October 17-18, 2024

Conference Schedule

Join us for six exciting conference tracks, with full agenda, session times, topics, and speakers coming soon. Don't miss out on these dynamic sessions! (More sessions continuously added)

Rust Language Track: Driving the Future of Safe and Performant Development

The Rust Language Track brings together key contributors from the Rust ecosystem, including experts in the language itself, compiler, developer tools, and members of the technical leadership council. This track offers an in-depth look at Rust from both a technical and community perspective, featuring insights from:

  • Major Rust project contributors, sharing the latest advancements and improvements in the Rust ecosystem
  • Real-world Rust application developers, who will discuss their experiences building performant and safe applications with Rust
  • Academic research, highlighting Rust’s safety guarantees and innovative language features that set it apart in modern programming
  • Project leaders, who will outline Rust’s future roadmap and vision, sharing what’s next for this rapidly growing language

Whether you're a seasoned Rust developer or new to the language, this track provides a unique opportunity to engage with the people driving Rust’s continued success and innovation.

Don’t miss out on this chance to explore the future of Rust and its impact on software development!

8:30
Registration
Time
8:30
Speakers
Time
8:30
9:30
GOSIM Keynote: Open Source Ecosystem in the Era of AGI
Time
9:30
Speakers
Tao Jiang
Open Source Ecosystem in the era of AGI... More Details Soon
Time
9:30
9:45
Opening Keynote: Mobile Solution by OpenHarmony
Time
9:45
Speakers
Yan Zhou
Huawei promotes the openness, concurrency, and security of mobile devices operating systems based on OpenHarmony, and combines the lightweight characteristics of devices to provide the industry with more effective parallel and concurrency solutions.
Time
9:45
10:05
GOSIM Hackathon Awards Ceremony
Time
10:05
Speakers
Winner of the GOSIM Hackathon will be announced.
Time
10:05
10:10
Morning Break
Time
10:10
Speakers
20 Minutes Break
Time
10:10
10:30
Register Allocation in Cranelift
Time
10:30
Register allocation is one of the most important stages in an optimizing compiler and is often responsible for a large fraction of the compilation time. This talk presents the register allocator used by Cranelift, a fast compiler backend written in Rust and used by the Rust compiler, with a focus on regalloc3, its newest iteration that is currently being developed.
Time
10:30
11:10
Through the Eyes of a Maintainer: How to Make Your Contribution More Successful
Time
11:10
Speakers
Ed Page
Whether its to the Rust Project, your favorite Rust tool, or a library you use, your first contribution can be intimidating. You are stepping into an area where you don't know the social dynamics or technical expectations. I'll be stepping you through the process of contributing from the maintainer's perspective so you can better understand how this works and how to make it more likely your contribution is both accepted and quickly.
Time
11:10
11:50
Rust Compiler Overview
Time
11:50
This talk will provide an overview of the Rust compiler: what it looks like, how it works, how its internals are structured, and how it is developed.
Time
11:50
12:30
Lunch Break
Time
12:30
Speakers
Time
12:30
14:00
Writing Your Own Rust Linter
Time
14:00
This talk will show you how to write your own rust linter using the rustc_tools crate which allows you to have access to the rust compiler API. It can be quite useful if you need specific lints for a project and cannot add them to clippy. It will also explain how the compiler API works and give a small example at the end showing it in action.
Time
14:00
14:40
Life as a Rust Project Contributor
Time
14:40
Join Jane Losare-Lusby as she takes you on a journey through her experiences as a contributor to the Rust project. Starting with her early work on Clippy, Jane has navigated various roles, including error-handling project group lead, member of the Library and Style teams, and as a board member of the Rust Foundation. After a brief hiatus to work on Truffle and June, she returned to contribute to the Rust Compiler team, where she is now focused on the development of locally stored metrics. In this talk, Jane will share personal insights on navigating challenges such as impostor syndrome, conflict resolution, burnout, and isolation. Jane will offer practical advice for newcomers looking to contribute to Rust, highlighting the importance of mentorship and connection, the dynamics of various Rust teams, and how to turn an idea—like her current metrics initiative—into a meaningful contribution. Whether you're a seasoned developer or just getting started, this talk will provide valuable perspectives on how to find your place and make an impact in the Rust community.
Time
14:40
15:20
Afternoon Break
Time
15:20
Speakers
30 Minutes Break
Time
15:20
15:50
Fireside Chat
Time
15:50
A cozy interview and Q&A with Dr. Josh Triplett, from the Rust Project language, library, and Cargo teams.
Time
15:50
16:30
Low-Level: Rust Education's "Next Big Thing"
Time
16:30
Speakers
Bart Massey
The Rust community has made a good start at teaching tools for introductory Rust. I will review the state of Rust education, especially in Universities, and talk about ways in which Rust-Edu and other groups hope to fill new needs in the low-level space.
Time
16:30
17:10
Typical Use of Rust Programs in Different Linking Ways in Trading Systems
Time
17:10
Speakers
Qiao Dan
This talk will delve into the fundamental concepts of workspace, target, and crate-type. It will explore three cases within a trading system, highlighting code dependencies and static links using -crate-type=lib, dynamic link with -crate-type=cdylib, and static link using -crate-type=rlib. Furthermore, the talk will expand on important aspects such as the performance differences between dynamic and static links and the trade-offs in terms of toolchain version flexibility when using rlib.
Time
17:10

October 18

Rust Programming Language
Time
Session
9:30
Ferrocene - The present and future of Rust in Safety
Time
9:30
Rust is a relative new language - released in 2015, it has taken the software programming word surprisingly swiftly. It is surprising that - not even 10 years later - we have ongoing projects and initiatives in the Functional Safety/SDV space. Rust is even seen as a front runner of open source safety initiatives. How did that happen? This talk provides an overview of currently ongoing initiatives, an assessment of challenges and an outlook into the future.
Time
9:30
10:10
How Different is Rust Code in the Linux Kernel?
Time
10:10
Speakers
Gary Guo
The Linux kernel is very different from ordinary software projects, and therefore you shouldn't expect kernel Rust code to be similar to the Rust code that you're familiar with! In this talk I'll go through the basic primitives in Rust for Linux, compare them with userspace Rust, and dive into rationales that lead to their design.
Time
10:10
10:50
Morning Break
Time
10:50
Speakers
20 Minutes Break
Time
10:50
11:10
An Outsider's Guide to the Rust Project
Time
11:10
Speakers
Jack Huey
The Rust project has grown both in members and in structure over the past years. This talk aims to give a bird's eye view of the teams and structure of the people that make up the Rust Project today, how the teams communicate and make collaborative decisions, and how new people and groups can get involved.
Time
11:10
11:50
Building Efficient and Scalable Distributed Systems: Implementing Zenoh with Rust for V2V, Autonomous Systems, and Robotics
Time
11:50
Speakers
Yuyuan Yuan
The need for efficient and scalable distributed systems is critical in the rapidly advancing fields of Vehicle-to-Vehicle (V2V) communication, autonomous systems, and robotics. Zenoh, an innovative protocol that integrates data in motion and data at rest, paired with the power of Rust, offers a compelling solution to these challenges. This talk will explore how to effectively implement Zenoh using Rust, unlocking new possibilities in edge computing, IoT, and real-time applications.
Time
11:50
12:30
Lunch Break
Time
12:30
Speakers
Time
12:30
14:00
Current Status, Challenges and Key Technologies on Rust OSes
Time
14:00
Speakers
Yu Chen
We Analyze the current development status and challenges faced by operating system kernels based on the Rust language. We also introduce the development history of an operating system designed and implemented in Rust by the Operating System Laboratory at Tsinghua University, as well as the current work on a component-based operating system kernel. The discussion also covers the difficulties encountered in designing and implementing a componentized operating system kernel, along with the technical attempts we are making. Additionally, we present our efforts in organizing operating system kernel competitions within Chinese universities and conducting open-source operating system bootcamps.
Time
14:00
14:40
Development Practice of Rust Language in vivo BlueOS
Time
14:40
Speakers
Shuang Chen
Vivo BlueOS is developed using Rust, a memory-safe systems programming language renowned for its high concurrency, strong security, and high efficiency. Rust provides a more stable and efficient runtime environment for operating systems. By leveraging Rust, BlueOS achieves smoother performance when handling complex tasks. This presentation will demonstrate how BlueOS implements core system services, such as graphics and multimedia, using the Rust language.
Time
14:40
15:20
Afternoon Break
Time
15:20
Speakers
30 Minutes Break
Time
15:20
15:50
How Rust Makes Open Source Embedded Drivers Easy
Time
15:50
Adding an existing open source driver for your embedded peripheral is pretty awful in C. Rust makes open source embedded drivers easy thanks to the language itself, and the Rust Embedded ecosystem.
Time
15:50
16:30
Leveraging Rust's Strengths in WasmTime
Time
16:30
The Rust programming language is well known for its safety guarantees, and even when a project is forced to use `unsafe` code it has many benefits to offer. This talk will dive into the Wasmtime project, a standalone WebAssembly engine, and how it leverages Rust's strengths to provide critical safety guarantees.
Time
16:30

App Development Track: The Future of Cross-Platform Innovation

Discover the most popular open-source frameworks and dive into the latest technological advancements and trends in app development! In the App Development Track, we’ll focus on: 

  • Flutter, React Native, Kotlin Multiplatform, Cangjie, and OpenHarmony, exploring innovative practices for building cross-platform applications
  • The latest progress of WebAssembly, offering unprecedented performance and flexibility in application development
  • Full-stack frameworks based on the Rust language, such as Makepad and Robius, revealing the opportunities and challenges of using Rust as a development language
  • How Generative AI is revolutionizing development workflows, enhancing creativity and efficiency
  • Valuable insights from super app developers, unlocking key strategies for building the future of applications


Whether you're an experienced developer or just starting, this track will provide in-depth analysis of cutting-edge technologies and practical insights to help you stay ahead in the fast-evolving world of app development!

Don’t miss the opportunity to explore the future of app development with industry leaders and innovators!

October 17

App Development
Time
Session
8:30
Registration
Time
8:30
Speakers
Time
8:30
9:30
GOSIM Keynote: Open Source Ecosystem in the Era of AGI
Time
9:30
Speakers
Tao Jiang
Open Source Ecosystem in the era of AGI... More Details Soon
Time
9:30
9:45
Opening Keynote: Mobile Solution by OpenHarmony
Time
9:45
Speakers
Yan Zhou
Huawei promotes the openness, concurrency, and security of mobile devices operating systems based on OpenHarmony, and combines the lightweight characteristics of devices to provide the industry with more effective parallel and concurrency solutions.
Time
9:45
10:05
GOSIM Hackathon Awards Ceremony
Time
10:05
Speakers
Winner of the GOSIM Hackathon will be announced.
Time
10:05
10:10
Morning Break
Time
10:10
Speakers
20 Minutes Break
Time
10:10
10:30
Next Generation App Development with Dioxus: Fullstack, Cross-Platform, Strongly Typed, and Blazingly Fast
Time
10:30
Dioxus is a Rust framework that makes it makes it easy for developers to ship web, desktop, and mobile apps with a single unified codebase. In this talk, we’ll be unveiling features like native mobile support, live hot-reloading, and a custom AI model.
Time
10:30
11:10
Flutter: Google’s UI toolkit for Mobile, Web, & Desktop Apps from a Single Codebase
Time
11:10
Speakers
Kevin Moore
Flutter is an open source framework for building beautiful, natively compiled, multi-platform applications from a single codebase.
Time
11:10
11:50
Blitz: Blurring the Lines Between Web and Native
Time
11:50
Speakers
Nico Burns
Introducing Blitz, a new cross-platform application runtime which combines web-compatible "write-once, run-everywhere " UI code with a small footprint and powerful system integrations when deploying to native (non-web) platforms.
Time
11:50
12:30
Lunch Break
Time
12:30
Speakers
Time
12:30
14:00
uni-app x: The next generation, pure native cross-platform framework
Time
14:00
Speakers
Hongbao Cui
In recent years, cross-platform development has long been a familiar development paradigm, but cross-platform development is still often criticized by developers? Why? How to improve the solution? In this speech, I will share the thinking and exploration of uni-app team on this problem, and expound their performance bottlenecks and optimization practices for the three platforms of native, H5 and mini program, and introduce uni-app x, a new cross-platform framework. uni-app x uses a TS-like DSL, compiles to swift/Kotlin/arkTS through Rust, and obtains a pure native App. On the premise of completely solving the performance problems of traditional Hybrid apps, Uni-App X uses the web ecology to liberate productivity and deliver quickly.
Time
14:00
14:40
Incremental Linking and Hot Code Reloading for Rust
Time
14:40
Slow edit-build-run dev cycles are a common complaint among Rust developers. The Wild project aims to help by building a linker in Rust with the goal of eventually supporting incremental linking and hot code reloading.
Time
14:40
15:20
Afternoon Break
Time
15:20
Speakers
30 Minutes Break
Time
15:20
15:50
Porting Servo to OpenHarmony
Time
15:50
A progress update on porting Servo, a web rendering engine written in Rust, to OpenHarmony
Time
15:50
16:30
GenUI: Declarative Rust Cross-platform Framework Based on Makepad
Time
16:30
Speakers
Yifei Sheng
GenUI is an innovative SFP front-end framework developed in the Rust language, was originally inspired by Vue3 and Makepad. Intended to help users write front-end projects using Rust more efficiently.
Time
16:30

October 18

App Development
Time
Session
9:30
Makepad: Visual Application Design in Rust
Time
9:30
Speakers
Rik Arends
In this talk I will show the new MIT licensed visual application IDE and design tool Makepad Studio. We will go through local-model AI assisted programming and an early peek at our visual UI designer. Makepad is a UI framework and IDE/designtool combination for Rust. Makepad runs natively on Web, Android, iOS, macOS, Linux and Windows using a GPU accelerated stack. With a focus on staying light-weight a complete application still compiles in 10 seconds with all dependencies on a modern computer.
Time
9:30
10:10
Kotlin Multiplatform: A Flexible Way for Cross-Platform Development
Time
10:10
Speakers
Shengyou Fan
Kotlin Multiplatform has become a robust tool for cross-platform development, enabling developers to use a single programming language to build applications across mobile, desktop, web, and server platforms. In this talk, I will introduce the key concepts of Kotlin Multiplatform, demonstrate the project architecture and development workflow, and showcase the toolchain and ecosystem that support it. Additionally, we will explore learning resources and the Kotlin community. By the end of the session, you'll see why Kotlin Multiplatform is an excellent choice for cross-platform development and understand its potential.
Time
10:10
10:50
Morning Break
Time
10:50
Speakers
20 Minutes Break
Time
10:50
11:10
Conquering Platforms with React Native
React Native has expanded beyond iOS and Android, incorporating more platforms such as Microsoft Windows and Huawei's OpenHarmony. Let's dive into the efforts of React Native's open-source community and see how it's done.
Time
11:10
11:50
Quake: Bridging the Build System Gap
Time
11:50
Software complexity has grown to outpace many of the build systems which support it. Modern applications require build-time features such as asset handling and cross-platform, multi-lingual compilation, but the tools used for these tasks rarely fit such a broad scope. quake addresses this issue head-on by offering an expressive, cross-platform build system compatible with any tech stack. We’ll also explore other popular build systems and techniques, as well as the challenges they face, to better understand our current landscape and what the future might hold.
Time
11:50
12:30
Lunch Break
Time
12:30
Speakers
Time
12:30
14:00
A New Cross-Platform Solution: ArkUI-X
Time
14:00
Speakers
Guoqi Yan
Introducing ArkUI-X, a new generation of cross-platform framework based on the Hongmeng native paradigm. Based on ArkUI-X, Hongmeng native applications can be quickly deployed to Android and iOS, achieving one-time development and multi-platform reuse
Time
14:00
14:40
Native HarmonyOS Application Development Based on Rust + Makepad on OHOS
Time
14:40
Two short presentations in one session: 1. Adaptation of Rust in native HarmonyOS application development and sharing of usage experience. 2. OpenHarmony provides the XCommonent component for graphics drawing and media data writing, and also provides callback functions for interface operation related events. Using these two capabilities, the EGL instructions generated in Makepad can be output to XCommonent to complete the display of graphics, and by registering the XCommonent operation event callback function, the screen operation events and screen coordinates can be converted into Makepad internal events and coordinates.
Time
14:40
15:20
Afternoon Break
Time
15:20
Speakers
30 Minutes Break
Time
15:20
15:50
Structured Text Generation with Incremental Parsers in Rust
Time
15:50
Speakers
Evan Almloff
LLMs are a powerful tool for understanding unstructured data. We will explore how Kalosm uses Rust for high-performance structured generation to extract structured data from model outputs. Along the way, we will dive into LLM internals and high-performance incremental parsers
Time
15:50
16:30
Masonry: An Imperative Widget Toolkit in Rust
Time
16:30
Speakers
Daniel McNab
Masonry enables the development of high-performance GUI frameworks by uncoupling state management from the underlying behaviour. Its open widget set enables greater flexibility than the web DOM. This will be a video recorded talk.
Time
16:30

AI Models & Infra Track: Pioneering the Next Generation of AI

In the AI Models & Infra Track, we bring together experts from the most influential open-source AI projects around the world. This track covers key topics shaping the future of AI, including:

  • AI foundation models and their role in building advanced, intelligent systems
  • On-device models that deliver AI-powered solutions directly on edge devices
  • Fully open-source models, enabling transparency, collaboration, and innovation in AI development
  • Large-scale AI training frameworks that support cutting-edge AI advancements
  • Edge inference frameworks, pushing AI capabilities to the edge for real-time solutions
  • Agentic frameworks, advancing AI's capacity for autonomy and decision-making
  • Fine-tuning techniques that customize models for specific applications
  • AI applications in vertical domains, showcasing how AI is transforming industries across the board

This track provides a deep dive into the innovations driving AI forward, offering valuable insights from the leaders shaping tomorrow's intelligent systems.

Don’t miss this opportunity to learn from the best and explore the future of AI development!

October 17

AI Models & Infra
Time
Session
8:30
Registration
Time
8:30
Speakers
Time
8:30
9:30
GOSIM Keynote: Open Source Ecosystem in the Era of AGI
Time
9:30
Speakers
Tao Jiang
Open Source Ecosystem in the era of AGI... More Details Soon
Time
9:30
9:45
Opening Keynote: Mobile Solution by OpenHarmony
Time
9:45
Speakers
Yan Zhou
Huawei promotes the openness, concurrency, and security of mobile devices operating systems based on OpenHarmony, and combines the lightweight characteristics of devices to provide the industry with more effective parallel and concurrency solutions.
Time
9:45
10:05
GOSIM Hackathon Awards Ceremony
Time
10:05
Speakers
Winner of the GOSIM Hackathon will be announced.
Time
10:05
10:10
Morning Break
Time
10:10
Speakers
20 Minutes Break
Time
10:10
10:30
OPEN-SORA: Democratizing Efficient Video Production for All
Time
10:30
Speakers
Yang You
The Open-Sora project is the world's first open-source reproduction solution similar to Sora, creating an efficient and low-cost video generation model. This presentation will begin by introducing the core concepts of Open-Sora, providing an in-depth analysis of its unique model architecture, and demonstrating how this architecture optimizes video generation efficiency while significantly reducing costs. The presentation will also discuss how Open-Sora adapts to different video resolutions and lengths while maintaining high-quality output, and how it addresses common data storage challenges in large model training. Additionally, it will cover optimizations in inference cost and memory usage. Open-Sora has received enthusiastic responses in the open-source community due to its fun and high-quality results. In this presentation, we will also showcase creative works from the open-source community and explore cutting-edge video generation technologies with you.
Time
10:30
11:10
Construction and Open-Sourcing of the Bilingual Large Models in the Aquila Series within the FlagOpen
Time
11:10
This presentation primarily introduces the construction and open-sourcing of the bilingual large models in the Aquila series, specifically including model pretraining, open-sourcing of the training framework, and open-sourcing of the datasets.
Time
11:10
11:50
InternLM: An Open-Source LLM with Full-Stack Toolchain
Time
11:50
Speakers
Jianfei Gao
InternLM is developed by Shanghai AI Laboratory. We keep open-sourcing high quality LLMs as well as a full-stack toolchain for development and application. In this talk, we will share the latest developments of InternLM and its related toolchain, covering areas such as Agents, Evaluation, and AI-Infra.
Time
11:50
12:30
Lunch Break
Time
12:30
Speakers
Time
12:30
14:00
Demystifying the LLM Training with the Fully Open-Source Moxin 7B Model
Time
14:00
Speakers
Yanzhi Wang
Recently, Large Language Models (LLMs) have undergone a significant transformation, marked by a rapid rise in both their popularity and capabilities. Open-source LLMs, such as LLaMA and Mistral, have made great contributions to the ever-increasing popularity of LLMs due to the ease to customize and deploy the models across various applications. Although LLMs offer unprecedented opportunities for research and innovation, its commercialization has raised concerns about transparency, reproducibility, and safety. Many open LLM models lack the necessary components (such as training code and data) for full understanding and reproducibility, and some use restrictive licenses whilst claiming to be “open-source”, which may hinder further innovations on LLMs. To mitigate this issue, we follow the Model Openness Framework (MOF), a ranked classification system that rates machine learning models based on their completeness and openness, following principles of open science, open source, open data, and open access. We present a truly open source LLM Moxin 7B and release pre-training code and configurations, training and fine-tuning data, and intermediate and final checkpoints, aiming to make continuous commitments to fully open-source LLMs.
Time
14:00
14:40
Towards Practical Multimodal Large Models
Time
14:40
Speakers
Tianyu Yu
As the inevitable path towards general artificial intelligence, multimodal large models have shown great potential for intelligent transition. It is not only at the forefront of academic exploration but also a catalyst for building a community with a shared future for mankind and promoting global cooperation. High deployment and inference costs, frequent hallucination behaviors, and scarce high-quality data all greatly restrict the development of multimodal large models. Starting from these key issues, MiniCPM-V has for the first time achieved comparable end-side multimodal understanding capabilities to closed-source commercial multimodal large models.
Time
14:40
15:20
Afternoon Break
Time
15:20
Speakers
30 Minutes Break
Time
15:20
15:50
CodeGeeX4: A Versatile Code Generation Model for AI Programming
Time
15:50
Speakers
Qinkai Zheng
We introduce CodeGeeX4-ALL-9B, the open-source version of the latest CodeGeeX4 model series. It is a multilingual code generation model continually trained on the GLM-4-9B, significantly enhancing its code generation capabilities. Using a single CodeGeeX4-ALL-9B model, it can support comprehensive functions such as code completion and generation, code interpreter, web search, function call, and repository-level code Q&A, covering various scenarios of software development. CodeGeeX4-ALL-9B has achieved highly competitive performance on public benchmarks, such as BigCodeBench and NaturalCodeBench. It is currently the most powerful code generation model with less than 10B parameters, even surpassing much larger general-purpose models, achieving the best balance in terms of inference speed and model performance.
Time
15:50
16:30
Functionary: Using tools with LLMs
Time
16:30
Functionary is an open-source LLM that is capable of using tools while maintaining conversational abilities. We will discuss how complex agentic tasks can be achieved with function calls and how we addressed challenges of function calling.
Time
16:30
17:10
Practical Exploration of XVERSE Large Language Model Series
Time
17:10
Speakers
John Xuan
This presentation primarily showcases the hands-on exploration of developing the XVERSE large language model series, covering various stages from ideation to realization and evolving from dense models to Mixture-of-Experts (MoE) models. In terms of data, we will outline the necessary datasets for effective training, describe preprocessing protocols, and discuss approaches for continual model improvement. Regarding the model aspect, there will be a focus on adapting from standard dense layers to implementing MoE structures, highlighting decision criteria for specifying expert capacities and allocating importance among them. Furthermore, addressing architectural facets, we will articulate methodologies designed to maximize training productivity and secure consistent learning performance.
Time
17:10

October 18

AI Models & Infra
Time
Session
9:30
vLLM: Easy, Fast, and Cheap LLM Serving for Everyone
Time
9:30
Speakers
Kaichao You
vLLM is a fast and easy-to-use library for LLM inference and serving. In this talk, I will briefly introduce the evolution of the vLLM project, the open-source community behind it, and highlight some features that are interesting to many users.
Time
9:30
10:10
The openMind application enablement kit brings a new training and inference experience for Ascend development
Time
10:10
Speakers
Yongqiang Ma
The continuous development of Ascend cannot be separated from the cooperation of the open source community. As the ecological window of Ascend, the openMind tool suite focuses on providing developers with lightweight capabilities from model distribution to integrated model training and inference, reducing the development threshold of NPUs. This sharing will provide the audience with exclusive features related to openMind including the improvement of the efficiency of fusion operators and the adaptive version management of containers, the adaptation capabilities of third-party libraries in the open source community including LLaMA-Factory, and the progress of cooperation with the Magic Leap community.
Time
10:10
10:50
Morning Break
Time
10:50
Speakers
20 Minutes Break
Time
10:50
11:10
LLaMA Factory: a Versatile Efficient Training Framework for LLMs
Time
11:10
Speakers
Yaowei Zheng
LLaMA Factory is an efficient, easy-to-use, and scalable open source framework for efficient training of large models. It has received more than 30,000 Stars in the GitHub open source community and has attracted the attention or application of many domestic and foreign companies. This sharing will analyze the construction motivation and component modules of LLaMA Factory in detail from the perspective of efficient training of large models, including the full-stack fine-tuning adaptation principles of hundreds of large models, LoRA operator optimization acceleration methods, and various fine-tuning trick integration ideas.
Time
11:10
11:50
Weaving Your AI App with MoFA
Time
11:50
Speakers
Zonghuan Wu
The AI era has risen with the emergence of large language models and will undoubtedly flourish with the maturation of AIOS. All of our efforts are aimed at this goal, focusing on advancing an open, local, and composable AI ecosystem. This talk will highlight the MoFA open-source project, a modular, composable, and programmable AI agent framework. Through an overview of our project suite, we will outline the blueprint for realizing a composable AIOS.
Time
11:50
12:30
AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation
Time
12:30
Speakers
Qingyun Wu
AutoGen is an open-source programming framework for agentic AI. It enables development of LLM applications using multiple agents that can converse with each other to solve tasks. In this talk, the speaker will provide an introduction of AutoGen, sharing the latest updates and ongoing research efforts spanning across key directions such as evaluation, interfaces, learning/optimization/teaching, and seamless integration with existing AI technologies. The speaker will also delve into open questions and future plans surrounding AutoGen. (This will be a pre-recorded talk)
Time
12:30
12:30
Lunch Break
Time
12:30
Speakers
Time
12:30
14:00
Navigating Open Source AI Governance: The Model Openness Framework and the EU AI Act
Time
14:00
Explore the Model Openness Framework (MOF) and its role in Open Source AI governance. Learn how MOF can guide compliance with new regulations like the AI Act, addressing key questions about component release and appropriate licensing in open source AI systems.
Time
14:00
14:40
AI Hub, the Next Generation Open Source Model Community for Chinese Developers
Time
14:40
Speakers
Zhikang Cui
How AI Hub builds a one-stop AI ecosystem for Chinese developers to help them innovate quickly
Time
14:40
15:20
Afternoon Break
Time
15:20
Speakers
30 Minutes Break
Time
15:20
15:50
A New Start with XLang™
Time
15:50
Speakers
Longjiang Li
XLang™ is a next-generation programming language crafted for AI and IoT applications, designed to deliver dynamic, high-performance computing. It excels in distributed computing and offers seamless integration with popular languages like C++, Python, and JavaScript, making it a versatile choice across diverse operating systems.
Time
15:50
16:30
AI Application Workflow Orchestration System Based on LLM
Time
16:30
The rapid development of large language models (LLMs) has led to the emergence of diverse architectures and frameworks, each introducing unique AI workloads and workflows. However, challenges such as efficient orchestration, workload reuse, and reducing the deployment barriers for LLM-based applications remain critical obstacles for enterprises. In this session, we will explore how Intel’s open-source initiative, OPEA (Open Platform for Enterprise AI), addresses these challenges. A key focus will be on the GMC (GenAI Microservices Connector), a dynamic orchestration tool designed to optimize AI workloads. Join us to learn how GMC simplifies LLM application deployment and enhances workflow efficiency.
Time
16:30

Embodied AI Track: Pioneering the Future of Intelligent Robotics

The Embodied AI track brings together the latest breakthroughs in robotics and artificial intelligence, offering a deep dive into cutting-edge technologies shaping the future of intelligent machines. Key topics include:

  • Progress on the open-source robotic operating system, advanced data collection mechanisms, and innovative model training frameworks crucial to the development of smarter, more capable robots.
  • The next-generation robotic dataset, blending physical and synthetic data to build robotic foundation models that enable universal robots to perform both industrial tasks and household chores.
  • Exploration of open-source computer vision projects, such as OpenCV, and the latest open-source visual language models, opening new doors for spatial intelligence in robotics.
  • The evolution of the open-source humanoid robot and advancements in the robot arm ecosystem, showcasing the future of versatile, collaborative robotic systems.

Join this track to explore how open-source innovation is driving the next wave of robotic capabilities, bringing us closer to a future where intelligent robots seamlessly integrate into industry and everyday life.

October 17

Embodied AI
Time
Session
8:30
Registration
Time
8:30
Speakers
Time
8:30
9:30
GOSIM Keynote: Open Source Ecosystem in the Era of AGI
Time
9:30
Speakers
Tao Jiang
Open Source Ecosystem in the era of AGI... More Details Soon
Time
9:30
9:45
Opening Keynote: Mobile Solution by OpenHarmony
Time
9:45
Speakers
Yan Zhou
Huawei promotes the openness, concurrency, and security of mobile devices operating systems based on OpenHarmony, and combines the lightweight characteristics of devices to provide the industry with more effective parallel and concurrency solutions.
Time
9:45
10:05
GOSIM Hackathon Awards Ceremony
Time
10:05
Speakers
Winner of the GOSIM Hackathon will be announced.
Time
10:05
10:10
Morning Break
Time
10:10
Speakers
20 Minutes Break
Time
10:10
10:30
OpenCV and Spatial Intelligence
Time
10:30
Speakers
Gary Bradski
I will describe the upcoming OpenCV release and current initiatives into "spatial intelligence" (AI + 3D pose in a space)
Time
10:30
11:10
Eclipse Zenoh: the Communication Backbone for Ubiquitous and Heterogeneous AI Use-Cases
Time
11:10
AI has become ubiquitous, enhancing many of our interactions with new technology. This omnipresence requires a powerful communication backbone allowing data to flow seamlessly. This presentation details Eclipse Zenoh, a next-gen communication middleware.
Time
11:10
11:50
On the Role of Open-Source Simulation in the Era of Embodied AI
Time
11:50
In the era of Embodied AI, open-source simulation platforms play a pivotal role in driving innovation and collaboration. This talk explores the balance between openness and closed software/APIs, emphasizing the need for flexibility and extensibility in developing robust simulation tools for Embodied AI. By examining CARLA as a case study, I highlight how open-source initiatives can foster community engagement, accelerate research, and provide adaptable tools that meet diverse research needs.
Time
11:50
12:30
Lunch Break
Time
12:30
Speakers
Time
12:30
14:00
Build an Open Source Operating System for Robots Based on OpenHarmony.
Time
14:00
Speakers
Yanxing Ba
Technological progress, an aging population and many other factors are driving the booming development of the robotics industry, but there is currently no official operating system that is fully oriented to robots on the market. OpenHarmony is an operating system for smart terminal devices that is oriented to the full-scenario, fully connected and fully intelligent era. Its technical features such as the distributed capabilities of the soft bus are highly consistent with the robot's demands for an OS. Shenzhen Kaihong, in conjunction with a number of leading companies and universities, launched an open source version of the robot operating system based on OpenHarmony in the open source community under the guidance of the OpenHarmony Technical Steering Committee. This topic will focus on the system architecture, technical features, industry applications and other aspects of this OS.
Time
14:00
14:00
The Dexterous Hand of Self-Evolution,The Key to Opening the Door to Embodied Intelligence
Time
14:00
Speakers
David Cui
The dexterous hand of self-evolution,The key to opening the door to embodied intelligence
Time
14:00
14:40
Task Modeling Method for Functional Unmanned Vehicle
Time
14:40
Speakers
Ming Cen
As a vehicle for smart missions such as delivery, selling, shuttling and security patrol, functional unmanned vehicles can be regarded as typical representatives of embodied intelligence. Considering of variety of smart tasks, designing a set of general method for task modeling and performing is of great significance to achieve the rapid development and deployment of functional unmanned vehicles.
Time
14:40
15:20
Afternoon Break
Time
15:20
Speakers
30 Minutes Break
Time
15:20
15:50
Robot technology driven by large models in the industrial field
Time
15:50
Speakers
Lei Yang
In the industrial field, there is a growing demand for data-driven robotics. The technical paths include advanced sensor tech, machine learning algorithms, and real-time data processing. However, we also face challenges like data security and interoperability.
Time
15:50
16:30
What Data Imperative for Action Learning of Embodied AI?
Time
16:30
Speakers
Yu Huang
After occurring of large scale language models (LLMs), the development of Embodied AI is seen as a fundamental path to achieving artificial general intelligence (AGI). However, data is currently a key bottleneck for advancement in Embodied AI, compared to that of LLMs. In this talk, the learning strategies are first analyzed, especially for embodied manipulation. Meanwhile, various existing public datasets in embodied AI are investigated. Eventually, requirements for an imperative embodied AI dataset are concluded and an upcoming synthesized video dataset (rendered by Unreal 5) called MVGameIR is introduced.
Time
16:30
17:10
Open World Embodied Large Models
Time
17:10
Speakers
Jiaming Liu
Multimodal Large Language Models (MLLMs) have demonstrated potential in visual instruction following across various tasks. Recently, some studies have integrated MLLMs into robotic manipulation, allowing robots to interpret multimodal information and predict low-level actions. While MLLM-based policies have shown promising progress, they may predict failure execution poses when faced with novel tasks or categories. Given these challenges, we raise a question: “Can we develop an end-to-end robotic agent that not only possesses manipulation skills but also effectively corrects low-level failure actions?” Drawing inspiration from Daniel Kahneman's assertion that "human thinking is divided into a fast system and a slow system, which separately represent intuitive processes and more logical reasoning," we introduce a series of research works that mimic a human-like thinking paradigm to address the above question.
Time
17:10

October 18

Embodied AI
Time
Session
9:30
AIRSHIP: Empowering Intelligent Robots through Embodied AI, Stronger United, Yet Distinct.
Time
9:30
Speakers
Shaoshan Liu
AIRSHIP has been developed to tackle the problem of software complexity in embodied AI. Its mission is to provide an easy-to-deploy software stack that empowers a wide variety of intelligent robots, thereby facilitating scalability and accelerating the commercialization of the embodied AI sector. AIRSHIP takes inspiration from Android, which played a crucial role in the mobile computing revolution by offering an open-source, flexible platform. Android enabled a wide range of device manufacturers to create smartphones and tablets at different price points, sparking rapid innovation and competition. This led to the widespread availability of affordable and powerful mobile devices. Android's robust ecosystem, supported by a vast library of apps through the Google Play Store, allowed developers to reach a global audience, significantly advancing mobile technology adoption. Similar to Android, AIRSHIP offers both software and hardware specifications, enabling robotic builders to develop complete embodied AI systems for a range of scenarios, including home, retail, and warehouse environments.
Time
9:30
10:10
A Mocap-based Data Factory for Embodied AI
Time
10:10
The presentation will begin by discussing the relationship between motion capture (human motion) and the field of embodied AI, introducing the workflow of Noitom’s motion capture solutions applied in the embodied AI robotics sector. It will emphasize the importance and challenges of mapping of human body to the robot body within the overall workflow. Additionally, it will explain the differences between four main approaches to generating robot datasets, and finally, by showcasing success story of the industry partners of Noitom, it will once again highlight the importance and feasibility of establishing large-scale embodied AI data factories (DataFactory).
Time
10:10
10:50
Morning Break
Time
10:50
Speakers
20 Minutes Break
Time
10:50
11:10
Super Easy Fine-Tuning of Robotic Arm AI Model with dora-rs
Time
11:10
Speakers
Xavier Tao
Making robot autonomous is getting easier and easier. In today's talk, we're going to see how in couple of steps we can get started with training a model that automate our task. This is thanks to dora-rs which makes finetuning AI model super easy.
Time
11:10
11:50
Rapier: One Step Further Toward Distributed Physics Simulation
Time
11:50
Rapier is a powerful open-source physics engine designed for Rust. In this talk we will present our latest findings on distributed physics simulation and how to leverage physics-engines-specific features to improve its performance.
Time
11:50
12:30
Lunch Break
Time
12:30
Speakers
Time
12:30
14:00
Embodied Collaborative Navigation and Interactive Learning
Time
14:00
Speakers
Huaping Liu
At present, complex and highly dynamic environments have put forward more urgent demands on robot perception and learning. Cluster systems have brought efficiency multiplication and application breakthroughs to multi-robot collaborative applications, but also brought huge challenges to the theoretical research and engineering application of group intelligence perception and learning. This report focuses on the two major multi-robot collaborative task requirements of situation understanding in adapting to the differences in the perception and action capabilities of heterogeneous robot platforms, and efficient and robust perception in wide-area dynamic scenes. It introduces the relevant research progress on how to use the perception and learning capabilities of heterogeneous multi-robots to achieve cluster efficiency and behavior emergence.
Time
14:00
14:40
OpenCSG Embodied Intelligent Data Platform
Time
14:40
Speakers
Dong Cao
OpenCSG has launched a powerful embodied intelligence data platform product: CSG DataFlow, which aims to provide data scientists and developers with efficient and comprehensive embodied intelligence application scenario dataset processing solutions.
Time
14:40
15:20
Afternoon Break
Time
15:20
Speakers
30 Minutes Break
Time
15:20
15:50
OpenLoong Community: Building an Open and Collaborative Humanoid Robot Open Source Ecosystem
Time
15:50
Speakers
Chong Tian
This presentation will explore the founding, development, achievements and future prospects of the OpenLoong community. As the leading humanoid robot open source community in China, OpenLoong is committed to promoting the development of humanoid robot embodied intelligence technology and promoting communication and cooperation among developers. The presentation will share the community's success stories, challenges, and how to build a more prosperous humanoid robot open source ecosystem through open collaboration.
Time
15:50
16:30
Breaking the Boundary - Exploring the Robot Endgame
Time
16:30
Speakers
Allen Zhang
Breaking the Boundary - Exploring the Robot Endgame
Time
16:30

Next Generation Internet: Building the Web of Tomorrow

The Next Generation Internet is reshaping how we think about web applications, identity, and global collaboration. This track will explore:

  • Decentralized applications like messaging and social networking that enhance individual control and privacy, empowering users through decentralized AI communities with improved data protection.
  • Open-source browsers designed for optimal performance, security, and embeddability, leading the way for the future of browser-based applications, including mini apps.
  • Trust-spanning protocols that unify centralized and decentralized identity systems, forming the foundation of the next generation of trust design at an internet-wide scale.
  • The critical role of open-source implementations of decentralized protocols and applications, which are essential for turning new internet standards into reality.

This track highlights the key innovations driving the future of the internet and how these technologies are fostering a more secure, decentralized, and collaborative web for all.

October 17

Next Generation Internet
Time
Session
8:30
Registration
Time
8:30
Speakers
Time
8:30
9:30
GOSIM Keynote: Open Source Ecosystem in the Era of AGI
Time
9:30
Speakers
Tao Jiang
Open Source Ecosystem in the era of AGI... More Details Soon
Time
9:30
9:45
Opening Keynote: Mobile Solution by OpenHarmony
Time
9:45
Speakers
Yan Zhou
Huawei promotes the openness, concurrency, and security of mobile devices operating systems based on OpenHarmony, and combines the lightweight characteristics of devices to provide the industry with more effective parallel and concurrency solutions.
Time
9:45
10:05
GOSIM Hackathon Awards Ceremony
Time
10:05
Speakers
Winner of the GOSIM Hackathon will be announced.
Time
10:05
10:10
Morning Break
Time
10:10
Speakers
20 Minutes Break
Time
10:10
10:30
Decentralized Technologies and Next-Generation Platforms
Time
10:30
Speakers
Alan Majer
Digital experiences are heavily mediated by centralized platforms. These dominant platforms face disruption by decentralized building blocks, AI, and algorithms - affording new opportunities to rebuild our platforms on open foundation, and reclaim control of our data and experiences. This presentation shares insights gleaned from research interviews with 23 people working toward that open future.
Time
10:30
11:10
Building for a User-Centered Internet
Time
11:10
Speakers
Rui Zhao
The Internet is increasingly centralised around a few big platforms, leading to numerous problems ignoring or harming users' benefits. In recent years, more attention is paid on decentralised architectures with more user focus. In this talk, I will briefly revisit the different lines of thinking, and focus on one promising approach, Solid (Social Linked Data), proposed by Web's inventor, Tim Berners-Lee, to re-decentralised the Web. I will cover its key principles, what it offers right now, and (our team's) explorations to expand its boarder. The talk will close with visions upon Solid and corresponding open questions to incentivise future evolution.
Time
11:10
11:50
Palpo Social Server: A self-hosted social server supporting Matrix protocol
Time
11:50
A Rust based self-hosted, open-source Matrix server for personal cloud deployment.
Time
11:50
12:30
Lunch Break
Time
12:30
Speakers
Time
12:30
14:00
From Blockchain to Web3: Building the Decentralized Network
Time
14:00
Speakers
Huawei Huang
Scalability has become a key research direction at the bottom of blockchain. The academic community has proposed several representative solutions to improve the scalability of blockchain, such as directed acyclic graph technology, sharding technology, new consensus mechanism, state channel, side chain, cross-chain and other solutions. Among them, sharding technology is considered to be a Layer 1 expansion solution with great potential to greatly improve the scalability of blockchain. This report first sorted out the research status of blockchain sharding technology, and then introduced the new DeFi protocol built by Professor Huang's research team based on sharded blockchain - BrokerFi.
Time
14:00
14:40
Robrix: a Pure Rust Multi-Platform Matrix Client and More
Time
14:40
Speakers
Kevin Boos
Robrix is a new Matrix chat client that aims to be a hub for federated services. By building atop the Robius app dev framework and the Makepad UI toolkit, Robrix works seamlessly across multiple platforms (including mobile) with no platform-specific code.
Time
14:40
15:20
Afternoon Break
Time
15:20
Speakers
30 Minutes Break
Time
15:20
15:50
Moxin: Enhancing the Decentralized Distribution Capabilities of Large Models
Time
15:50
Speakers
Shu Cai
With the development of artificial intelligence, large language models (LLMs) are increasingly used in various industries, but the centralized distribution of model files still faces problems such as high infrastructure costs and limited access. Moxin aims to solve these challenges by distributing model files in a decentralized way and focusing on efficient and secure transmission of model files.
Time
15:50
16:30
Compression Carcinized: Drop-in Compatible Compression in Rust
Time
16:30
Data compression is used when loading almost anything on the web. This talk is about how we implement zlib-rs and other compression algorithms in rust so that they are drop-in compatible with their C equivalent.
Time
16:30

October 18

Next Generation Internet
Time
Session
9:30
Wasm as the Universal Runtime for the API-First and Agentic Internet
Time
9:30
Speakers
Michael Yuan
Wasm as emerged as a universal runtime environment that is cross-platform, hardware portable and language agnostic. It can be embedded into many different types of applications or SaaS as widgets or extensions. In this talk, we will discuss popular use cases of Wasm in programming the next-gen Internet from serverless functions, workflow automation, web3 smart contracts to AI agents.
Time
9:30
10:10
The Trust Spanning Protocol and Its Role for Authentic AI
Time
10:10
Speakers
Wenjing Chu
The lack of authenticity plagues today's Internet. Imagine a world overrun with AGI agents that we can not trust, let alone control! Time is now when we must lay a solid foundation of trust for the future of AI and Internet. The success of AI adoption is not possible without the framework with which people, organizations and society can establish technical and human level trust and control. In this talk, Wenjing will introduce the cutting edge information authenticity technology around the Trust Spanning Protocol (TSP) and how it plays a pivotal role in enabling authentic AI in content generation, reasoning, and autonomous agents.
Time
10:10
10:50
Morning Break
Time
10:50
Speakers
20 Minutes Break
Time
10:50
11:10
The Global Acceptance Network: Implementing Decentralized Digital Trust Infrastructure at Scale
Time
11:10
The Global Acceptance Network (GAN) is a new layer of digital public infrastructure for the Internet — comparable in some ways to the Internet’s current Domain Name System (DNS). The GAN will enable people, organizations, and “things” (e.g., devices, software, AI bots) to form and maintain sustainable digital trust relationships using open standard digital wallets, digital credentials, and trust registries. This new utility is designed to address many of the current ills of the Internet, including spam, phishing, fraud, malware, ransomware, and other persistent security, privacy, and trust issues that have plagued the Internet because its original design did not include this trust layer.
Time
11:10
11:50
What is a WebAssembly Component (and why)
Time
11:50
WebAssembly Components, part of the component-model proposal to WebAssembly, promise a framework of developing portable and composable applications both inside and outside the browser. Components preserve the best parts of WebAssembly such as its portability, security, and performance, while also bringing the benefits of higher-level API design and composability to applications. Come learn about the component model and what it might be able to do for you!
Time
11:50
12:30
Lunch Break
Time
12:30
Speakers
Time
12:30
14:00
The Role of Open Source Education in the Next Generation Internet (NGI)
Time
14:00
Open source talent training in colleges of economics, management and law
Time
14:00
14:40
Browser Tailoring and Thinking
Time
14:40
Speakers
Bin Wang
Explore and build a compact browser kernel
Time
14:40
15:20
Afternoon Break
Time
15:20
Speakers
30 Minutes Break
Time
15:20
15:50
Solving Concurrency with (In-)formal Methods in Servo
Time
15:50
Servo is a Web engine, and one of its goal is parallelism, which creates challenges related to concurrency which aren’t covered by the Rust compiler. TLA plus is a mathematical language for modelling concurrent algorithms, making it a useful tool to improve the correctness of Servo. Recent and practical use cases will be discussed in this talk.
Time
15:50
16:30
Bye WebSockets and WebRTC! Boost Web Comms with Rust, QUIC, WebTransport and HTTP3
Time
16:30
In this talk, we'll explore the future of real-time web communication by moving beyond WebSockets and WebRTC. Learn how Rust, QUIC, WebTransport, and HTTP/3 offer enhanced security, scalability, and performance. Using a videoconferencing system as a case study, we’ll demonstrate how these technologies can revolutionize modern web communication.
Time
16:30

Next Generation Media: Revolutionizing Visual and Interactive Experiences

The Next Generation Media track will dive into the cutting-edge technologies transforming how we create and interact with digital content. Key topics include:

  • The WebGPU standard and its open-source implementation, a groundbreaking unified computing kernel for graphics, physics simulations, and generative AI applications.
  • The latest open-source generative AI models and tools for generating high-quality image, video, and 3D content, unlocking new possibilities for creative expression.
  • Emerging generative media applications, showcasing how AI-driven content creation is redefining industries like entertainment, design, and marketing.
  • The next wave of AI-powered XR and wearable devices, set to revolutionize the consumer device landscape by merging immersive experiences with smart technology.

Join us to explore the future of media creation, interactive experiences, and the technologies that are shaping the next generation of digital content.

October 17

Next Generation Media & Device
Time
Session
8:30
Registration
Time
8:30
Speakers
Time
8:30
9:30
GOSIM Keynote: Open Source Ecosystem in the Era of AGI
Time
9:30
Speakers
Tao Jiang
Open Source Ecosystem in the era of AGI... More Details Soon
Time
9:30
9:45
Opening Keynote: Mobile Solution by OpenHarmony
Time
9:45
Speakers
Yan Zhou
Huawei promotes the openness, concurrency, and security of mobile devices operating systems based on OpenHarmony, and combines the lightweight characteristics of devices to provide the industry with more effective parallel and concurrency solutions.
Time
9:45
10:05
GOSIM Hackathon Awards Ceremony
Time
10:05
Speakers
Winner of the GOSIM Hackathon will be announced.
Time
10:05
10:10
Morning Break
Time
10:10
Speakers
20 Minutes Break
Time
10:10
10:30
WGPU: Modern GPU Programming In Rust
Time
10:30
Speakers
Jim Blandy
GPUs have great performance—if you can make your problem fit their computational model. Hyperthreading, local memory spaces, mandatory vectorization, and other quirks all affect how GPU code should be written. WGPU is a cross-platform API for GPU coding.
Time
10:30
11:10
Introduction to the Latest Progress of the Khronos Vulkan Standard
Time
11:10
Speakers
Kangying Cai
Introduces the Khronos Vulkan standard, the latest progress of the standard, and some new Vulkan features.
Time
11:10
11:50
WGMath: Reusable WebGPU Shaders for Cross-Platform GPU Scientific Computing
Time
11:50
Using WebGpu, WGMath aims to bring GPU scientific computing to all platforms by empowering all scientific communities with reusable WGSL shaders. We discuss the fundamentals of this ecosystem, current and future features for LLM, geometry, and physics.
Time
11:50
12:30
Lunch Break
Time
12:30
Speakers
Time
12:30
14:00
Open-Sourced Text-to-Video Model: CogVideoX
Time
14:00
Speakers
Yuxuan Zhang
We introduce CogVideoX, a large-scale diffusion transformer model designed for generating videos based on text prompts. Results show that CogVideoX demonstrates state-of-the-art performance across both multiple machine metrics and human evaluations. The model weight of CogVideoX is publicly available at https://github.com/THUDM/CogVideo.
Time
14:00
14:40
Open-Sora Plan: An open source plan for large-scale video generation models based on domestic computing power
Time
14:40
Speakers
Xinhua Cheng
Open-Sora Plan is an open source plan for large domestic video generation models that supports the entire calculation on Ascend computing power. This speech introduces the progress, details and planning of Open-Sora Plan.
Time
14:40
15:20
Afternoon Break
Time
15:20
Speakers
30 Minutes Break
Time
15:20
15:50
The Development and Evolution of FaceChain in the Digital Human Domain
Time
15:50
Speakers
Baigui Sun
FaceChain is an innovative open-source project in the field of digital humans, leading the new wave with advancements in facial perception and understanding, digital portrait generation, and talking video creation. With just a single photo, users can create personalized digital avatars and generate unique portrait works using a diverse range of styles and template libraries. FaceChain offers a full-stack digital portrait generation capability, including a 10-second training-free digital portrait generation feature and a DIY style model quick training function, and is currently building an ecosystem for digital portrait content. In addition, FaceChain has also laid out plans and produced outputs for generating talking videos and virtual try-on experiences. With its open-source spirit and vibrant community, FaceChain has driven rapid technological development and widespread application. This topic will delve into the technical features of FaceChain, community building, commercial application prospects, and its evolutionary path in the digital human domain.
Time
15:50
16:30
Pointrix: a differentiable point-based rendering libraries
Time
16:30
Over the past year, significant progress has been made in differentiable point cloud rendering techniques, represented by 3D Gaussian splatting. This report will introduce Pointrix, a differentiable point cloud rendering optimization framework developed by our research group. Pointrix consists of a modular Python API, an efficient CUDA backend, and a user-friendly real-time rendering GUI interface, making it an open-source framework for point cloud rendering that is easy to extend. Pointrix offers detailed quick-start guides and rich example configurations, allowing users to implement specific algorithms by making changes at the Python level. It provides researchers with a portable environment for secondary development. Additionally, Pointrix supports a range of mainstream 3DGS-related work, offering strong support for research and applications in the field of differentiable point cloud rendering.
Time
16:30

October 18

Next Generation Media & Device
Time
Session
9:30
Metaverse Standards Forum Overview
Time
9:30
Speakers
Dennis Fu
Metaverse Standards Forum is one of the most important and also largest global industry consortium focusing on Metaverse standards development. The organization is dedicated to an open metaverse development. This overview will give audiences information regarding its history and vision, as well as a snapshot of its current activities, and progress.
Time
9:30
10:10
OminiX: Towards Unified Library and Acceleration Framework for Generative AI Models on Different Hardware Platforms
Time
10:10
Speakers
Yanzhi Wang
In the generative AI era, general users need to apply different base models, finetuned checkpoints, and LoRAs. Also the data privacy and real-time requirement will favor on-device, local deployment of large-scale generative AI models. It is desirable to develop a "plug-and-play" framework such that users can download any generative AI model, click and run on their own device. This poses significant challenge to the current AI deployment frameworks, which are typically time-consuming and requires human expertise of hardware and code generation. We present our effort of OminiX, which is a first step towards unified library and acceleartion of generative AI models across various hardware platforms. Integrating our unique front-end library and back-end instantaneous acceleration techniques, which will be open-source soon, we show capability of plug-and-play deployment and state-of-the-art acceleration of various generative AI models, starting from image generation, large language models, multi-model language models, speech generation and voice cloning, real-time chatting engine, real-time translation, video generation, real-time avatar, to name a few. This can be achieved without server, just on everyone's own platform.
Time
10:10
10:50
Morning Break
Time
10:50
Speakers
20 Minutes Break
Time
10:50
11:10
Neural Wrist: Wearable Intelligence
Time
11:10
Speakers
Mona Chen
Neural Wrist technology represents an innovative approach to wearable devices, specializing in the collection and analysis of physiological electrical signal data. This advanced technology typically employs sensors that are worn around the wrist to detect and record various physiological parameters such as heart rate, blood pressure, and electrodermal activity. The data collected provides valuable insights into the user’s health status and can be used for a myriad of applications, including fitness tracking, medical diagnosis, and health monitoring.
Time
11:10
11:50
Wearable AI: The Future of Seamless Interaction
Time
11:50
Speakers
Siyuan Qi
The forthcoming generation of wearable AI is poised to redefine daily interactions with technology through all-day comfort and intuitive interfaces. These devices will feature advanced displays that blend seamlessly into the user’s field of view, providing information in a minimally invasive manner. Interactions with these wearables will be as natural as human conversation, with AI-driven personalization that adapts to individual preferences and behaviors. They will serve as central nodes, linking users to a network of smart devices, ensuring a smooth and integrated experience across various aspects of life. In essence, the next wave of wearable AI will offer a personalized, connected experience that is both unobtrusive and central to daily life.
Time
11:50
12:30
Lunch Break
Time
12:30
Speakers
Time
12:30
14:00
SPAD Image Sensor: The Foundation for 2D & 3D Fusion
Time
14:00
Speakers
Changtao Hou
Using a detector based on the SPAD single-photon detector form, the sensor can simultaneously output point cloud (distance) and image (RGB color), achieving a perfect fusion of 2D and 3D.
Time
14:00
14:40
Empowering Immersive Mixed Reality Experiences with OpenXR
Time
14:40
Speakers
Zhipeng Liu
First, introduce the basics of OpenXR; Then, introduce the basics of MR (Mixed Reality); Finally, introduce how to use OpenXR to develop immersive MR experiences.
Time
14:40
15:20
Afternoon Break
Time
15:20
Speakers
30 Minutes Break
Time
15:20
15:50
3D True Video: Challenges and Practices
Time
15:50
Speakers
Shan Gao
Share the development practices, technical applications and challenges of 3D true-to-life video
Time
15:50