ArXivCSExplorer
☆☆Bookmarks🏆RSSHow to UseFAQ
Built with and by Teycir Ben Soltane•
How to Use•FAQ•GitHub•arXiv.org•
Share:

20 results for “model width”

CS papers only

Hybrid search: Keyword + semantic, ranked by combined score.ⓘ

Want pure semantic search? Try claim verification →

cs.CLEmpiricalRecentJun 16, 2026

Variable-Width Transformers

Zhaofeng Wu, Oliver Sieberling, Shawn Tan, Rameswar Panda +2 more

This paper proposes a new architecture for transformer-based language models called 'former', which allocates capacity non-uniformly across network depth by maintaining wider early and late layers whi…

View →
cs.DScs.CCTheoreticalRecentJun 18, 2026

Computing Twin-Width via Treedepth and Vertex Integrity

Robert Ganian, Mathis Rocton

The authors develop fixed-parameter algorithms for computing or approximating twin-width under different parameterizations.

View →
cs.LGcs.AIcs.CLEmpiricalRecentJun 22, 2026

Tapered Language Models

Reza Bayat, Ali Behrouz, Aaron Courville

This paper explores the effect of non-uniform parameter allocation in language models and introduces a new architectural principle called Tapered Language Models (TLMs).

View →
cs.LGcs.AImath.NAEmpiricalRecentJun 22, 2026

Solve for the Hyperparameter, Skip the Search: Kolmogorov-Optimal Scaling Laws for Spline Regression

Yong Yi Bay, Kathleen A. Yearick

The paper presents a method, KORE, to find the optimal resolution for spline regression without exhaustive search.

View →
cs.LGmath.PRstat.MLTheoreticalRecentJul 7, 2026

Quantitative Gaussian-Process limits of Tensor Programs

Andrea Agazzi, Eloy Mosig García, Dario Trevisan

This paper provides explicit error bounds for the infinite-width Gaussian-process limit of random neural networks using tensor programs and quantitative convergence theory in Wasserstein distance.

View →
cs.CVEmpiricalRecentJun 22, 2026

Lightweight Neural Framework for Robust 3D Volume and Surface Estimation from Multi-View Images

Diego E. Farchione, Ramzi Idoughi, Peter Wonka

A new framework for accurately estimating volume and surface area from multi-view images with low computational cost.

View →
cs.ARcs.AIcs.LGEmpiricalRecentOct 27, 2026

AnalogScout: Simulation-Efficient PVT-Aware Analog Sizing with LLM Agents

Yichen Long, Keren Zhu

AnalogScout is a simulation-efficient PVT-aware sizing framework driven by a large language model agent that reduces power consumption in analog device design.

View →
cs.CVcs.AIRecentJun 1, 2026

Parameter-Efficient Fine-Tuning of Large Pretrained Models for Instance Segmentation Tasks

Nermeen Abou Baker, David Rohrschneider, Uwe Handmann

This paper investigates the application of Parameter-Efficient Fine-Tuning (PEFT) methods, specifically adapters and LoRA, to large pretrained models for instance segmentation, demonstrating that thes…

View →
cs.AIRecentMay 27, 2026

MUSE: Benchmarking Manufacturable, Functional, and Assemblable Text-to-CAD Generation

Xiaoyu Dong, Zhi Li, Xiao-Ming Wu

The paper introduces MUSE, a comprehensive benchmark that evaluates Text-to-CAD generation by assessing complex assemblies based on functionality, manufacturability, and assemblability, moving beyond…

View →
cs.AIcond-mat.mtrl-sciEmpiricalRecentOct 31, 2026

LM6-CastBench: A Geometry-Aware Simulation Dataset for Transient Thermal Prediction in LM6 Sand Casting

Suhani Kolhatkar, Anand Khandekar

This paper introduces LM6-CastBench, an open simulation dataset for geometry-aware learning of transient thermal fields in LM6 aluminum sand casting.

View →
eess.SPEmpiricalRecentSep 25, 2026

Variable Dimension IMM and GPB2 Filters for Tracking in Multiple Model Systems

Roberto Pérez-Pérez, Ángel F. García-Fernández

This paper presents mathematically principled filters for multiple-model systems with states of different dimensionality.

View →
cs.CRcs.CVMay 10, 2026

On the Generation and Mitigation of Harmful Geometry in Image-to-3D Models

Yule Liu, Yilong Yang, Jiale Teng, Hanze Jia +10 more

The paper systematically measures the risk of current image-to-3D models generating harmful geometries, finding that these models are effective at reconstruction and existing safeguards are insufficie…

View →
cs.CERecentMay 31, 2026

MsFEM-Inspired CNNs with Transfer Learning for Multiscale Model Reduction

Xuehan Zhang, Lijian Jiang, Eric T. Chung

The paper proposes MITL, an MsFEM-inspired transfer learning strategy for CNN-based reduced-order models, enabling efficient and adaptable approximation of multiscale systems with minimal retraining.

View →
cs.SEcs.PLEmpiricalRecentJun 9, 2026

Web-Native Graphical EMF Model Editors

Susanne Göbel, Ralf Lämmel

Graphical model editing framework for EMF models

View →