BioGPT: Generative Pre-trained Transformer for Biomedical Text Generation and Mining

microsoft/biogpt 19 Oct 2022

Pre-trained language models have attracted increasing attention in the biomedical domain, inspired by their great success in the general natural language domain.

Document Classification Language Modelling +3

1,589
3.49 stars / hour

TEXTure: Text-Guided Texturing of 3D Shapes

TEXTurePaper/TEXTurePaper 3 Feb 2023

In this paper, we present TEXTure, a novel method for text-guided generation, editing, and transfer of textures for 3D shapes.

Image Generation text-guided-generation

157
1.92 stars / hour

Multimodal Chain-of-Thought Reasoning in Language Models

amazon-science/mm-cot 2 Feb 2023

By incorporating the vision features in both stages, the model is able to generate effective rationales that contribute to answer inference.

312
1.73 stars / hour

Mixture of Diffusers for scene composition and high resolution image generation

albarji/mixture-of-diffusers 5 Feb 2023

Diffusion methods have been proven to be very effective to generate images while conditioning on a text prompt.

Image Generation

83
1.71 stars / hour

Hard Prompts Made Easy: Gradient-Based Discrete Optimization for Prompt Tuning and Discovery

YuxinWenRick/hard-prompts-made-easy 7 Feb 2023

In the text-to-image setting, the method creates hard prompts for diffusion models, allowing API users to easily generate, discover, and mix and match image concepts without prior knowledge on how to prompt the model.

132
1.71 stars / hour

BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models

salesforce/lavis 30 Jan 2023

The cost of vision-and-language pre-training has become increasingly prohibitive due to end-to-end training of large-scale models.

Image Captioning Image Retrieval +5

2,106
1.23 stars / hour

Discovering Symbolic Models from Deep Learning with Inductive Biases

glouppe/info8010-deep-learning NeurIPS 2020

The technique works as follows: we first encourage sparse latent representations when we train a GNN in a supervised setting, then we apply symbolic regression to components of the learned model to extract explicit physical relations.

Symbolic Regression

924
0.97 stars / hour

Top-Down Beats Bottom-Up in 3D Instance Segmentation

samsunglabs/td3d 6 Feb 2023

Most 3D instance segmentation methods exploit a bottom-up strategy, typically including resource-exhaustive post-processing.

 Ranked #1 on 3D Instance Segmentation on S3DIS (using extra training data)

3D Instance Segmentation Semantic Segmentation

41
0.95 stars / hour

ClimaX: A foundation model for weather and climate

microsoft/ClimaX 24 Jan 2023

Most state-of-the-art approaches for weather and climate modeling are based on physics-informed numerical models of the atmosphere.

Self-Supervised Learning Weather Forecasting

33
0.83 stars / hour

Dual PatchNorm

lucidrains/musiclm-pytorch 2 Feb 2023

We propose Dual PatchNorm: two Layer Normalization layers (LayerNorms), before and after the patch embedding layer in Vision Transformers.

1,182
0.77 stars / hour