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
Pixel-Level Pavement Distress Assessment Using Instance Segmentation	Logan Dewick, Bibesh Pyakurel, Kong Pheng Yang	http://arxiv.org/abs/2605.26095v1	cs.CV	May 25, 2026	None	RDR80	Automated pavement distress assessment requires more than image-level classification or coars...
Uniform Diffusion Models Revisited: Leave-One-Out Denoiser and Absorbing State Reformulation	Samson Gourevitch, Yazid Janati, Dario Shariatian	http://arxiv.org/abs/2605.22765v1	cs.LG, stat.ML	May 21, 2026	Detected	RDR80	Discrete diffusion models are often trained through clean-data prediction, but the prediction...
Physics Is All You Need? A Case Study in Physicist-Supervised AI Development of Scientific Software	Nhat-Minh Nguyen	http://arxiv.org/abs/2605.30353v1	cs.AI, astro-ph.CO	May 28, 2026	None	RDR79	Are AI agents tools, co-authors, or researchers? We present a quantified case study ($N=1$):...
VideoMLA: Low-Rank Latent KV Cache for Minute-Scale Autoregressive Video Diffusion	Hidir Yesiltepe, Jiazhen Hu, Tuna Han Salih Meral	http://arxiv.org/abs/2605.30351v1	cs.CV, cs.AI	May 28, 2026	None	RDR79	Long-rollout causal video diffusion has converged on a fixed-size sliding-window KV cache, wi...
DynaFLIP: Rethinking Robotics Perception via Tri-Modal-Dynamics Guided Representation	Jusuk Lee, Seungjae Lee, Jonghun Shin	http://arxiv.org/abs/2605.30350v1	cs.RO, cs.LG	May 28, 2026	None	RDR79	Robot manipulation critically depends on perception that preserves the action-relevant aspect...
Tiny but Trusted: Efficient Vision-Language Reasoning for Time-Series Anomaly Detection	Xiaona Zhou, Muntasir Wahed, Tianjiao Yu	http://arxiv.org/abs/2605.30344v1	cs.AI	May 28, 2026	None	RDR79	Recent advances in Vision-Language Models (VLMs) have achieved impressive performance across...
Demystifying Data Organization for Enhanced LLM Training	Yalun Dai, Yangyu Huang, Tongshen Yang	http://arxiv.org/abs/2605.30334v1	cs.AI, cs.CL	May 28, 2026	Detected	RDR79	Large Language Models (LLMs) have revolutionized various fields, yet their training efficienc...
From Pixels to Words -- Towards Native One-Vision Models at Scale	Haiwen Diao, Jiahao Wang, Penghao Wu	http://arxiv.org/abs/2605.28820v1	cs.CV	May 27, 2026	Detected	RDR79	Current vision-language models (VLMs) typically stitch together separate image encoders and l...
Beyond Binary: Sim-to-Real Dexterous Manipulation with Physics-Grounded Contact Representation	Jiahe Pan, Stelian Coros, Jitendra Malik	http://arxiv.org/abs/2605.28812v1	cs.RO, cs.AI	May 27, 2026	None	RDR79	A primary bottleneck in contact-rich manipulation is the difficulty of collecting real-world...
Algorithmic Monocultures in Hiring	Rishi Bommasani, Sarah H. Bana, Kathleen A. Creel	http://arxiv.org/abs/2605.27371v1	cs.CY, cs.AI	May 26, 2026	None	RDR79	Many employers screen job applicants with algorithms built by the same few algorithm vendors....