LIVE-Last scan updating-53 sources active-903 signals today-RESEARCH PVBVR-Pro: A Scalable and Verifiable Suite for Native Visual Reasoning
AI Radar
Subscribe
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
2,287
matching records
Shown
10
current page
Top radar
81
http://arxiv.org/abs/2608.24823v1
Code links
4
on this page
PaperAuthorsarXiv IDCategoriesPublishedCodeRadarSummary
BioKERN: Biological Kernel Regularization for Histology-to-Transcriptomics Neighborhood RetrievalSeungik Cho, Betul Orcan-Ekmekcihttp://arxiv.org/abs/2608.24823v1cs.LG, q-bio.QMAug 25, 2026NoneRDR81Spatially resolved biology requires representations that preserve biological neighborhood str...
MoE-based Feature Adapter for Prompt-free Binary Coronary Artery Segmentation in X-ray AngiographyLin Xi, Yingliang Mahttp://arxiv.org/abs/2608.24783v1cs.CVAug 25, 2026NoneRDR81Accurate segmentation of coronary arteries in X-ray angiography videos is essential for quant...
Prime Agent: A Self-Improving RLM HarnessSeth Karten, Alex L. Zhang, Kevin Thomashttp://arxiv.org/abs/2608.23552v1cs.AI, cs.CLAug 24, 2026DetectedRDR86Language models are sequential processors, but long-horizon agency requires external informat...
SRPO: Self-Reflective Policy Optimization for Long-Horizon ReasoningJialong Liu, Yuling Shi, Ning Yanghttp://arxiv.org/abs/2608.23493v1cs.AIAug 24, 2026DetectedRDR86Self-reflection is a powerful mechanism for credit assignment in human learning, converting s...
When Adaptation Hurts: Connecting Representational Drift to OOD Failures in MedSAM Fine-TuningMarko Haralović, Sounic Akkaraju, Carlo Barettahttp://arxiv.org/abs/2608.21300v1cs.CVAug 21, 2026DetectedRDR82Foundation models for medical image segmentation, like prompt-based MedSAM, generalize well a...
EnSI-RAG: Entity-Structure-Indexed Retrieval-Augmented Generation for Long-Document Question AnsweringXuanyu Meng, Jiashuo Sun, Jash Rajesh Parekhhttp://arxiv.org/abs/2608.21252v1cs.CL, cs.AIAug 21, 2026DetectedRDR84Question answering (QA) over long, connected documents remains challenging because relevant e...
Token-Level AdvertisingHanbing Liu, Bowei Zhang, Changyuan Yuhttp://arxiv.org/abs/2608.27382v1cs.GT, cs.LGAug 27, 2026NoneRDR77Generative AI is transforming how people access information, challenging traditional advertis...
Stageboost: Recommending Signals Based on Counterfactual EstimationDarpan Singhal, Matan Mandelbrod, Tal Franjihttp://arxiv.org/abs/2608.27366v1cs.IR, cs.AIAug 27, 2026NoneRDR65Signals are short textual or visual snippets displayed on the eBay View-Item (VI) page, provi...
$R^3$: Training Robots to Reason in Natural Language via Reinforcement LearningLehong Wu, Yuxiao Qu, Zheyuan Huhttp://arxiv.org/abs/2608.26053v1cs.RO, cs.AIAug 26, 2026NoneRDR79Reasoning in language allows foundation models to spend more test-time compute on hard proble...
Automatic Model Card Generation Using an LLMTajkia Rahman Toma, Balreet Grewal, Cor-Paul Bezemerhttp://arxiv.org/abs/2608.24807v1cs.SE, cs.AIAug 25, 2026NoneRDR73Model cards are structured documents that summarize key information about machine learning mo...