysis of various mathematical strands and their influences on projective modules showcases methodological strength. Nonetheless, its historical focus may have limited applicability in more modern, application-driven mathematical research.

An Information Theoretic Analysis of Ghost Modulation

Side channels have become an essential component of many modern information-theoretic schemes. The emerging field of cross technology communications (CTC) provides practical methods for creating inten...

Useful Fields:

The paper presents a novel theoretical framework for Ghost Modulation, a significant advancement in cross technology communications. The development of a cost-effective solution that utilizes existing hardware is particularly noteworthy, as it has strong practical implications. The methodological rigor is emphasized through mathematical modeling and simulation results, which bolster the validity of the proposed techniques. The novelty of addressing the challenges associated with asymmetric binary crossover erasure channels further enhances the paper's relevance.

Occupation time statistics for non-Markovian random walks

We study the occupation time statistics for non-Markovian random walkers based on the formalism of the generalized master equation for the Continuous-Time Random Walk. We also explore the case when th...

Useful Fields:

The article presents a thorough investigation into occupation time statistics within non-Markovian random walks, leveraging generalized master equations and stochastic resetting. The analysis of PDFs, moments, and limiting distributions showcases methodological rigor and provides directed insights that could advance theoretical frameworks in statistical physics and related fields. The revisiting of the arcsine law adds a novel dimension to the study of random processes, indicating potential implications for future explorations into resetting dynamics and anomalous diffusion.

On the $AdS_3$ Virasoro-Shapiro Amplitude

We consider tree-level scattering amplitudes for four string tachyons on $AdS_3 \times {\cal N}$ with pure NSNS fluxes. We show that in a small curvature expansion, properly defined, the ampli...

Useful Fields:

The article presents a novel analysis of tree-level scattering amplitudes in the context of string theory on AdS spaces, bridging elements of higher-dimensional theories with a specific focus on AdS_3. The use of multiple polylogarithms presents a new mathematical framework for understanding these amplitudes, which could generate significant interest and further research in the field of string theory and mathematical physics. The methodological rigor in defining the amplitudes is commendable, although the target audience may be limited to a niche within theoretical physics.

Bayesian quantum estimation of the separation of two incoherent point sources

We address the estimation problem of the separation of two arbitrarily close incoherent point sources from the quantum Bayesian point of view, i.e., when a prior probability distribution function (PDF...

Useful Fields:

The article explores a novel quantum-based estimation problem, which adds significant value to quantum metrology and imaging techniques. The rigorous comparison of methods within the Bayesian framework enhances methodological rigor and opens new avenues for practical applications. Its innovative approach to estimation with prior information could lead to significant advancements in imaging precision.

Enhancing Foundation Models for Time Series Forecasting via Wavelet-based Tokenization

How to best develop foundational models for time series forecasting remains an important open question. Tokenization is a crucial consideration in this effort: what is an effective discrete vocabulary...

Useful Fields:

This article presents an innovative approach to tokenization specifically designed for time series forecasting, addressing a key challenge in the development of foundational models. The wavelet-based tokenizer (WaveToken) not only enhances model learning in time series contexts but also demonstrates empirical superiority over existing methods. Its rigorous evaluation across a diverse set of 42 datasets boosts its credibility and relevance. The potential applicability to various forecasting scenarios increases its significance for future research.

CompCap: Improving Multimodal Large Language Models with Composite Captions

How well can Multimodal Large Language Models (MLLMs) understand composite images? Composite images (CIs) are synthetic visuals created by merging multiple visual elements, such as charts, posters, or...

Useful Fields:

This article presents a novel approach to enhancing Multimodal Large Language Models (MLLMs) specifically for understanding composite images, a growing area of interest that has been underexplored compared to natural images. The introduction of Composite Captions (CompCap) and the creation of a sizeable dataset (CompCap-118K) are significant contributions that address critical gaps in current research. The empirical validation through fine-tuning MLLMs is robust, adding credibility to the findings. Overall, the novelty of the dataset and methodology, combined with the empirical results, demonstrates strong potential for advancing the field.

Lack of Rest-frame UV Variability in Little Red Dots Based on HST and JWST Observations

Variability is a fundamental signature for active galactic nuclei (AGN) activity, and serve as an unbiased indicator for rapid instability happened near the center supermassive black hole (BH). Previo...

Useful Fields:

The article presents novel findings on the variability of little red dots (LRDs) in the context of AGN activity, using robust observational data from JWST and HST. The lack of variability in these high-redshift galaxies is significant as it challenges some existing models about AGN activity. The study's implications for identifying low-mass AGNs at high redshifts are particularly noteworthy, showcasing potential advancements in understanding the role of supermassive black holes in early galaxy formation.

Simultaneous identification of the parameters in the plasticity function for power hardening materials : A Bayesian approach

In this paper, we study simultaneous determination of the strain hardening exponent, the shear modulus and the yield stress in an inverse problem. First, we analyze the direct and the inverse problems...

Useful Fields:

The article presents a novel Bayesian approach to simultaneously identify parameters in plasticity functions, which is crucial for understanding and modeling material behavior in engineering. The methodological rigor in applying Bayesian inference to an inverse problem is significant, as it enhances accuracy in parameter estimation. The use of numerical examples demonstrates applicability, which is important for practical implementation. However, the impact is somewhat limited by the initial analysis only considering specific types of materials without broader validation across different scenarios.

Automated, Unsupervised, and Auto-parameterized Inference of Data Patterns and Anomaly Detection

With the advent of data-centric and machine learning (ML) systems, data quality is playing an increasingly critical role in ensuring the overall quality of software systems. Data preparation, an essen...

Useful Fields:

The article presents a novel approach to automated anomaly detection without the need for supervised learning or parameter setting. The reported F1 scores and performance improvements over existing systems demonstrate significant methodological rigor. The system's capability to work with uncleaned data adds substantial practical value, making it applicable across various domains.

Conditions for uniform in time convergence: applications to averaging, numerical discretisations and mean-field systems

We establish general conditions under which there exists uniform in time convergence between a stochastic process and its approximated system. These standardised conditions consist of a local in time ...

Useful Fields:

This article presents novel and general conditions for uniform in time convergence across stochastic processes and their approximations, addressing a relevant challenge in mathematical analysis and numerical methods. Its methodological rigor and broad applicability make it potentially impactful, especially in advancing related numerical techniques and theoretical frameworks in various fields.

Some splitting and rigidity results for sub-static spaces

In this paper we study the rigidity problem for sub-static systems with possibly non-empty boundary. First, we get local and global splitting theorems by assuming the existence of suitable compact min...

Useful Fields:

The paper addresses important aspects of rigidity and splitting theorems in differential geometry, specifically relating to sub-static spaces, which are less explored areas. The methodological approach appears to be rigorous, and the extension of existing results provides valuable advancements. The implications of boundary integral inequalities and connections to Einstein field equations are particularly noteworthy, as these are central to theoretical physics and may inspire further interdisciplinary research.

MAmmoTH-VL: Eliciting Multimodal Reasoning with Instruction Tuning at Scale

Open-source multimodal large language models (MLLMs) have shown significant potential in a broad range of multimodal tasks. However, their reasoning capabilities remain constrained by existing instruc...

Useful Fields:

The article presents a novel approach to enhancing the reasoning capabilities of multimodal large language models through the construction of a large-scale instruction-tuning dataset. Its rigorous methodology, significant improvements in performance benchmarks, and the focus on detailed rationales mark it as a strong contribution to the field. The findings could inspire future research on dataset construction and multimodal AI applications, increasing its relevance.

Field Sources for Wormholes With Multiple Throats/Anti-throats

In this work, we studied the characteristics of wormholes with multiple throats/anti-throats in the context of general relativity. The presence of these structures is verified through the minima and m...

Useful Fields:

This article presents a novel exploration of wormholes with multiple throats/anti-throats, which is relatively under-explored in the literature. The rigorous application of general relativity to analyze these exotic structures, along with the use of embedding diagrams, adds methodological depth to the research. Importantly, the findings on energy conditions and scalar field interactions provide valuable insights into the stability and characteristics of such spacetimes. This work has strong implications for theoretical physics and cosmology, particularly in studying the fundamental aspects of spacetime and potential implications for traversable wormholes, enhancing its relevance for future research.

Penetrative rotating magnetoconvection subject to lateral variations in temperature gradients

Convection-driven flows in planetary interiors exhibit rich dynamics owing to multiple spatio-temporally varying forcing conditions and physical constraints. In particular, the churning of liquid meta...

Useful Fields:

The article offers significant insights into penetrating rotating magnetoconvection, specifically in the context of the Earth's outer core, an area critical for understanding geomagnetic field dynamics. It presents novel findings regarding thermal heterogeneity and stratification effects, supported by robust computational and theoretical methods. The closed-form expressions developed enhance its practical applicability, promising to guide future geophysical modeling. The study's interdisciplinary implications are substantial, as it connects fluid dynamics with planetary science, making it a valuable reference for both theoretical exploration and practical applications in geophysics.