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

current page

Top radar

http://arxiv.org/abs/2608.27455v1

Code links

on this page

Paper	Authors	arXiv ID	Categories	Published	Code	Radar	Summary
CritICL: Inference-Time Weak-to-Strong Generalization from Small Language Model Failure Modes	Yufan Wu, Yinghui He, Zhengyi Hu	http://arxiv.org/abs/2608.27455v1	cs.CL	Aug 27, 2026	Detected	RDR83	Recent advances in inference-time scaling have significantly improved the reasoning performan...
Puro-2B: Poor Lab's Qwen2-1.5B Trained on RTX 5090 within $5090	Kairong Luo, Jiarui Cui, Yaorui Yin	http://arxiv.org/abs/2608.27370v1	cs.CL, cs.LG	Aug 27, 2026	Detected	RDR87	Language model pretraining has become almost synonymous with prohibitive cost, placing it out...
RedEvoAgent: Automatic Red-Teaming Agent with Experience-Driven Skill Evolution	Junjie Zhang, Hui Liu, Kecheng Chen	http://arxiv.org/abs/2608.27439v1	cs.CR, cs.AI	Aug 27, 2026	None	RDR82	LLM-based agents are increasingly deployed in product-level execution harnesses, where jailbr...
Retrieval Heads Meet Vision: Uncovering How VLMs Locate and Extract Visual Information	Chanho Park, Daehyeon Choi, Jihyun Lee	http://arxiv.org/abs/2608.27417v1	cs.CV	Aug 27, 2026	None	RDR84	Vision-language models (VLMs) can locate an image region referred to by a text prompt and rou...
Reconstructing Humans and Objects in Interaction using Large Reconstruction Models	Agniv Chatterjee, Georgios Pavlakos	http://arxiv.org/abs/2608.27407v1	cs.CV	Aug 27, 2026	None	RDR83	Estimation of Human-Object Interactions in 3D (3D HOI) is a fundamental problem in 3D compute...
CLAP: Cross-Embodiment Video World Models are Zero-Shot Physical Simulators	Kechen Liu, Ola Shorinwa	http://arxiv.org/abs/2608.27406v1	cs.RO, cs.AI	Aug 27, 2026	None	RDR86	State-of-the-art action-conditioned video models are typically restricted to a single robot e...
Making Clinical Language Models Auditable: Concept-Guided Fine-Tuning for Robust Prediction	Jin Mu, Guanhua Chen	http://arxiv.org/abs/2608.27397v1	cs.CL, cs.AI	Aug 27, 2026	None	RDR84	Clinical language models can achieve strong in-hospital accuracy yet fail under deployment sh...
LeVJEPA: Efficient & Scalable Video Pretraining without the Heuristics	Lukas Kuhn, Lucas Maes, Giuseppe Serra	http://arxiv.org/abs/2608.27395v1	cs.CV, cs.AI	Aug 27, 2026	None	RDR87	Video carries the temporal structure of the physical world, yet learning representations from...
Persona-Execution Separation: An Architecture Pattern for Evolving LLM Agents under Execution Audit	Yisen Xi	http://arxiv.org/abs/2608.27427v1	cs.SE, cs.AI	Aug 27, 2026	None	RDR77	Large language model (LLM) agents in governed organizations must let the persona (instruction...
RATIO: A Benchmark for Retrieval Across Typed Ideation Operations in Scientific Literature	Maayan Sharon, Tom Hope	http://arxiv.org/abs/2608.27394v1	cs.CL, cs.IR	Aug 27, 2026	None	RDR84	Retrieved scientific literature can serve as inspiration for both human and AI scientists. In...