theblackcat102

LLaVA

[NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.

nougat

Implementation of Nougat Neural Optical Understanding for Academic Documents

LightGlue

LightGlue: Local Feature Matching at Light Speed (ICCV 2023)

CTranslate2

Fast inference engine for Transformer models

Medusa

Medusa: Simple Framework for Accelerating LLM Generation with Multiple Decoding Heads

Detic

Code release for "Detecting Twenty-thousand Classes using Image-level Supervision".

resource-stream

CUDA related news and material links

awesome-mixture-of-experts

A collection of AWESOME things about mixture-of-experts

CMMLU

CMMLU: Measuring massive multitask language understanding in Chinese

cleora

Cleora AI is a general-purpose model for efficient, scalable learning of stable and inductive entity embeddings for heterogeneous relational data.

adept-inference

Inference code for Persimmon-8B

self-correction-llm-papers

This is a collection of research papers for Self-Correcting Large Language Models with Automated Feedback.

CLIP-SAM

Experiment on combining CLIP with SAM to do open-vocabulary image segmentation.

qmoe

Code for the paper "QMoE: Practical Sub-1-Bit Compression of Trillion-Parameter Models".

neurips_llm_efficiency_challenge

NeurIPS Large Language Model Efficiency Challenge: 1 LLM + 1GPU + 1Day

ModuleFormer

ModuleFormer is a MoE-based architecture that includes two different types of experts: stick-breaking attention heads and feedforward experts. We released a collection of ModuleFormer-based Language Models (MoLM) ranging in scale from 4 billion to 8 billion parameters.

ToolkenGPT

ToolkenGPT: Augmenting Frozen Language Models with Massive Tools via Tool Embeddings - NeurIPS 2023 (oral)

TransformerPrograms

[NeurIPS 2023] Learning Transformer Programs

ml-calibration

relplot: Utilities for measuring calibration and plotting reliability diagrams

SemDeDup

Code for "SemDeDup", a simple method for identifying and removing semantic duplicates from a dataset (data pairs which are semantically similar, but not exactly identical).

barrel-rec-pytorch

Bingo

GPU-Puzzles

Solve puzzles. Learn CUDA.

ievals

Official github repo for TMMLU+, Large scale traditional chinese massive multitask language understanding

hades

Fast singularity detection with kernel

Beyond-Neural-Scaling

Implementation of Beyond Neural Scaling beating power laws for deep models and prototype-based models

arc-agents

Experiments with LLMs on the Abstraction and Reasoning Corpus (ARC)

SFABD

Semantic Fusion Augmentation and Semantic Boundary Detection: A Novel Approach to Multi-Target Video Moment Retrieval (SFABD)

pscan_kernel

json-schema-corpus

Corpus of over 80thousand JSON Schema documents, collected from open source GitHub repositories.