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
10
current page
Top radar
70
http://arxiv.org/abs/2610.03698v1
Code links
0
on this page
| Paper | Authors | arXiv ID | Categories | Published | Code | Radar | Summary |
|---|---|---|---|---|---|---|---|
| Decoding the Functional Roles of Register and High-Norm Patch Tokens in Vision Transformers | Neel Varma, Andrew Rufail, Dipika Khullar | http://arxiv.org/abs/2610.03698v1 | cs.CV | Oct 2, 2026 | None | Score70 | Self-supervised Vision Transformers (ViTs), such as DINOv2, learn rich visual representations... |
| Pivot-SD: Efficient Self-Distillation for Masked Diffusion Language Models | Seo Hyun Kim, Sunwoo Hong, Younwoo Choi | http://arxiv.org/abs/2610.03665v1 | cs.LG, cs.CL | Oct 2, 2026 | None | Score77 | Masked diffusion language models (dLMs) offer a promising parallel alternative to autoregress... |
| ProAR: Learning Prospective Reasoning with Autoregressive Video Models | Linghui Shen, Tinghui Zhu, Sheng Zhang | http://arxiv.org/abs/2610.03664v1 | cs.CV | Oct 2, 2026 | None | Score68 | Autoregressive (AR) video models excel at causal generation, but their reliance on next-chunk... |
| Forecasting from Counterfactual Simulator Rollouts: A Sim2Real Evaluation | Angel Wang, Dominique Perrault-Joncas, Alvaro Maggiar | http://arxiv.org/abs/2610.03662v1 | cs.LG | Oct 2, 2026 | None | Score75 | Deploying a new decision policy creates a cold-start problem for prediction models whose targ... |
| Where to Look Matters: Rethinking Sub-Volume Sampling in 3D Medical Self-supervised Learning | Junkai Liu, Le Zhang | http://arxiv.org/abs/2610.13915v1 | cs.CV | Oct 9, 2026 | None | Score59 | Self-supervised learning on 3D medical images commonly relies on random sub-volume sampling d... |
| Tracking the Invisible: Self-Supervised Wheel Track Detection with Thermal-RGB Fusion | Yiming Yang, Jeremy P. Bos | http://arxiv.org/abs/2610.13872v1 | cs.CV, cs.RO | Oct 9, 2026 | None | Score69 | Accurate detection of drivable area is essential for autonomous driving and advanced driver a... |
| A Balanced Data Diet: Addressing the Exploration Bottleneck in Mega-Scale RL for Robot Control | Octi Zhang, Mateo Guaman Castro, Patrick Yin | http://arxiv.org/abs/2610.12465v1 | cs.RO, cs.LG | Oct 8, 2026 | None | Score72 | General-purpose robots must perform a wide range of tasks from agile locomotion to dexterous... |
| Hybrid Cinematography: Previsualizing and Managing Hallucination Risk in Generative Video Reshooting | Nhan, Tran, Neal Wadhwa | http://arxiv.org/abs/2610.12455v1 | cs.HC, cs.CV | Oct 8, 2026 | None | Score67 | On a film set, the camera move is committed during a take. Generative video reshooting lets f... |
| Two-Level Softmax Sampling Done Right: Correcting Bias from Size Imbalance and Dispersion | Walid Bendada, Guillaume Salha-Galvan | http://arxiv.org/abs/2610.10483v1 | cs.LG, cs.IR | Oct 7, 2026 | None | Score74 | Sampling from a softmax distribution is a fundamental operation in machine learning, but its... |
| Reinforcement Learning with Conformal Action Sets: An Application to Sequential Recommendation | Wenwen Si, Honghao Wei | http://arxiv.org/abs/2610.08743v1 | cs.LG, cs.AI | Oct 6, 2026 | None | Score74 | Sequential recommenders typically use a fixed slate size even though the number of useful alt... |