ArXiv TLDR
26,162 papers summarized

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 Extension
Computer Vision

FurE: Efficient Instance-Specific 3D Fur Reconstruction without Animal-Fur Datasets

FurE reconstructs realistic 3D animal fur from multi-view images using a novel strand-based method, achieving 10x speedup without animal-fur datasets.

2609.35770
Natural Language Processing

Telescopic Language Models

Telescopic Language Models (TLMs) are single, nested Transformers that function as valid LMs at various capacities, trained with stochastic prefix supervision.

2609.35769
Computer Vision

PDMD: Projected Distribution Matching Distillation for Video Diffusion Models

PDMD stabilizes video diffusion model distillation by projecting out critic errors, improving sample quality and efficiency with minimal code changes.

2609.35768
Computer Vision

Learning Native Reflection in Unified Models with Interleaved Reinforcement Learning

This paper introduces UMM-Reflection, an RL method for unified multimodal models to self-repair image generations through interleaved reflection and revision.

2609.35767
Natural Language Processing

Retrieving Biblical Intertextual References in Karen Blixen's Seven Gothic Tales

This paper evaluates computational methods for retrieving biblical intertextual references in Karen Blixen's 'Seven Gothic Tales,' comparing sparse and dense models.

2609.35765
Computer Vision

Reliability-Gated Fusion of Consumer Head and Foot IMUs for Lower-Body 3D Pose

This paper introduces a reliability-gated fusion model for consumer head and foot IMUs, improving 3D lower-body pose estimation by learning to trust sensor channels.

2609.35764
Machine Learning

Unifying Distributional Training for One-Step Visual Generation

This paper unifies distributional training for one-step visual generation and introduces MGFlow, a new method achieving state-of-the-art results.

2609.35763
Robotics

DexRoam: Learning Mobile Bimanual Dexterous Manipulation from Egocentric Whole-Body Human Demonstrations

DexRoam enables robots to learn complex mobile bimanual manipulation from egocentric human demonstrations, improving success rates and data efficiency.

2609.35761
Machine Learning

TokenCast: Forecasting Token Consumption During LLM Agent Execution

TokenCast predicts LLM agent token consumption by learning composable cost representations, improving forecast accuracy and budget control.

2609.35760
Natural Language Processing

Scaling Long-Form Story Generation via Narrative State Tracking

NstAgent is a training-free framework that enables LLMs to generate consistent, high-quality long-form stories up to 100K words by tracking narrative state.

2609.35759
Machine Learning

Neural Harmonic Measure Operator

NHMO is a neural solver for elliptic PDEs on variable-shape domains, using a transformer-based boundary kernel for efficient solutions.

2609.35752
Machine Learning

How to Loop MoE: Flatten the Experts, Untie the Attention

Foil introduces a novel method for looping Mixture-of-Experts (MoE) models, improving expert utilization and pretraining loss through flattened experts and untied attention.

2609.35751

📬 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:

arxiv.org/abs/2401.12345
↓
arxivtldr.org/abs/2401.12345
1

Paste a link

Enter any arXiv URL or paper ID

2

AI summarizes

Get a TLDR, key bullets, and why it matters

3

Read & share

Get key insights in seconds, share with colleagues