This week in Multimodal

Latent Reasoning in Continuous Space for Unified Multimodal ModelsOct 30 · ▲ 24
SemEllipsoid: Semantic-aware Shape Abstraction with EllipsoidOct 30
ReLift: Dense 3D Tracking as Lifted Trajectory RepairOct 30 · ▲ 22

This week

Get the morning digest

The top stories in the fields you follow, at 08:00 your time.

Latest in AI and developer tools

Today's trail

updated 18m ago

Last hour Today

Pulse · tech-insider.org

tech-insider.org · Nov 3

French Tech Central Day 2026: Nov. 19 Facts

French Tech Central Day 2026 is listed for Nov. 19. See what is confirmed, what is not, and the 2026 French startup funding data.

Pulse

Top picks this weekarXiv

arXiv

Oct 30

GazeRIB: Systematic Evaluation of Multi-Reader Gaze-Guided Chest X-ray Learning

Radiologists' visual attention can supervise medical image learning, but how reader panel composition affects its value remains untested. We evaluated GazeRIB (Gaze-guided Radiologist-Informed Blackout), which masks unviewed regions during training and uses unmasked images at inference. Eight experts repeatedly read the same 80 chest X-rays (CXRs). We compared individual readers with three-, five- and eight-reader panels across seven architectures and 10 seeds, evaluating 910 models on a separate 2,897-CXR cohort for four-class severity prediction. Every panel improved macro F1 over the no-gaze baseline by 5.33-7.47 percentage points (pp; Bonferroni-adjusted p < 0.007); no individual reader reached significance on this endpoint. Three readers captured 73-80% of the eight-reader gain in macro F1, the area under the receiver-operating-characteristic curve (ROC-AUC) and the area under the precision-recall curve (AU-PRC). On external validation without retraining, all four panels improved ROC-AUC on SIIM-FISABIO-RSNA by +1.93 to +4.24 pp (p <= 0.004); on MIMIC-Eye no panel differed significantly from baseline, whereas the single-reader average declined (-1.40 pp, p = 0.008). Reader…

Top picks

All arXiv Hugging Face GitHub Hacker News

Latest Top this week

GitHub

Apr 23

Today's trail

Most active this week

indirect prompt injection · prompt injectionagents37 this week
recursive self improvement · airecursive self improvement32 this week
clinical reasoning · clinical decisionclinical reasoning31 this week
neural operators · physics informed neuralneural operators29 this week
agents · toolagents26 this week
code generation · language models llmsagents25 this week
graph neural · graph learningneural operators24 this week
catastrophic forgetting · continual learningInference & serving24 this week
coding agents · coding agentagentsNew22 this week
deep neural · neural networksneural operators21 this week

Rising topics

Rising tools

Soup+1225% downloads · 17k/wk+1225.2%
jcode+233% downloads · 25.5k/wk+233.1%
billion-context+259% downloads · 1.1M/wk+258.8%
LMDeploy+116% downloads · 13.1k/wk+116.3%
opencode+47% downloads · 5.2M/wk+47.3%
Scalar+72% downloads · 533k/wk+71.6%
better-auth+78% downloads · 674.2k/wk+77.8%
Qwen Code+58% downloads · 140.4k/wk+58.0%

All tools