research note
BAMIFun addresses a well-known but under-solved problem in functional data analysis: when trajectories are sparsely or irregularly observed (common in wearable device, microbiome, and neuroimaging …
research note
This paper presents the first comprehensive large-scale analysis of the video piracy ecosystem on Telegram, focusing on structural, operational, and behavioral patterns of 1,057 piracy-related chan…
research note
CA-SQL addresses a specific and well-characterized failure mode in LLM-based Text-to-SQL pipelines: on the hardest queries in the BIRD benchmark, existing inference-time methods fail not because th…
research note
CUTS-GPR addresses the long-standing dual bottleneck of Gaussian process regression (GPR): the cubic cost in the number of training points N and the exponential cost in dimensionality D
research note
This paper tackles the core training bottleneck of Spiking Neural Networks (SNNs): the non-differentiability of the spike (Heaviside) function forces practitioners to use surrogate gradients, which…
research note
GRAPHLCP addresses the problem of uncertainty quantification for graph neural networks (GNNs) using conformal prediction (CP)
research note
This paper addresses a gap in the pinching antenna systems (PASS) literature: virtually all prior work assumes users are stationary or quasi-static, yet the optimal placement of pinching points alo…
research note
This paper addresses the challenge of controlling and detecting 'hidden' multipolar orders — specifically quadrupolar and octupolar states — in spin-orbit-coupled Mott insulators with d2 electronic…
research note
Mazocarta is not primarily a new deckbuilder mechanic; it is an engineering artifact meant to prove that a procedural game can be built so the exact same rules core drives play, simulation, tests, …
research note
This paper studies the matching complexity of regular expressions with backreferences (REWBs) through the lens of fine-grained complexity, with an eye toward ReDoS-safe regex engines
research note
PET-Adapter addresses a core practical bottleneck in deep-learning-based PET image reconstruction: models trained on simulation or phantom data degrade substantially when applied to real clinical d…
research note
QUANTAS 2 is a distributed-algorithm simulator aimed at closing a practical gap in the original QUANTAS: it keeps the fast, round-based abstract model but adds a concrete socket-execution mode and …
