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

1,327

1 arXiv sources

Top radar

http://arxiv.org/abs/2607.13941v1

Published 24h

by paper date

Rate limit

1/3s

arXiv policy

Paper	Authors	arXiv ID	Categories	Published	Code	Radar	Summary
SemRF: A Semantic Reference Frame for Residual-Stream Dynamics in Language Models	Jian Gu, Aldeida Aleti, Chunyang Chen	http://arxiv.org/abs/2606.32022v1	cs.LG, cs.CL	Jun 30, 2026	None	RDR64	Residual-stream analysis asks how language-model computation evolves across depth, but interm...
DOPD: Dual On-policy Distillation	Xinlei Yu, Gen Li, Qingyi Si	http://arxiv.org/abs/2606.30626v1	cs.AI	Jun 29, 2026	None	RDR64	On-policy distillation (OPD) offers superior capacity transfer by supervising student-sampled...
Automated Background Swapping for Robustness against Spurious Backgrounds	Cesar Roder, Kajetan Schweighofer	http://arxiv.org/abs/2606.32018v1	cs.CV, cs.LG	Jun 30, 2026	None	RDR74	Classifiers based on Deep Neural Networks exhibit strong performance across domains, yet can...
GPU-Parallel Linearization Error Bounds for Real-Time Robust Optimal Control of Nonlinear and Neural Network Dynamics	Jeffrey Fang, Keyi Shen, Anutam Srinivasan	http://arxiv.org/abs/2607.01203v1	eess.SY, cs.AI	Jul 1, 2026	None	RDR64	This paper studies real-time robust optimal control for uncertain nonlinear systems, where li...
Decision-Aware Training for Sample-Based Generative Models	Kornelius Raeth, Nicole Ludwig	http://arxiv.org/abs/2607.01171v1	cs.LG, stat.ML	Jul 1, 2026	None	RDR64	Sample-based generative models are increasingly used for probabilistic forecasting in high-st...
AdaJEPA: An Adaptive Latent World Model	Ying Wang, Oumayma Bounou, Yann LeCun	http://arxiv.org/abs/2606.32026v1	cs.LG, cs.AI	Jun 30, 2026	None	RDR64	Latent world models enable planning from high-dimensional observations by predicting future s...
Random Reshuffling Dominates Stochastic Gradient Descent	Zijian Liu	http://arxiv.org/abs/2606.32005v1	math.OC, cs.LG	Jun 30, 2026	None	RDR74	Stochastic Gradient Descent ($\textsf{SGD}$) is one of the most classical optimization algori...
Radial Suppression Accelerates Algorithmic Generalization: A Geometric Analysis of Delayed Generalization	Srijan Tiwari, Aditya Chauhan, Manjot Singh	http://arxiv.org/abs/2606.32000v1	cs.LG, cs.AI	Jun 30, 2026	None	RDR64	Why do neural networks memorize algorithmic training data long before they generalize? We pre...
Pessimism's Paradox: Conservative Offline Training Amplifies Reward Hacking During Online Adaptation in Reasoning Models	Subramanyam Sahoo, Aman Chadha, Vinija Jain	http://arxiv.org/abs/2606.30627v1	cs.LG, cs.AI	Jun 29, 2026	None	RDR64	Conservative offline training is widely advocated as a safe foundation for subsequent online...
Neural Certificate Pricing for Combinatorial Optimization Problems	Jingyi Chen, Xinyuan Zhang, Xinwu Qian	http://arxiv.org/abs/2607.01185v1	cs.LG	Jul 1, 2026	None	RDR63	Combinatorial optimization (CO) problems are difficult because certifiable discrete structure...