Research works

Optimizing 3D Scene Reconstruction with Differentiable Ray Tracing

A fundamental limitation of classical rendering pipelines is that their discrete, non-differentiable structure makes them unsuitable for gradient-based parameter recovery. This paper presents a differentiable ray tracing framework that directly addresses this gap, enabling end-to...

3-2-1: Lift Off! Co-Planning Strategies When “There’s No Time!”

Collaborative planning is both a High-Leverage Practice and evidence-based practice for inclusive education. However, co-teachers frequently cite a lack of time as the greatest barrier to effective collaboration and instructional impact. This article presents a practical, flexibl...

Estimating Spatial and Temporal Patterns of Residential Power Outages in Massachusetts from 2013 to 2022 Using Public Records

Abstract Power outages are a growing threat to human health. Extreme weather events and strains on the electrical grid can cut off access to critical health-supporting medical equipment. However, the full scope of health impacts related to power outages has not been quantified, i...

Risk Theory in a Finite Customer-Pool Setting

This paper investigates an insurance model with a finite number of major clients and a large number of small clients, where the dynamics of the latter group are modeled by a spectrally positive Levy process. We begin by analyzing this general model, in which the inter-arrival tim...

Nickel-catalyzed enantioselective reductive vinyl-alkynylation of unactivated alkenes with bromoalkynes

Alkynyl groups are highly valuable functionalities, prompting chemists to continually seek efficient alkynylation reactions. Traditional asymmetric alkynylation strategies using terminal alkynes often rely on complex chiral phosphine ligands, stoichiometric amounts of silver reag...

From lectures to learning outcomes: meaningful integration of AI-generated content in pre-clerkship medical training

Abstract Background Large Language Models (LLMs) can efficiently synthesize educational content, yet few studies have evaluated standardized, LLM-powered curricular interventions and their effects on medical student learning. Methods In this single-institution study, lecture-spec...

The Impacts of Zonal and Meridional Winds on the Observed Occurrence Rate of Equatorial Plasma Bubbles

Abstract Post‐sunset equatorial plasma bubble (EPB) formation is believe to be influenced by factors including thermospheric neutral winds. This study characterizes pre‐sunset neutral winds observed by ICON on EPB occurrences observed by GOLD over the east Atlantic. 3 years of da...

First Sounds, Second Language: How Word Onsets Spark Language Co‐Activation in 3‐Year‐Old Bilinguals

ABSTRACT Adult and young bilinguals co‐activate their languages to different degrees, even in entirely monolingual tasks. However, many studies use translation pairs with high phonological overlap (e.g., the English‐GERMAN: “milk”‐“MILCH”), making it difficult to discern the orig...

A Survey on Clustered Federated Learning: Taxonomy, Analysis and Applications

As Federated Learning (FL) expands, the challenge of non-independent and identically distributed (non-IID) data becomes critical. Clustered Federated Learning (CFL) addresses this by training multiple specialized models, each representing a group of clients with similar data dist...

Cellular automata can really solve the parity problem

Abstract Determining properties of an arbitrary binary sequence is a challenging task if only local processing is allowed. Among these properties, the determination of the parity of 1s by distributed consensus has been a recurring endeavour in the context of automata networks. In...

Cycling speed and weather: Roles of cyclist weather-sensitivity, spatial and infrastructural conditions

Weather significantly influences cycling speed, and this influence increases with climate change. It is related to cycling attractiveness, but little evidence exists. This study examines this influence, considering the self-selection bias that some people avoid cycling during bad...

Multi-Bin Batching for Increasing LLM Inference Throughput

As large language models (LLMs) grow in popularity for their diverse capabilities, improving the efficiency of their inference systems has become increasingly critical. Batching LLM requests is a critical step in scheduling the inference jobs on servers (e.g. GPUs), enabling the...

The Moduli of the Universal Geometry of Heterotic Moduli

Abstract We study the moduli of the universal geometry of $$d=4$$ d = 4 $$N=1$$ N = 1 heterotic vacua. Universal geometry refers to a family of heterotic vacua fibred over the moduli space. The universal geometry mimics aspects of the original heterotic vacua, in particular holom...

A Novel polymer planar biomimetic membrane with self-differentiating intestinal-like structural for drug permeability prediction

Accurately and efficiently assessing the drug permeability are emerging as the big challenge early in new drug discovery and development, thus various models have been developed. Here, a polymer planar biomimetic membrane permeation assay (PPBMPA) model was constructed in a trans...

Disordered Gibbs measures and Gaussian conditioning

Abstract We study the law of a random field $$f_N(\varvec{\sigma })$$ f N ( σ ) evaluated at a random sample from the Gibbs measure associated to a Gaussian field $$H_N(\varvec{\sigma })$$ H N ( σ ) . In the high-temperature regime, we show that bounds on the probability that $$f...

Koma Density Geometry ── Connection Conditions between Density-Derived Informational Structure and Effective Geometry

This paper organizes the mathematical scope for extracting informational structure from the Koma density kernel $\rho$ and connecting it to an effective geometry. We introduce a Fisher-information-type functional for the density kernel, and consider its local gradient structure $...

Location-Mixed Flea Profiles and Born-State Convergence: Robustness of the Two-Scale Classification in the Two-State Flea Model

We study the robustness of the two-scale Born-state convergence classification in the exact two-state flea model when the fixed scaled source profile is replaced by genuinely h-dependent location mixtures. For a fixed normalized kernel k in L1(R), we consider profiles g_h = k * n...

The Dualbind Gravity Model ── Curvature Emergence from Information Placement Density (Koma Theory)

This paper presents the Koma Gravity Model — the gravitational branch of Koma Theory — as a structural hypothesis for the problem domain of emergent gravity. In this model, the spatial placement density of quantum information units is expressed as a dimensionless information fiel...

The Dualbind Gravity Model ── Curvature Emergence from Information Placement Density (Koma Theory)

This paper presents the Koma Gravity Model — the gravitational branch of Koma Theory — as a structural hypothesis for the problem domain of emergent gravity. In this model, the spatial placement density of quantum information units is expressed as a dimensionless information fiel...

A7U: Distributed Planck Bundles, Access-Relative Purity, and the Matter-Wave Visibility Bridge

When a pure object is only pure for a closed access cut The completed distributed bundle closes one source budget: \[ \boxed{Q_w^{\mathrm{vis}}+Q_w^{\mathrm{hid}}=2\pi.} \] Version: 4.1 Concept DOI: 10.5281/zenodo.20097247 Author: Ali Attar, ORCID 0009-0008-9931-2691 Website: qua...