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/2610.03702v1

Code links

on this page

Paper	Authors	arXiv ID	Categories	Published	Code	Radar	Summary
LESSER: Post-Training Data Selection with Output-Layer Gradients	Lyuxin David Zhang, Eric Wong, Surbhi Goel	http://arxiv.org/abs/2610.03702v1	cs.LG	Oct 2, 2026	None	Score70	The choice of post-training data for large language models substantially affects downstream p...
Markov Chain Convergence in Kernel-Induced Geometry	Oleksii Kachaiev, Luc Brogat-Motte, Silvia Villa	http://arxiv.org/abs/2610.13916v1	math.PR, stat.ML	Oct 9, 2026	None	Score58	We study the convergence of Markov chains in the geometry induced by Hilbertian embeddings as...
Base Models Can Reason By Taking a Cue From Training Data	Sophie L. Wang, Amil Dravid, Rulin Shao	http://arxiv.org/abs/2610.06851v1	cs.LG, cs.AI	Oct 5, 2026	None	Score65	In this paper, we study how training data creates associations between the tokens at the star...
Anatomy-aware Fine-grained Multimodal Fusion for Laryngopharyngeal Cancer T-Staging Prediction Using CT and Radiology Report	Xingyue Zhao, Yanzhou Su, Fang Zhang	http://arxiv.org/abs/2610.06837v1	cs.CV	Oct 5, 2026	None	Score61	Accurate T-staging is crucial for guiding personalized treatment strategies for laryngopharyn...
Direct Intermediate Initialization for Tilted Diffusion Samplers	Gregory D. Bellchambers	http://arxiv.org/abs/2610.06834v1	stat.ML, cs.LG	Oct 5, 2026	None	Score73	Some diffusion posterior samplers construct Gaussian-tilted intermediate distributions along...
Block Disentanglement in CRL: Bridging Identifiability and Visual State Estimation	Emre Acartürk, Pranamya Kulkarni, Puranjay Datta	http://arxiv.org/abs/2610.06809v1	cs.LG, stat.ML	Oct 5, 2026	None	Score64	Causal representation learning (CRL) is the process of recovering causally-related latent var...
MC-Sparse: Deconstructing and Closing the Dense-Sparse Attention Gap in Diffusion Transformers	Jiarui Chen, Zeqiang Lai, Jiangshan Wang	http://arxiv.org/abs/2610.06801v1	cs.CV, cs.AI	Oct 5, 2026	None	Score64	Sparse attention is a primary approach to reducing the latency of diffusion transformers in l...
PhaseMatcher: Autoregressive Phase-Set Identification with Spectral Decomposition	Zhonglong Peng, Qiuliang Liu, Chang Chen	http://arxiv.org/abs/2610.05844v1	cs.CE, cond-mat.mtrl-sci	Oct 5, 2026	None	Score73	Recovering complete phase sets from powder X-ray diffraction (PXRD) is challenging when weak-...
Constructive Bounds for the Topological Expressivity of Neural Networks	Kosio Beshkov	http://arxiv.org/abs/2610.13931v1	math.AT, cs.LG	Oct 9, 2026	None	Score61	The expressivity of a neural network $Φ$ describes the class of functions that a network can...
Finding Gaussian Structure in Bosonic States	Alvan Arulandu, Sitan Chen, Ziyun Chen	http://arxiv.org/abs/2610.06810v1	quant-ph, cs.DS	Oct 5, 2026	None	Score64	We study agnostic tomography of pure bosonic Gaussian states: given copies of an arbitrary $n...