DataDreamer: A Tool for Synthetic Data Generation and Reproducible LLM Workflows

datadreamer-dev/datadreamer 16 Feb 2024

The rapid rise to prominence of these models and these unique challenges has had immediate adverse impacts on open science and on the reproducibility of work that uses them.

Synthetic Data Generation

528
0.67 stars / hour

MambaIR: A Simple Baseline for Image Restoration with State-Space Model

csguoh/mambair 23 Feb 2024

Recent years have witnessed great progress in image restoration thanks to the advancements in modern deep neural networks e. g. Convolutional Neural Network and Transformer.

Image Restoration

74
0.67 stars / hour

The Unreasonable Effectiveness of Eccentric Automatic Prompts

stanfordnlp/dspy 9 Feb 2024

Given the combinatorial complexity, and thus computation time, of experimenting with hand-tuning prompts for large black-box models, we then compared the performance of the best "positive thinking" prompt against the output of systematic prompt optimization.

GSM8K

7,550
0.64 stars / hour

Coarse-to-Fine Latent Diffusion for Pose-Guided Person Image Synthesis

YanzuoLu/CFLD 28 Feb 2024

While existing methods simply align the person appearance to the target pose, they are prone to overfitting due to the lack of a high-level semantic understanding on the source person image.

Pose-Guided Image Generation

59
0.63 stars / hour

SC-GS: Sparse-Controlled Gaussian Splatting for Editable Dynamic Scenes

yihua7/SC-GS 4 Dec 2023

During learning, the location and number of control points are adaptively adjusted to accommodate varying motion complexities in different regions, and an ARAP loss following the principle of as rigid as possible is developed to enforce spatial continuity and local rigidity of learned motions.

Novel View Synthesis

202
0.62 stars / hour

What Makes Good In-Context Examples for GPT-$3$?

stanfordnlp/dsp 17 Jan 2021

Inspired by the recent success of leveraging a retrieval module to augment large-scale neural network models, we propose to retrieve examples that are semantically-similar to a test sample to formulate its corresponding prompt.

Few-Shot Learning Natural Language Understanding +4

7,445
0.61 stars / hour

FuseChat: Knowledge Fusion of Chat Models

fanqiwan/fusellm 25 Feb 2024

Recently, \textsc{FuseLLM} introduced the concept of knowledge fusion to transfer the collective knowledge of multiple structurally varied LLMs into a target LLM through lightweight continual training.

233
0.59 stars / hour

Long-Context Language Modeling with Parallel Context Encoding

princeton-nlp/cepe 26 Feb 2024

We further introduce a CEPE variant that can extend the context window of instruction-tuned models with only unlabeled data, and showcase its effectiveness on LLAMA-2-CHAT, leading to a strong instruction-following model that can leverage very long context on downstream tasks.

In-Context Learning Instruction Following +1

64
0.55 stars / hour

Retrieval-Augmented Generation for AI-Generated Content: A Survey

hymie122/rag-survey 29 Feb 2024

Furthermore, we introduce the benchmarks for RAG, discuss the limitations of current RAG systems, and suggest potential directions for future research.

Information Retrieval Retrieval

67
0.55 stars / hour

YOLO-World: Real-Time Open-Vocabulary Object Detection

ailab-cvc/yolo-world 30 Jan 2024

The You Only Look Once (YOLO) series of detectors have established themselves as efficient and practical tools.

Instance Segmentation Language Modelling +4

2,509
0.53 stars / hour