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
MuellerPT: Decomposition Driven Pretraining for Dense Learning in Mueller Polarimetry	Adam Tlemsani, Yingdian Li, Maxime Giot	http://arxiv.org/abs/2605.23840v1	cs.CV	May 22, 2026	None	RDR82	Mueller matrix imaging provides rich, physically meaningful contrast for biomedical tissue an...
Which Way Did It Move? Diagnosing and Overcoming Directional Motion Blindness in Video-LLMs	Jongseo Lee, Hyuntak Lee, Sunghun Kim	http://arxiv.org/abs/2605.22823v1	cs.CV	May 21, 2026	Detected	RDR82	Video Large Language Models (Video-LLMs) have made rapid progress on temporal video understan...
SoundnessBench: Can Your AI Scientist Really Tell Good Research Ideas from Bad Ones?	Sy-Tuyen Ho, Minghui Liu, Huy Nghiem	http://arxiv.org/abs/2605.30329v1	cs.LG	May 28, 2026	None	RDR81	Autonomous AI research agents aim to accelerate scientific discovery by automating the resear...
PEFT-Arena: Understanding Parameter-Efficient Finetuning from a Stability-Plasticity Perspective	Yangyi Huang, Ruotian Peng, Zeju Qiu	http://arxiv.org/abs/2605.28819v1	cs.LG, cs.CL	May 27, 2026	None	RDR81	Parameter-efficient finetuning (PEFT) has become the standard approach for adapting large lan...
Can Large Language Models Handle Discourse Particles? A Case Study of Colloquial Malay	Mariah Al Giptiah Binte Yusoff, Jakin Tan, Bocheng Chen	http://arxiv.org/abs/2605.28782v1	cs.CL	May 27, 2026	None	RDR81	Discourse particles, such as \textit{well} and \textit{kind of}, are crucial components that...
The Abstraction Gap in Vision-Language Causal Reasoning	Chinh Hoang, Mohammad Rashedul Hasan	http://arxiv.org/abs/2605.28779v1	cs.CL, cs.CV	May 27, 2026	None	RDR81	Vision-language models (VLMs) generate fluent causal explanations, but current evaluations ca...
Global Structure-from-Motion Meets Feedforward Reconstruction	Linfei Pan, Johannes Schönberge, Marc Pollefeys	http://arxiv.org/abs/2605.26103v1	cs.CV	May 25, 2026	Detected	RDR81	Structure-from-Motion -- the process of simultaneously estimating camera poses and 3D scene s...
Claw-Anything: Benchmarking Always-On Personal Assistants with Broader Access to User's Digital World	Yusong Lin, Xinyuan Liang, Haiyang Wang	http://arxiv.org/abs/2605.26086v1	cs.AI	May 25, 2026	None	RDR81	Large language model agents are increasingly envisioned as always-on personal assistants with...
DeltaBox: Scaling Stateful AI Agents with Millisecond-Level Sandbox Checkpoint/Rollback	Yunpeng Dong, Jingkai He, Yuze Hou	http://arxiv.org/abs/2605.22781v1	cs.OS, cs.AI	May 21, 2026	None	RDR81	LLM-powered AI agents require high-frequency state exploration (e.g., test-time tree search a...
Benchmarking Single-Factor Physical Video-to-Audio Generation	Tingle Li, Siddharth Gururani, Kevin J. Shih	http://arxiv.org/abs/2605.30339v1	cs.CV, cs.MM	May 28, 2026	None	RDR80	Generative video-to-audio (V2A) models produce highly plausible soundtracks, but it remains u...