Research radar

arXiv preprints, filtered for developer impact.

Concise summaries of new AI papers, ranked for how likely they are to change how production systems are built. arXiv collection remains rate-limited.

Papers tracked

180

1 arXiv sources

Top radar

http://arxiv.org/abs/2605.30352v1

Published 24h

by paper date

Rate limit

1/3s

arXiv policy

Paper	Authors	arXiv ID	Categories	Published	Code	Radar	Summary
Gamma-World: Generative Multi-Agent World Modeling Beyond Two Players	Fangfu Liu, Kai He, Tianchang Shen	http://arxiv.org/abs/2605.28816v1	cs.CV	May 27, 2026	None	RDR63	World models for interactive video generation have largely focused on single-agent settings,...
Forgetting in Language Models: Capacity, Optimization, and Self-Generated Replay	Martin Marek, Dongkyu Cho, Shikai Qiu	http://arxiv.org/abs/2605.26097v1	cs.LG	May 25, 2026	None	RDR63	Models trained on a new task typically degrade on prior tasks, a phenomenon known as forgetti...
G3T Up! Gravity Aligned Coordinate Frames Simplify Pointmap Processing	Bharath Raj Nagoor Kani, Noah Snavely	http://arxiv.org/abs/2605.27372v1	cs.CV	May 26, 2026	None	RDR62	Modern feed-forward 3D reconstruction methods like VGGT predict pixel-aligned pointmaps in ca...
PARE: Pruning and Adaptive Routing for Efficient Video Generation	Yutong Wang, Yunke Wang, Tianfan Xue	http://arxiv.org/abs/2605.27336v1	cs.CV	May 26, 2026	None	RDR62	Video Diffusion Transformers (DiTs) generate high-quality videos but demand substantial compu...
Helix4D: Complex 4D Mesh Generation	Jiraphon Yenphraphai, Jianqi Chen, Jian Wang	http://arxiv.org/abs/2605.26109v1	cs.CV	May 25, 2026	None	RDR62	Current video-to-4D methods struggle with complex topology changes, transparent materials, th...
Complete-muE: Optimal Hyperparameter Transfer and Scaling for MoE Models	Hongwu Peng, Ohiremen Dibua, Yuanjun Xiong	http://arxiv.org/abs/2605.23893v1	cs.LG	May 22, 2026	None	RDR62	We propose Complete-muE, a framework which targets hyperparameter transfer across dense FFN a...
Move on Muon : A Hamiltonian probability gradient flow perspective of Muon optimizer	Aratrika Mustafi, Soumya Mukherjee, Bharath K. Sriperumbudur	http://arxiv.org/abs/2605.23871v1	stat.ML, cs.LG	May 22, 2026	None	RDR62	We develop a gradient flow on the space of probability measures defined on matrix-valued para...
Strong Teacher Not Needed? On Distillation in LLM Pretraining	Taiming Lu, Zhuang Liu	http://arxiv.org/abs/2605.23857v1	cs.LG, cs.CL	May 22, 2026	None	RDR62	Knowledge distillation generally assumes a strong-to-weak relationship where stronger teacher...
FAME: Failure-Aware Mixture-of-Experts for Message-Level Log Anomaly Detection	Huanchi Wang, Zihang Huang, Yifang Tian	http://arxiv.org/abs/2605.22779v1	cs.SE, cs.LG	May 21, 2026	None	RDR62	Production systems generate millions of log lines daily, yet most anomaly detectors operate a...
Veda: Scalable Video Diffusion via Distilled Sparse Attention	Shihao Han, Hao Yang, Xinting Hu	http://arxiv.org/abs/2605.30325v1	cs.CV	May 28, 2026	None	RDR61	Scaling Diffusion Transformers to generate high-resolution, long videos is constrained by the...