Wenjie Zheng @wjzheng_nlp

🚨New paper!🚨 Still worried about the low quality of your rule-cleaned pre-training corpora? Try 🫐 ProX! 1. Dramatically boosts pre-training corpus quality with a language model that generates executable programs. 2. A 1.7B model, trained on corpus refined by 🫐 ProX with…

Fan Zhou @FaZhou_998

a year ago

🚨New paper!🚨 Still worried about the low quality of your rule-cleaned pre-training corpora? Try 🫐 ProX! 1. Dramatically boosts pre-training corpus quality with a language model that generates executable programs. 2. A 1.7B model, trained on corpus refined by 🫐 ProX with… https://t.co/CQlI9lbuuf

6 39 192 63K 122

Download Video

[CV] CinePile: A Long Video Question Answering Dataset and Benchmark arxiv.org/abs/2405.08813 - The paper introduces CinePile, a large-scale video question answering dataset with ~305k questions covering temporal comprehension, human-object interactions, reasoning about…

keyboard_arrow_left Previous keyboard_arrow_right Next

work full of sincerity, welcome everyone to follow. 🥳🥳🥳

Qiming Xie @grayground_x

a year ago

work full of sincerity, welcome everyone to follow. 🥳🥳🥳

1 2 15 2K 4

Excited to share our #ACL2024 Findings paper "EmpathicStories++: A Multimodal Dataset for Empathy towards Personal Experiences" 🧵(1/7) Dataset request: mitmedialab.github.io/empathic-stori…

Introducing MetaGPT's Data Interpreter: Open Source and Better "Devin". Data Interpreter has achieved state-of-the-art scores in machine learning, mathematical reasoning, and open-ended tasks, and can analyze stocks, imitate websites, and train models. Data Interpreter is an…

keyboard_arrow_left Previous keyboard_arrow_right Next

A Survey on Data Selection for Language Models Presents a comprehensive review of existing literature on data selection methods and related research areas, providing a taxonomy of existing approaches arxiv.org/abs/2402.16827

🔴 stream: YOLO-World Q&A + coding in less than 15 minutes, I start my first YT stream; I'll be talking about YOLO-World and answering your questions that you left under my last YT video stop by to say hello link: youtube.com/live/lF1BtQL16… ↓ some of the topics we will cover

Thrilled to release our new version (v0.2) of MathPile, a cleaner version through our efforts to fix some issues!🥳 More importantly, we also released commercial-use version, namely MathPile_Commercial(huggingface.co/datasets/GAIR/…)🥳🥳🥳

Aran Komatsuzaki @arankomatsuzaki

2 years ago

Thrilled to release our new version (v0.2) of MathPile, a cleaner version through our efforts to fix some issues!🥳 More importantly, we also released commercial-use version, namely MathPile_Commercial(huggingface.co/datasets/GAIR/…)🥳🥳🥳

6 76 397 50K 250

Download Image

Quadratic attention has been indispensable for information-dense modalities such as language... until now. Announcing Mamba: a new SSM arch. that has linear-time scaling, ultra long context, and most importantly--outperforms Transformers everywhere we've tried. With @tri_dao 1/

A team just made OpenAI Whisper 6x faster, 49% smaller, while keeping 99% of the accuracy. The model is already available on the HuggingFace Transformers library: model_id = "distil-whisper/distil-large-v2" You can also use their web UI to transcribe from URLs, files, or…

How to #RLHF for LLMs: #PPO or #DPO? Introducing #BPO (black-box prompt optimization) to align LLMs without model training. 1) ChatGPT + BPO > ChatGPT 2) GPT-4 + BPO > GPT-4 3) Vicuna + BPO > Vicuna + PPO/DPO 4) Vicuna + DPO + BPO > Vicuna + DPO arxiv.org/pdf/2311.04155…

mPLUG-Owl2: Revolutionizing Multi-modal Large Language Model with Modality Collaboration paper page: huggingface.co/papers/2311.04… Multi-modal Large Language Models (MLLMs) have demonstrated impressive instruction abilities across various open-ended tasks. However, previous methods…

TEAL: Tokenize and Embed ALL for Multi-modal Large Language Models paper page: huggingface.co/papers/2311.04… Despite Multi-modal Large Language Models (MM-LLMs) have made exciting strides recently, they are still struggling to efficiently model the interactions among multi-modal…

🚨 Unveiling GPT-4V(ision)'s mind! We're breaking down how even the brightest Visual Language Models get it wrong! With our new 'Bingo' benchmark, we shed light on the two common types of hallucinations in GPT-4V(ision): bias and interference. Led by @cuichenhang @AiYiyangZ

CogVLM: Visual Expert for Pretrained Language Models paper page: huggingface.co/papers/2311.03… introduce CogVLM, a powerful open-source visual language foundation model. Different from the popular shallow alignment method which maps image features into the input space of language…

OtterHD: A High-Resolution Multi-modality Model paper page: huggingface.co/papers/2311.04… present OtterHD-8B, an innovative multimodal model evolved from Fuyu-8B, specifically engineered to interpret high-resolution visual inputs with granular precision. Unlike conventional models…

We're rolling out new features and improvements that developers have been asking for: 1. Our new model GPT-4 Turbo supports 128K context and has fresher knowledge than GPT-4. Its input and output tokens are respectively 3× and 2× less expensive than GPT-4. It’s available now to…

FlashDecoding++: Faster Large Language Model Inference on GPUs paper page: huggingface.co/papers/2311.01… As the Large Language Model (LLM) becomes increasingly important in various domains. However, the following challenges still remain unsolved in accelerating LLM inference: (1)…

LLaVA-Interactive: An All-in-One Demo for Image Chat, Segmentation, Generation and Editing paper page: huggingface.co/papers/2311.00… LLaVA-Interactive is a research prototype for multimodal human-AI interaction. The system can have multi-turn dialogues with human users by taking…

MM-VID: Advancing Video Understanding with GPT-4V(ision) paper page: huggingface.co/papers/2310.19… present MM-VID, an integrated system that harnesses the capabilities of GPT-4V, combined with specialized tools in vision, audio, and speech, to facilitate advanced video understanding.…