Research & Innovation

Global: New Open‑Source Toolkit SciEvalKit Aims to Standardize AI Evaluation Across Scientific DisciplinesResearchers have released SciEvalKit, an open‑source benchmarking toolkit designed to evaluate artificial‑intelligence models for scientific tasks. The preprint…

Global: Study Analyzes Feedback Complexity in Sparse Dictionary LearningResearchers led by Akash Kumar released a study on February 8, 2025 that examines how sparse superposed features can be recovered through…

Global: Deep Learning Boosts Valuation Accuracy for First‑Time Art SalesA study posted to arXiv on December 28, 2025, by Jianping Mei, Michael Moses, Jan Waelty, and Yucheng Yang investigates how…

Global: New Transformer Model Enhances Fashion Item DetectionResearchers have introduced the Holistic Detection Transformer (Holi-DETR), a model designed to identify fashion items within outfit images by exploiting multiple layers of…

Global: Alpha-R1 Reasoning Model Improves Factor Screening in Dynamic MarketsResearchers affiliated with FinStep AI introduced Alpha‑R1, an 8‑billion‑parameter reasoning model designed to evaluate the relevance of quantitative investment factors amid…

Global: Convergence Guarantees for Asynchronous Pipeline Training on Analog In‑Memory AcceleratorsResearchers who authored the arXiv preprint (arXiv:2410.15155v2) have presented a theoretical analysis of stochastic gradient descent (SGD) applied to large…

Global: Meta-Learning Enables Precise Geometry Prediction in Laser-Directed Energy DepositionBackgroundResearchers have identified that predicting bead geometry in laser-directed energy deposition (L‑DED) is complicated by limited and diverse experimental data, which…

Global: New Framework Merges Contextual and Causal Bayesian Optimization On Jan 29 2023 a team of researchers led by Vahan Arsenyan published a preprint on arXiv that proposes a unified…

Global: New Framework Reveals Instability in Model Explanations Across Training RunsResearchers posted a study on arXiv in December 2025 that introduces a diagnostic tool designed to assess whether explanations generated…

Global: New Benchmark Assesses Uncertainty Techniques for Domain-Agnostic Image SegmentationA recent preprint posted on arXiv on December 29, 2025, introduces a benchmark called UncertSAM that evaluates how post‑hoc uncertainty estimation…