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

3,367

matching records

Shown

current page

Top radar

http://arxiv.org/abs/2610.03716v1

Code links

on this page

Paper	Authors	arXiv ID	Categories	Published	Code	Radar	Summary
MoSE3: Learning World-Space SE(3) at Every Pixel	Jiahuan Cheng, Zhiyi Li, Tian Xia	http://arxiv.org/abs/2610.03716v1	cs.CV	Oct 2, 2026	None	Score74	Dense 3D point tracking has been a prominent paradigm for modeling motion in dynamic scenes,...
What Should World Models Forget? Stratified Retention for Continual Adaptation	Nishit Anand, Ramani Duraiswami, Dinesh Manocha	http://arxiv.org/abs/2610.03713v1	cs.LG, cs.AI	Oct 2, 2026	None	Score81	Continual learning treats degradation on previously seen data as evidence of failure, a conve...
FlowHMR: Physically Plausible Motion Capture from Video	Zhanke Wang, Chengfeng Zhao, Qing Shuai	http://arxiv.org/abs/2610.03691v1	cs.CV	Oct 2, 2026	None	Score77	We present FlowHMR, a framework for recovering physically plausible global 3D human motion fr...
SigLIP2 for aerial fire risk classification	Yunus Serhat Bıçakçı	http://arxiv.org/abs/2610.03689v1	cs.CV	Oct 2, 2026	None	Score81	We examine the transfer of a pretrained SigLIP2 image encoder to seven class fire risk classi...
Goal-Oriented Bundle Management and Flow Control for Deep-Space Communications via Reinforcement Learning	Sajjad Baghaeea, Aimin Lib, Shanmugarajan Balasubramanian	http://arxiv.org/abs/2610.13923v1	cs.NI, cs.LG	Oct 9, 2026	None	Score62	Deep-space communications operate under intermittent connectivity, long propagation delays, a...
On the estimation and validity of AI time horizons---a statistical look at the METR plot	Drew T. Nguyen, William Fithian	http://arxiv.org/abs/2610.12466v1	cs.AI	Oct 8, 2026	None	Score73	METR's 50\% time horizon measures the human completion time of software tasks that an AI solv...
Decentralized SGD under Heavy-Tailed Noise: Optimal Convergence Rates and the Role of Gradient Clipping	Aleksandar Armacki, Haoyuan Cai, Ali H. Sayed	http://arxiv.org/abs/2610.10527v1	math.OC, cs.LG	Oct 7, 2026	None	Score64	Heavy-tailed noise has been widely observed in modern machine learning, motivating the use of...
Best Arm Identification for Bandits with Shifting Means	Lukas Zierahn, Wouter M. Koolen, Shubhada Agrawal	http://arxiv.org/abs/2610.10488v1	stat.ML, cs.LG	Oct 7, 2026	None	Score73	We study the best arm identification problem in a stochastic environment with a novel form of...
FoldBack: Self-Correcting Masked Generative Policy for Long-Horizon Garment Folding	Lipeng Zhuang, Shiyu Fan, Yingdong Ru	http://arxiv.org/abs/2610.10462v1	cs.RO, cs.AI	Oct 7, 2026	None	Score65	We present FoldBack, a self-correcting masked generative policy for long-horizon garment fold...
Composing What Each Teacher Learned: Multi-Teacher On-Policy Distillation through Teacher-Relative Shifts	Hejian Sang, Zhengze Zhou, Shayan Mohajer Hamidi	http://arxiv.org/abs/2610.10460v1	cs.LG, cs.AI	Oct 7, 2026	None	Score73	Multi-teacher on-policy distillation (MOPD) is used in two settings. In common-domain composi...