AI summaries of arXiv papers
Paste any arXiv URL or paper ID. Or just swap arxiv.org → arxivtldr.org in your address bar.
Add to Chrome — Free ExtensionToday's Papers
The evolution of galaxy dust scaling relations in the COLIBRE simulations
COLIBRE simulations reveal the evolution of galaxy dust scaling relations across cosmic time, showing silicate dominance and resolution-dependent limits.
Pass the Baton: Trajectory-Relayed On-Policy Distillation
Relay-OPD improves on-policy distillation by using a teacher to correct student "prefix failures" during training, boosting performance and efficiency.
INTACT: Isomorphic Intent-to-Action Learning for Search-Free World Models
INTACT introduces an isomorphic intent-to-action learning approach for search-free world models, achieving high success with fast, direct policy inference.
$π\mathbf{R}^2$: Reactive Real-time Flow Policies
$π\mathbf{R}^2$ makes large backbone manipulation policies reactive and real-time by splitting sensory input and adapting to varying hardware latency.
Spend Experts Where You Are Unsure: Confidence-Adaptive Routing for Mixture-of-Experts LoRA
CARE dynamically adjusts the number of active experts in MoE-LoRA based on token uncertainty, improving performance and efficiency.
S2A2: Audio-Visual Imitation Learning for Manipulation Tasks Using Acoustic Spatial Information
S2A2 is a new audio-visual imitation learning framework that uses acoustic spatial information for robots to perform manipulation tasks requiring sound source localization.
Re-thinking Mammography Transfer Learning: The Dataset-Informed Transfer Learning (DITL) Framework for Breast Cancer Screening and Lesion Diagnosis
DITL is a new transfer learning framework for mammography that improves breast cancer screening and lesion diagnosis by adapting to dataset characteristics.
VetClaw: An Edge-Cloud Multimodal Agentic System for Veterinary Disease Screening
VetClaw is an edge-cloud multimodal agentic system for early veterinary disease screening, improving zero-shot classification with symptom-guided inputs.
Desktop-Delta Bench: Do Computer-Use Models Understand Desktop GUI Transitions?
Desktop-Delta Bench introduces a new benchmark to diagnose if computer-use agents understand desktop GUI transitions, revealing gaps in current models.
Reinformed Dreamer: An Asymmetric World Model Efficiently Trained through Latent Guidance
Reinformed Dreamer is a new asymmetric model-based RL algorithm that uses latent guidance to efficiently learn better representations, outperforming prior approaches.
Wonder: Video World Model Done Better
Wonder is a real-time, camera-controllable video world model that creates interactive, explorable virtual environments from images or videos.
Falling Behind Drives Unsafe Development in an Idealised AI Race Experiment
An AI race experiment shows that fear of falling behind and opponent actions, not just risk preferences, drive unsafe development choices.
Browse by Category
Browse by author →Browse by Topic
All topics →📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.
The URL swap trick
Reading a paper on arxiv.org? Just change the domain to arxivtldr.org in your address bar:
Paste a link
Enter any arXiv URL or paper ID
AI summarizes
Get a TLDR, key bullets, and why it matters
Read & share
Get key insights in seconds, share with colleagues