new

Get trending papers in your email inbox once a day!

Get trending papers in your email inbox!

Daily Papers

byAK and the research community

Oct 26

Submitted by

akhaliq

A Picture is Worth a Thousand Words: Principled Recaptioning Improves Image Generation

·
5 authors

1

Submitted by

akhaliq

CommonCanvas: An Open Diffusion Model Trained with Creative-Commons Images

·
9 authors

Submitted by

akhaliq

DreamCraft3D: Hierarchical 3D Generation with Bootstrapped Diffusion Prior

·
7 authors

Submitted by

akhaliq

QMoE: Practical Sub-1-Bit Compression of Trillion-Parameter Models

·
2 authors

Submitted by

akhaliq

An Early Evaluation of GPT-4V(ision)

·
7 authors

Submitted by

akhaliq

Wonder3D: Single Image to 3D using Cross-Domain Diffusion

·
11 authors

Submitted by

akhaliq

ConvNets Match Vision Transformers at Scale

·
4 authors

Submitted by

akhaliq

LLM-FP4: 4-Bit Floating-Point Quantized Transformers

·
5 authors

Submitted by

akhaliq

Detecting Pretraining Data from Large Language Models

·
8 authors

Submitted by

akhaliq

CLEX: Continuous Length Extrapolation for Large Language Models

·
5 authors

Submitted by

akhaliq

TiC-CLIP: Continual Training of CLIP Models

·
8 authors

Submitted by

akhaliq

TD-MPC2: Scalable, Robust World Models for Continuous Control

·
3 authors

Submitted by

akhaliq

LightSpeed: Light and Fast Neural Light Fields on Mobile Devices

·
7 authors