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

3,367

matching records

Shown

current page

Top radar

http://arxiv.org/abs/2610.05830v1

Code links

on this page

Paper	Authors	arXiv ID	Categories	Published	Code	Radar	Summary
Measurement-First Auditing of Agentic Leaderboards: Contamination Susceptibility, Matched-Control Re-evaluation, and Scorer Validation	Dishu Yang, Qi Su, Hongbo Qin	http://arxiv.org/abs/2610.05830v1	cs.AI	Oct 5, 2026	None	Score87	Agentic leaderboards increasingly evaluate systems on public benchmarks whose task statements...
Prediction Is Not a Sufficient Memory State	Robert Jenkinson Alvarez	http://arxiv.org/abs/2610.13881v1	cs.LG	Oct 9, 2026	None	Score72	A memory that reproduces every current prediction may still have forgotten how to learn from...
DreamTrue: Action-Faithful Robot World Model with Counterfactual Post-Training	Junyan Li, Ruizhi Li, Yu Liu	http://arxiv.org/abs/2610.12468v1	cs.RO, cs.CV	Oct 8, 2026	None	Score76	We present DreamTrue, a multi-view, cross-embodiment robot world model for action-faithful an...
VioLA: Learning Generalist Humanoid Control Policies from Human Data	Mert Albaba, Jens Beißwenger, Anna Manasyan	http://arxiv.org/abs/2610.12435v1	cs.RO, cs.LG	Oct 8, 2026	None	Score76	Teaching a humanoid to follow instructions with its whole body runs into two obstacles. Its a...
RoboRSI: Stable, efficient, and reusable robot self-evolution in complex real-world environments	Zimo Wen, Yijin Chen, Yuxuan Cao	http://arxiv.org/abs/2610.12424v1	cs.RO, cs.AI	Oct 8, 2026	None	Score72	A generalist robot should not only perform diverse tasks but also improve through experience,...
Decoupling Exploration from Optimization in RLVR	Saif Punjwani, Micah Goldblum	http://arxiv.org/abs/2610.10536v1	cs.LG, cs.AI	Oct 7, 2026	None	Score73	Modern language models undergo reinforcement learning with verifiable rewards (RLVR) on top o...
EngramEdit: Decoupled Knowledge Updates in LLMs through Conditional Memory	Hongru Cai, Ran Wei, Wenjie Wang	http://arxiv.org/abs/2610.10533v1	cs.CL	Oct 7, 2026	None	Score66	Conditional memory architectures such as DeepSeek Engram use input n-grams to look up learned...
Rephrase Before You Act: Characterizing and Mitigating Language Sensitivity in Vision-Language-Action Models	Mikey Watts, Yuchen Cui	http://arxiv.org/abs/2610.10526v1	cs.RO, cs.CL	Oct 7, 2026	None	Score79	Vision-language-action models (VLAs) are strikingly sensitive to instruction phrasing and do...
GRACE: Generation-aware latent compression for efficient video generation	Jiyoung Kim, Paul Hyunbin Cho, Jisu Nam	http://arxiv.org/abs/2610.10524v1	cs.CV	Oct 7, 2026	None	Score72	Highly compressed video autoencoders offer an effective way to accelerate video diffusion mod...
Learning to Read the Contextual Tokens in Diffusion Transformers	Omer Dahary, Etai Sella, Hadar Averbuch-Elor	http://arxiv.org/abs/2610.06844v1	cs.CV, cs.AI	Oct 5, 2026	None	Score80	Multimodal Diffusion Transformers (MM-DiTs) jointly process visual and textual representation...