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

1,289

1 arXiv sources

Top radar

http://arxiv.org/abs/2607.13027v1

Published 24h

by paper date

Rate limit

1/3s

arXiv policy

Paper	Authors	arXiv ID	Categories	Published	Code	Radar	Summary
Dual Latent Memory in Vision-Language-Action Models for Robotic Manipulation	Hongyu Qu, Jianzhe Gao, Xiaobin Hu	http://arxiv.org/abs/2607.07608v1	cs.RO, cs.CV	Jul 8, 2026	None	RDR61	Mainstream Vision-Language-Action (VLA) models predict actions primarily from the current obs...
The Large Cancer Assistant (LCA): A Model-Agnostic Orchestration Framework for Scalable Clinical Decision Support in Oncology	Ghassen Marrakchi, Basarab Matei	http://arxiv.org/abs/2607.06531v1	cs.AI, cs.LG	Jul 7, 2026	None	RDR66	- Objective: Multimodal deep learning models in oncology are currently limited by monolithic...
The Effects of Synthetic Data and Label Distribution on Canola Branch Counting	Amirsalar Darvishpour, Mikolaj Cieslak, Adam Runions	http://arxiv.org/abs/2607.09630v1	cs.CV	Jul 10, 2026	None	RDR69	Collecting annotated plant images for automated phenotyping is often slow and expensive. Plan...
MPFlow: Learning Budgeted Max-Flow Optimization on the Lightning Network with Deep Graph Reinforcement Learning	Harrison Rush, Vincent Davis, Simone Antonelli	http://arxiv.org/abs/2607.08703v1	cs.LG	Jul 9, 2026	None	RDR62	We address liquidity placement in the Bitcoin Lightning Network (LN): given a fixed budget, w...
How Data Shapes RoPE Frequency Usage: From Positional Scale Matching to Length Generalization	Xinyi Wu, Siyuan Liu, Ali Jadbabaie	http://arxiv.org/abs/2607.07678v1	cs.LG	Jul 8, 2026	None	RDR64	Rotary Position Embeddings (RoPE) provide transformers with a fixed grid of positional freque...
Guidance Breaks the Fitted Operator: A Terminal-Fitted Repair for Classifier-Free Guidance	Shiheng Zhang	http://arxiv.org/abs/2607.07665v1	cs.LG, math.NA	Jul 8, 2026	None	RDR64	Classifier-free guidance (CFG) is the standard way to strengthen class-conditioning in diffus...
RL Post-Training Builds Compositional Reasoning Strategies	Azwar Abdulsalam, Nishil Patel, Andrew Saxe	http://arxiv.org/abs/2607.07646v1	cs.AI, cs.CL	Jul 8, 2026	None	RDR64	Does RL post-training merely amplify primitive skills already latent in a base model, or can...
On the feasibility of dependency parsing of non-human sequences without a gold standard. Is evaluation possible in other species?	Ramon Ferrer-i-Cancho, Catherine Hobaiter, Thore Bergman	http://arxiv.org/abs/2607.06542v1	cs.CL	Jul 7, 2026	None	RDR80	Dependency parsing consists of finding a tree representation for a sequence. Unsupervised dep...
GraphBU: MILP Instance Generation with Graph-Native Block Units	Xiaolei Guo, Chenyu Zhou, Jianghao Lin	http://arxiv.org/abs/2607.06532v1	cs.LG, math.OC	Jul 7, 2026	None	RDR75	Mixed-integer linear programming (MILP) instances used for solver development are hard to obt...
RSF-GLLM: Bridging the Semantic Gap in Multi-Hop Knowledge Graph QA via Recurrent Soft-Flow and Decoupled LLM Generation	Sambaran Bandyopadhyay, Ananth Muppidi	http://arxiv.org/abs/2607.06527v1	cs.CL, cs.AI	Jul 7, 2026	None	RDR65	Multi-hop Question Answering over Knowledge Graphs faces a critical challenge: traditional re...