AIGC周刊 | 第27期

时间: 2023.5.8-2023.5.14

本周大事记

1. Meta发布可跨越六种感官的大模型ImageBind,已开源

论文: dl.fbaipublicfiles.com

代码: github.com

read more

2. Google 发布大模型PaLM的更新版本PaLM2,部分任务超越GPT-4

介绍:blog.google

技术报告: ai.google

read more

最新技术:

  1. Cognitive Reframing of Negative Thoughts through Human-Language Model Interaction

    论文:arxiv.org

  2. Avatar Fingerprinting for Authorized Use of Synthetic Talking-Head Videos

    论文: arxiv.org

    主页: research.nvidia.com

  3. A Suite of Generative Tasks for Multi-Level Multimodal Webpage Understanding 论文: arxiv.org

  4. Composite Motion Learning with Task Control

    论文: arxiv.org

    代码: github.com

  5. Residual Prompt Tuning: Improving Prompt Tuning with Residual Reparameterization

    论文: arxiv.org

    代码: github.com

  6. X-LLM: Bootstrapping Advanced Large Language Models by Treating Multi-Modalities as Foreign Languages

    论文: arxiv.org

  7. Controllable Light Diffusion for Portraits

    论文: arxiv.org

  8. Locally Attentional SDF Diffusion for Controllable 3D Shape Generation

    论文: arxiv.org

  9. FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performance

    论文: arxiv.org

  10. Code Execution with Pre-trained Language Models a

    论文: arxiv.org

  11. 1.Sketching the Future (STF): Applying Conditional Control Techniques to Text-to-Video Models

    论文: arxiv.org

    主页:sketchingthefuture.github.io

  12. HumanRF: High-Fidelity Neural Radiance Fields for Humans in Motion

    论文:arxiv.org

    主页: synthesiaresearch.github.io

  13. Generative AI meets 3D: A Survey on Text-to-3D in AIGC Era

    论文:arxiv.org

  14. Do LLMs Understand User Preferences? Evaluating LLMs On User Rating Prediction

    论文: arxiv.org

  15. EfficientViT: Memory Efficient Vision Transformer with Cascaded Group Attention

    论文: arxiv.org

    代码: github.com

  16. Simple Token-Level Confidence Improves Caption Correctness

    论文: arxiv.org

  17. Not All Languages Are Created Equal in LLMs: Improving Multilingual Capability by Cross-Lingual-Thought Prompting

    论文: arxiv.org

  18. InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning

    论文: arxiv.org

  19. Chain-of-Dictionary Prompting Elicits Translation in Large Language Models

    论文: arxiv.org

  20. Bot or Human? Detecting ChatGPT Imposters with A Single Question

    论文: arxiv.org

  21. V2Meow: Meowing to the Visual Beat via Music Generation

    论文: arxiv.org

  22. Exploiting Diffusion Prior for Real-World Image Super-Resolution

    论文: arxiv.org

    主页: iceclear.github.io

  23. TMR: Text-to-Motion Retrieval Using Contrastive 3D Human Motion Synthesis

    主页: mathis.petrovich.fr

  24. ArtGPT-4: Artistic Vision-Language Understanding with Adapter-enhanced MiniGPT-4

    论文: arxiv.org

    模型: huggingface.co

  25. Towards best practices in AGI safety and governance: A survey of expert opinion

    论文: arxiv.org

  26. Can ChatGPT Forecast Stock Price Movements? Return Predictability and Large Language Models

    论文:papers.ssrn.com

  27. Language models can explain neurons in language models

    主页:openai.com

  28. MEGABYTE: Predicting Million-byte Sequences with Multiscale Transformers 论文: arxiv.org

  29. 巨型AI模型背后的分布式训练技术

    1. zhuanlan.zhihu.com

    2. zhuanlan.zhihu.com

  30. 涌现:21 世纪科学的统一主题

    mp.weixin.qq.com

课程:

  1. LLM Bootcamp – Spring 2023

    fullstackdeeplearning.com

  2. Introduction to Data-Centric AI

    dcai.csail.mit.edu

  3. huggingface blog更新:

    1. Text-to-Video: The Task, Challenges and the Current State

      huggingface.co

      中文版:mp.weixin.qq.com

    2. Run a Chatgpt-like Chatbot on a Single GPU with ROCm

      huggingface.co

    3. Assisted Generation: a new direction toward low-latency text generation

      huggingface.co

商业:

  1. 海外大数据统计至少有4个基于gpt的聊天应用,月收入超过100万美元

    twitter.com

  2. Midjourney 官方中文版 · 内测申请

    mp.weixin.qq.com

  3. 陆奇最新演讲全文实录、完整PPT和视频:大模型带来的新范式

    mp.weixin.qq.com

案例:

  1. Open LLM Leaderboard

    huggingface.co

  2. threestudio: 创建3D内容的统一框架

    github.com

  3. ControlNet大更新:仅靠提示词就能精准P图,保持画风不变

    mp.weixin.qq.com

  4. huggingface 发布Transformers Agent

    huggingface.co

source
The article is sourced from the internet. Click the “Source” button to view the original content. If there is any copyright infringement, please contact our team for removal.

Share this article
Shareable URL
Prev Post

How to start

Next Post

Use this tactic to drive revenue on your SEO campaigns by Cynthia Ramsaran

Read next

AIGC Weekly | #46

Top Papers of the week(Dec 11 – Dec 17) 1.Mathematical discoveries from program search with large language…
Subscribe to our newsletter
Get notified of the best deals on our WordPress themes.