ucyang (Unchun Yang)

upvoted a paper about 16 hours ago

A Technical Report for Polyglot-Ko: Open-Source Large-Scale Korean Language Models

Paper • 2306.02254 • Published Jun 4, 2023 • 10

upvoted a collection about 18 hours ago

Instruction Pre-Training

Collection

8 items • Updated 8 days ago • 23

upvoted an article about 24 hours ago

Article

Cosmopedia: how to create large-scale synthetic data for pre-training Large Language Models

Mar 20

• 31

upvoted an article 1 day ago

Article

Ethics and Society Newsletter #6: Building Better AI: The Importance of Data Quality

5 days ago

• 17

upvoted 2 collections 1 day ago

LLM Compiler

Collection

Meta LLM Compiler is a state-of-the-art LLM that builds upon Code Llama with improved performance for code optimization and compiler reasoning. • 4 items • Updated 1 day ago • 97

Gemma 2 Release

Collection

10 items • Updated 1 day ago • 80

upvoted an article 1 day ago

Article

Decoding GPT-4'o': In-Depth Exploration of Its Mechanisms and Creating Similar AI.

By

•

May 21

• 26

upvoted a paper 2 days ago

Multimodal Needle in a Haystack: Benchmarking Long-Context Capability of Multimodal Large Language Models

Paper • 2406.11230 • Published 12 days ago • 34

upvoted 2 papers 3 days ago

LLaNA: Large Language and NeRF Assistant

Paper • 2406.11840 • Published 12 days ago • 17

VideoLLM-online: Online Video Large Language Model for Streaming Video

Paper • 2406.11816 • Published 12 days ago • 20

upvoted 2 collections 3 days ago

Cambrian Data

Collection

3 items • Updated 4 days ago • 6

Cambrian-1 Models

Collection

6 items • Updated 1 day ago • 14

upvoted 7 papers 3 days ago

Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs

Paper • 2406.16860 • Published 5 days ago • 45

BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions

Paper • 2406.15877 • Published 7 days ago • 39

DataComp-LM: In search of the next generation of training sets for language models

Paper • 2406.11794 • Published 12 days ago • 39

MMDU: A Multi-Turn Multi-Image Dialog Understanding Benchmark and Instruction-Tuning Dataset for LVLMs

Paper • 2406.11833 • Published 12 days ago • 60

upvoted 2 articles 3 days ago

Article

Fine-tuning Florence-2 - Microsoft's Cutting-edge Vision Language Models

5 days ago

• 106

Article

Data Is Better Together: A Look Back and Forward

9 days ago

• 13

upvoted 5 papers 4 days ago

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

Paper • 2310.16656 • Published Oct 25, 2023 • 39

Phased Consistency Model

Paper • 2405.18407 • Published May 28 • 44

Chameleon: Mixed-Modal Early-Fusion Foundation Models

Paper • 2405.09818 • Published May 16 • 110

Mixture-of-Agents Enhances Large Language Model Capabilities

Paper • 2406.04692 • Published 22 days ago • 49

CRAG -- Comprehensive RAG Benchmark

Paper • 2406.04744 • Published 22 days ago • 37

upvoted a collection 5 days ago

Moirai-1.0-R models

Collection

6 items • Updated 6 days ago • 24

upvoted an article 5 days ago

Article

BM25 for Python: Achieving high performance while simplifying dependencies with BM25S⚡

By

•

3 days ago

• 22

upvoted a paper 7 days ago

Instruction Pre-Training: Language Models are Supervised Multitask Learners

Paper • 2406.14491 • Published 9 days ago • 74

upvoted an article 8 days ago

Article

Enhancing Image Model Dreambooth Training Through Effective Captioning: Key Observations

By

•

10 days ago

• 11

upvoted an article 9 days ago

Article

Goodbye cold boot - how we made LoRA inference 300% faster

Dec 5, 2023

• 1

upvoted a collection 9 days ago

MosaicBERT

Collection

A collection of BERT-based models of different sequence lengths trained on the C4 dataset. Details: https://mosaicbert.github.io/ • 5 items • Updated Dec 27, 2023 • 4

upvoted 4 papers 9 days ago

GEB-1.3B: Open Lightweight Large Language Model

Paper • 2406.09900 • Published 15 days ago • 18

Glyph-ByT5-v2: A Strong Aesthetic Baseline for Accurate Multilingual Visual Text Rendering

Paper • 2406.10208 • Published 15 days ago • 21

OmniCorpus: A Unified Multimodal Corpus of 10 Billion-Level Images Interleaved with Text

Paper • 2406.08418 • Published 17 days ago • 28

Needle In A Multimodal Haystack

Paper • 2406.07230 • Published 18 days ago • 51

upvoted an article 10 days ago

Article

BigCodeBench: Benchmarking Large Language Models on Solving Practical and Challenging Programming Tasks

11 days ago

• 30

upvoted a paper 10 days ago

Florence-2: Advancing a Unified Representation for a Variety of Vision Tasks

Paper • 2311.06242 • Published Nov 10, 2023 • 65

upvoted a collection 10 days ago

Florence

Collection

9 items • Updated 14 days ago • 127

upvoted a paper 10 days ago

How Do Large Language Models Acquire Factual Knowledge During Pretraining?

Paper • 2406.11813 • Published 12 days ago • 28

upvoted an article 10 days ago

Article

From DeepSpeed to FSDP and Back Again with Hugging Face Accelerate

16 days ago

• 24

upvoted a collection 12 days ago

DeepSeek-VL

Collection

DeepSeek-VL model series • 4 items • Updated Mar 14 • 10

upvoted a paper 12 days ago

VideoLLaMA 2: Advancing Spatial-Temporal Modeling and Audio Understanding in Video-LLMs

Paper • 2406.07476 • Published 18 days ago • 30

upvoted a collection 14 days ago

Nemotron 4 340B

Collection

Nemotron-4: open models for Synthetic Data Generation (SDG). Includes Base, Instruct, and Reward models. • 4 items • Updated 15 days ago • 145

upvoted a collection 15 days ago

Magpie-Pro

Collection

Dataset built with Meta Llama 3 70B. Models are fine-tuned from Llama 3 8B. • 8 items • Updated 13 days ago • 14

upvoted 2 papers 16 days ago

Accessing GPT-4 level Mathematical Olympiad Solutions via Monte Carlo Tree Self-refine with LLaMa-3 8B

Paper • 2406.07394 • Published 18 days ago • 16

VALL-E 2: Neural Codec Language Models are Human Parity Zero-Shot Text to Speech Synthesizers

Paper • 2406.05370 • Published 21 days ago • 12

upvoted 4 collections 17 days ago

Yi-1.5 (2024/05)

Collection

10 items • Updated May 20 • 82

Gemma release

Collection

Groups the Gemma models released by the Google team. • 40 items • Updated 1 day ago • 317

BERT release

Collection

Regroups the original BERT models released by the Google team. Except for the models marked otherwise, the checkpoints support English. • 8 items • Updated 1 day ago • 17

RecurrentGemma Release

Collection

8 items • Updated 1 day ago • 37

upvoted an article 17 days ago

Article

Space secrets security update

29 days ago

• 50

upvoted an article 18 days ago

Article

BrAIn: next generation neurons?

By

•

23 days ago

• 15

upvoted 3 papers 19 days ago

BitsFusion: 1.99 bits Weight Quantization of Diffusion Model

Paper • 2406.04333 • Published 23 days ago • 36

Parrot: Multilingual Visual Instruction Tuning

Paper • 2406.02539 • Published 25 days ago • 35

Audio Mamba: Bidirectional State Space Model for Audio Representation Learning

Paper • 2406.03344 • Published 24 days ago • 15

upvoted a collection 19 days ago

LLaVA-NeXT-Video

Collection

Some powerful video models. • 5 items • Updated 3 days ago • 25

upvoted a paper 19 days ago

MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmark

Paper • 2406.01574 • Published 26 days ago • 42

upvoted a collection 20 days ago

MAmmoTH2

Collection

Scaling up instruction data from the web for to build better LLMs • 11 items • Updated May 26 • 7

upvoted a paper 21 days ago

Show, Don't Tell: Aligning Language Models with Demonstrated Feedback

Paper • 2406.00888 • Published 26 days ago • 29

Unchun Yang

AI & ML interests

Organizations

ucyang's activity

Cosmopedia: how to create large-scale synthetic data for pre-training Large Language Models

Ethics and Society Newsletter #6: Building Better AI: The Importance of Data Quality

Decoding GPT-4'o': In-Depth Exploration of Its Mechanisms and Creating Similar AI.

Fine-tuning Florence-2 - Microsoft's Cutting-edge Vision Language Models

Data Is Better Together: A Look Back and Forward

BM25 for Python: Achieving high performance while simplifying dependencies with *BM25S*⚡

Enhancing Image Model Dreambooth Training Through Effective Captioning: Key Observations

Goodbye cold boot - how we made LoRA inference 300% faster

BigCodeBench: Benchmarking Large Language Models on Solving Practical and Challenging Programming Tasks

From DeepSpeed to FSDP and Back Again with Hugging Face Accelerate

Space secrets security update

BrAIn: next generation neurons?

BM25 for Python: Achieving high performance while simplifying dependencies with BM25S⚡