正在加载视频...

视频加载失败

返回首页

Working on multimodal instruction tuning and finding it hard to scale? Building Web/GUI agents but data is too narrow? Introducing 🚀MultiUI: 7.3M multimodal instructions from 1M webpage UIs, offering diverse data to boost text-rich visual understanding. Key takeaways: 🌟WebUI-trained models show major gains in visual web understanding and agent...

xiangyue96's profile picture

Xiang Yue

6,579 subscribers

57,736 次观看 • 1 年前 •via X (Twitter)

科学技术新闻政治教育
LIVE

Anya Rossi• Live Now

Private livecam show

10 条评论

Xiang Yue 的头像
Xiang Yue1 年前

Overview of MultiUI, a 7M multimodal instruction-tuning dataset built from a diverse collection of Webpage UIs. The model UIX, trained on MultiUI, generalizes effectively to a broad range of unseen scenarios, including GUI understanding (web and mobile interfaces) and, surprisingly, non-GUI tasks such as document and chart understanding.

Xiang Yue 的头像
Xiang Yue1 年前

Methods: We generated examples by defining representative tasks and prompting text LLMs to create general instructions for various webpages. To enhance diversity, we used strategies like pairing with existing multimodal instructions as in-context examples.

Xiang Yue 的头像
Xiang Yue1 年前

Here is an example of different training tasks in our dataset.

Xiang Yue 的头像
Xiang Yue1 年前

We adopted a two-stage training strategy where we first train the LLaVA models on our 95% MultiUI samples and then we further train the model on a mix of general instructions (e.g., LLaVA) and a small portion (5%) of MultiUI samples. The stage training is mostly useful compared with simply merging the two training datasets. Scaling up different task samples generally leads to better performance.

Xiang Yue 的头像
Xiang Yue1 年前

Further ablation studies show that different tasks have mutual benefits to increase the different abilities of models.

Xiang Yue 的头像
Xiang Yue1 年前

Generating multimodal data with the help of text-based LLMs is an interesting research direction. Web and other potential sources with rendered text would be ideal platforms for exploring such methods. This project took a huge effort to finish in academia. We hope that open-sourcing our work could accelerate the advancement of open science in the field of multimodal LLMs and important downstream applications like multimodal GUI agents!

Zhe Gan 的头像
Zhe Gan1 年前

Nice work! We will also have one paper to be released in the coming week on UI understanding. 😄

randy 的头像
randy1 年前

Exceptionally well done!

Lyman 的头像
Lyman1 年前

nice work

Lyman 的头像
Lyman1 年前

nice work

相关视频

🚀Introducing VisualWebBench: A Comprehensive Benchmark for Multimodal Web Page Understanding and Grounding. 🤔What's this all about? Why this benchmark? > Back in Nov 2023, when we released MMMU ( a comprehensive multimodal understanding benchmark, we received feedback that it included very few UI screenshots. Considering the growing importance of UI understanding, especially with the rise of powerful agents like Devin ( which is built on the strong vision capability of #GPT4, we recognized the need for a benchmark focused on UI screenshot understanding.📸👀 > Multimodal #LLMs have significantly boosted web agents' performance on benchmarks like Mind2Web and WebArena. For instance, the SeeAct agent ( showcases the power of integrating vision into web agents. However, these benchmarks primarily evaluate the end-to-end task execution ability of web agents rather than their understanding of web pages. 🌉 Bridging the Gap with VisualWebBench > To provide a comprehensive evaluation of multimodal LLMs' web page understanding capabilities, we introduce VisualWebBench. Our benchmark spans 139 websites 🌐 across 12 domains 🏷️ and 87 sub-domains 🔍, ensuring a diverse and representative dataset. It assesses MLLMs at three levels: website-level, element-level, and action-level 📊, and encompasses seven tasks designed to evaluate understanding, OCR, grounding, and reasoning abilities 🧠💡. 😮 Surprising Findings > 🎉 Open-source models are catching up: Even though closed-source MLLMs are still leading the leaderboard, we are happy to see open-source models like LLaVA 1.6 34B achieve comparable performance to Gemini Pro. > 🧠 Grounding ability, crucial for developing MLLM-based web applications, is a weakness for most MLLMs. > 🖼️ Importance of Image Resolution: The limited image resolution handling capabilities of most open-source MLLMs restrict their utility in web scenarios, where rich text and elements are prevalent. > 🧱 Relatively strong correlation with general understanding benchmarks like MMMU but weak correlation with web agent benchmarks like Mind2Web. Web agent benchmarks primarily evaluate the end-to-end task execution ability of web agents, which involves a series of actions to accomplish a goal. In contrast, VisualWebBench emphasizes evaluating the foundational skills of MLLMs such as understanding and grounding web page elements. 💡Fun Fact > Claude Sonnet is better than Opus on our benchmark :) 🎓 Conclusion > VisualWebBench serves as a valuable resource for the community, driving research and development in the field of multimodal web page understanding and grounding. As MLLMs continue to evolve and improve, we look forward to seeing new applications and breakthroughs. We believe that our benchmark will contribute to the development of more powerful MLLMs in the web domain, ultimately leading to a more intuitive and efficient user experience on the web. Kudos to the student leads Junpeng Liu Yifan Song and the team Bill Yuchen Lin, Wai Lam, Graham Neubig, Yuanzhi Li! 👏 Check out more details in the Junpeng's thread👇
1:26

🚀Introducing VisualWebBench: A Comprehensive Benchmark for Multimodal Web Page Understanding and Grounding. 🤔What's this all about? Why this benchmark? > Back in Nov 2023, when we released MMMU ( a comprehensive multimodal understanding benchmark, we received feedback that it included very few UI screenshots. Considering the growing importance of UI understanding, especially with the rise of powerful agents like Devin ( which is built on the strong vision capability of #GPT4, we recognized the need for a benchmark focused on UI screenshot understanding.📸👀 > Multimodal #LLMs have significantly boosted web agents' performance on benchmarks like Mind2Web and WebArena. For instance, the SeeAct agent ( showcases the power of integrating vision into web agents. However, these benchmarks primarily evaluate the end-to-end task execution ability of web agents rather than their understanding of web pages. 🌉 Bridging the Gap with VisualWebBench > To provide a comprehensive evaluation of multimodal LLMs' web page understanding capabilities, we introduce VisualWebBench. Our benchmark spans 139 websites 🌐 across 12 domains 🏷️ and 87 sub-domains 🔍, ensuring a diverse and representative dataset. It assesses MLLMs at three levels: website-level, element-level, and action-level 📊, and encompasses seven tasks designed to evaluate understanding, OCR, grounding, and reasoning abilities 🧠💡. 😮 Surprising Findings > 🎉 Open-source models are catching up: Even though closed-source MLLMs are still leading the leaderboard, we are happy to see open-source models like LLaVA 1.6 34B achieve comparable performance to Gemini Pro. > 🧠 Grounding ability, crucial for developing MLLM-based web applications, is a weakness for most MLLMs. > 🖼️ Importance of Image Resolution: The limited image resolution handling capabilities of most open-source MLLMs restrict their utility in web scenarios, where rich text and elements are prevalent. > 🧱 Relatively strong correlation with general understanding benchmarks like MMMU but weak correlation with web agent benchmarks like Mind2Web. Web agent benchmarks primarily evaluate the end-to-end task execution ability of web agents, which involves a series of actions to accomplish a goal. In contrast, VisualWebBench emphasizes evaluating the foundational skills of MLLMs such as understanding and grounding web page elements. 💡Fun Fact > Claude Sonnet is better than Opus on our benchmark :) 🎓 Conclusion > VisualWebBench serves as a valuable resource for the community, driving research and development in the field of multimodal web page understanding and grounding. As MLLMs continue to evolve and improve, we look forward to seeing new applications and breakthroughs. We believe that our benchmark will contribute to the development of more powerful MLLMs in the web domain, ultimately leading to a more intuitive and efficient user experience on the web. Kudos to the student leads Junpeng Liu Yifan Song and the team Bill Yuchen Lin, Wai Lam, Graham Neubig, Yuanzhi Li! 👏 Check out more details in the Junpeng's thread👇

Xiang Yue

56,697 次观看 • 2 年前

New short course: Building Multimodal Search and RAG", by Weaviate AI Database's Sebastia(N_) Witalec ✊🏽✊🏾✊🏿. Contrastive learning is used to train models to map vectors into an embedding space by pulling similar concepts closer together and pushing dissimilar concepts away from each other. This technique is also used to train multimodal embedding models that capture semantic similarity across different modalities like text, images, and audio. These multimodal embeddings can be used to build multimodal search and RAG systems. In this course, you'll learn how contrastive learning works, and how to add multimodality to RAG – so your models can draw on diverse, relevant context to answer questions. For example, a query about a financial report might synthesize information from text snippets, graphs, tables, and slides. You will also learn how visual instruction tuning lets you integrate image understanding into language models, and build a multi-vector recommender system using Weaviate’s open-source vector database. Please sign up here:
3:25

New short course: Building Multimodal Search and RAG", by Weaviate AI Database's Sebastia(N_) Witalec ✊🏽✊🏾✊🏿. Contrastive learning is used to train models to map vectors into an embedding space by pulling similar concepts closer together and pushing dissimilar concepts away from each other. This technique is also used to train multimodal embedding models that capture semantic similarity across different modalities like text, images, and audio. These multimodal embeddings can be used to build multimodal search and RAG systems. In this course, you'll learn how contrastive learning works, and how to add multimodality to RAG – so your models can draw on diverse, relevant context to answer questions. For example, a query about a financial report might synthesize information from text snippets, graphs, tables, and slides. You will also learn how visual instruction tuning lets you integrate image understanding into language models, and build a multi-vector recommender system using Weaviate’s open-source vector database. Please sign up here:

Andrew Ng

104,451 次观看 • 2 年前

Achieve SOTA on multimodal, complex tasks with minimal supervision! Introducing In-Context Abstraction Learning (ICAL): a method that extracts insights from visual demos and human feedback to continually teach multimodal LLM agents 🚀🤖🧠 ICAL learns from sub-optimal demonstrations and human feedback by transforming noisy trajectories into optimized, annotated examples that improve multimodal decision-making in new domains. ICAL is the first to show state-of-the-art performance across three challenging domains: we obtain a 34% relative improvement on TEACh household robot instruction following from dialogue, a 59% relative improvement on Visual Web Arena for autonomous visual web agents, and beat few-shot GPT-4V and supervised baselines on Ego4D action forecasting.
0:30

Achieve SOTA on multimodal, complex tasks with minimal supervision! Introducing In-Context Abstraction Learning (ICAL): a method that extracts insights from visual demos and human feedback to continually teach multimodal LLM agents 🚀🤖🧠 ICAL learns from sub-optimal demonstrations and human feedback by transforming noisy trajectories into optimized, annotated examples that improve multimodal decision-making in new domains. ICAL is the first to show state-of-the-art performance across three challenging domains: we obtain a 34% relative improvement on TEACh household robot instruction following from dialogue, a 59% relative improvement on Visual Web Arena for autonomous visual web agents, and beat few-shot GPT-4V and supervised baselines on Ego4D action forecasting.

Gabriel Sarch

19,340 次观看 • 2 年前

JARVIS-1: Open-World Multi-task Agents with Memory-Augmented Multimodal Language Models paper page: Achieving human-like planning and control with multimodal observations in an open world is a key milestone for more functional generalist agents. Existing approaches can handle certain long-horizon tasks in an open world. However, they still struggle when the number of open-world tasks could potentially be infinite and lack the capability to progressively enhance task completion as game time progresses. We introduce JARVIS-1, an open-world agent that can perceive multimodal input (visual observations and human instructions), generate sophisticated plans, and perform embodied control, all within the popular yet challenging open-world Minecraft universe. Specifically, we develop JARVIS-1 on top of pre-trained multimodal language models, which map visual observations and textual instructions to plans. The plans will be ultimately dispatched to the goal-conditioned controllers. We outfit JARVIS-1 with a multimodal memory, which facilitates planning using both pre-trained knowledge and its actual game survival experiences. In our experiments, JARVIS-1 exhibits nearly perfect performances across over 200 varying tasks from the Minecraft Universe Benchmark, ranging from entry to intermediate levels. JARVIS-1 has achieved a completion rate of 12.5% in the long-horizon diamond pickaxe task. This represents a significant increase up to 5 times compared to previous records. Furthermore, we show that JARVIS-1 is able to self-improve following a life-long learning paradigm thanks to multimodal memory, sparking a more general intelligence and improved autonomy.
0:22

JARVIS-1: Open-World Multi-task Agents with Memory-Augmented Multimodal Language Models paper page: Achieving human-like planning and control with multimodal observations in an open world is a key milestone for more functional generalist agents. Existing approaches can handle certain long-horizon tasks in an open world. However, they still struggle when the number of open-world tasks could potentially be infinite and lack the capability to progressively enhance task completion as game time progresses. We introduce JARVIS-1, an open-world agent that can perceive multimodal input (visual observations and human instructions), generate sophisticated plans, and perform embodied control, all within the popular yet challenging open-world Minecraft universe. Specifically, we develop JARVIS-1 on top of pre-trained multimodal language models, which map visual observations and textual instructions to plans. The plans will be ultimately dispatched to the goal-conditioned controllers. We outfit JARVIS-1 with a multimodal memory, which facilitates planning using both pre-trained knowledge and its actual game survival experiences. In our experiments, JARVIS-1 exhibits nearly perfect performances across over 200 varying tasks from the Minecraft Universe Benchmark, ranging from entry to intermediate levels. JARVIS-1 has achieved a completion rate of 12.5% in the long-horizon diamond pickaxe task. This represents a significant increase up to 5 times compared to previous records. Furthermore, we show that JARVIS-1 is able to self-improve following a life-long learning paradigm thanks to multimodal memory, sparking a more general intelligence and improved autonomy.

AK

141,440 次观看 • 2 年前

Explore state-of-the-art multimodal prompting in our new short course Large Multimodal Model Prompting with Gemini, taught by Erwin Huizenga in collaboration with Google Cloud. One interesting insight from this course: with multimodal models, prompt structure matters significantly. Placing text inputs, such as a patient's medical history, before image inputs, like an X-ray, can enhance the model's ability to contextualize and interpret visual data effectively. In other contexts, such as image captioning, you may get better results by putting the image first. Multimodal models behave differently than text-only LLMs, and effective prompting for models varies depending on the model you’re using. In this course you’ll learn how to effectively prompt Gemini models. Gemini's multimodal capabilities also enable new approaches in AI application development, for example: - The Gemini library handles various video formats (MP4, MOV, MPEG), streamlining applications using these formats. - Large context window (up to 1 million tokens) enables processing of extensive content, like analyzing multiple 50-minute videos simultaneously. - Function calling feature integrates real-time data (e.g., current exchange rates) into model responses. The course demonstrates building multimodal applications with real-world examples including document analyzers that reason across text and graphs simultaneously, video content extractors that find and timestamp specific information from multiple hours of footage, and automated expense report systems processing receipt images while cross-referencing company policies. Sign up here:
2:49

Explore state-of-the-art multimodal prompting in our new short course Large Multimodal Model Prompting with Gemini, taught by Erwin Huizenga in collaboration with Google Cloud. One interesting insight from this course: with multimodal models, prompt structure matters significantly. Placing text inputs, such as a patient's medical history, before image inputs, like an X-ray, can enhance the model's ability to contextualize and interpret visual data effectively. In other contexts, such as image captioning, you may get better results by putting the image first. Multimodal models behave differently than text-only LLMs, and effective prompting for models varies depending on the model you’re using. In this course you’ll learn how to effectively prompt Gemini models. Gemini's multimodal capabilities also enable new approaches in AI application development, for example: - The Gemini library handles various video formats (MP4, MOV, MPEG), streamlining applications using these formats. - Large context window (up to 1 million tokens) enables processing of extensive content, like analyzing multiple 50-minute videos simultaneously. - Function calling feature integrates real-time data (e.g., current exchange rates) into model responses. The course demonstrates building multimodal applications with real-world examples including document analyzers that reason across text and graphs simultaneously, video content extractors that find and timestamp specific information from multiple hours of footage, and automated expense report systems processing receipt images while cross-referencing company policies. Sign up here:

Andrew Ng

74,060 次观看 • 2 年前

We’re excited to announce the release and open-source of HunyuanImage 3.0 — the largest and most powerful open-source text-to-image model to date, with over 80 billion total parameters, of which 13 billion are activated per token during inference.The effect is completely comparable to the industry’s flagship closed-source model.🚀🚀🚀 HunyuanImage 3.0 originates from our internally developed native multimodal large language model, with fine-tuning and post-training focused on text-to-image generation. This unique foundation gives the model a powerful set of capabilities: ✅Reason with world knowledge ✅Understand complex, thousand-word prompts ✅Generate precise text within images Different from traditional DiT architecture image generation models, HunyuanImage 3.0’s MoE architecture uses a Transfusion-based approach to deeply couple Diffusion and LLM training for a single, powerful system. Built on Hunyuan-A13B, HunyuanImage 3.0 was trained on a massive dataset: 5 billion image-text pairs, video frames, interleaved image-text data, and 6 trillion tokens of text corpora. This hybrid training across multimodal generation, understanding, and LLM capabilities allows the model to seamlessly integrate multiple tasks. Whether you're an illustrator, designer, or creator, this is built to slash your workflow from hours to minutes. HunyuanImage 3.0 can generate intricate text, detailed comics, expressive emojis, and lively, engaging illustrations for educational content. The current release focuses solely on text-to-image generation and future updates will include image-to-image, image editing, multi-turn interaction, and more. 👉🏻Try it now: 🔗GitHub: 🤗Hugging Face:
2:09

We’re excited to announce the release and open-source of HunyuanImage 3.0 — the largest and most powerful open-source text-to-image model to date, with over 80 billion total parameters, of which 13 billion are activated per token during inference.The effect is completely comparable to the industry’s flagship closed-source model.🚀🚀🚀 HunyuanImage 3.0 originates from our internally developed native multimodal large language model, with fine-tuning and post-training focused on text-to-image generation. This unique foundation gives the model a powerful set of capabilities: ✅Reason with world knowledge ✅Understand complex, thousand-word prompts ✅Generate precise text within images Different from traditional DiT architecture image generation models, HunyuanImage 3.0’s MoE architecture uses a Transfusion-based approach to deeply couple Diffusion and LLM training for a single, powerful system. Built on Hunyuan-A13B, HunyuanImage 3.0 was trained on a massive dataset: 5 billion image-text pairs, video frames, interleaved image-text data, and 6 trillion tokens of text corpora. This hybrid training across multimodal generation, understanding, and LLM capabilities allows the model to seamlessly integrate multiple tasks. Whether you're an illustrator, designer, or creator, this is built to slash your workflow from hours to minutes. HunyuanImage 3.0 can generate intricate text, detailed comics, expressive emojis, and lively, engaging illustrations for educational content. The current release focuses solely on text-to-image generation and future updates will include image-to-image, image editing, multi-turn interaction, and more. 👉🏻Try it now: 🔗GitHub: 🤗Hugging Face:

Tencent Hy

413,175 次观看 • 1 年前

Google presents AudioPaLM: A Large Language Model That Can Speak and Listen paper page: introduce AudioPaLM, a large language model for speech understanding and generation. AudioPaLM fuses text-based and speech-based language models, PaLM-2 [Anil et al., 2023] and AudioLM [Borsos et al., 2022], into a unified multimodal architecture that can process and generate text and speech with applications including speech recognition and speech-to-speech translation. AudioPaLM inherits the capability to preserve paralinguistic information such as speaker identity and intonation from AudioLM and the linguistic knowledge present only in text large language models such as PaLM-2. We demonstrate that initializing AudioPaLM with the weights of a text-only large language model improves speech processing, successfully leveraging the larger quantity of text training data used in pretraining to assist with the speech tasks. The resulting model significantly outperforms existing systems for speech translation tasks and has the ability to perform zero-shot speech-to-text translation for many languages for which input/target language combinations were not seen in training. AudioPaLM also demonstrates features of audio language models, such as transferring a voice across languages based on a short spoken prompt.
4:10

Google presents AudioPaLM: A Large Language Model That Can Speak and Listen paper page: introduce AudioPaLM, a large language model for speech understanding and generation. AudioPaLM fuses text-based and speech-based language models, PaLM-2 [Anil et al., 2023] and AudioLM [Borsos et al., 2022], into a unified multimodal architecture that can process and generate text and speech with applications including speech recognition and speech-to-speech translation. AudioPaLM inherits the capability to preserve paralinguistic information such as speaker identity and intonation from AudioLM and the linguistic knowledge present only in text large language models such as PaLM-2. We demonstrate that initializing AudioPaLM with the weights of a text-only large language model improves speech processing, successfully leveraging the larger quantity of text training data used in pretraining to assist with the speech tasks. The resulting model significantly outperforms existing systems for speech translation tasks and has the ability to perform zero-shot speech-to-text translation for many languages for which input/target language combinations were not seen in training. AudioPaLM also demonstrates features of audio language models, such as transferring a voice across languages based on a short spoken prompt.

AK

290,517 次观看 • 3 年前

Today we're announcing the open-source release of HunyuanVideo-Foley, our new end-to-end Text-Video-to-Audio (TV2A) framework for generating high-fidelity audio.🚀 This tool empowers creators in video production, filmmaking, and game development to generate professional-grade audio that precisely aligns with visual dynamics and semantic context, addressing key challenges in V2A generation.🔊 Key Innovations: 🔹Exceptional Generalization: Trained on a massive 100k-hour multimodal dataset, the model generates contextually-aware soundscapes for a wide range of scenes, from natural landscapes to animated shorts. 🔹Balanced Multimodal Response: Our innovative multimodal diffusion transformer (MMDiT) architecture ensures the model balances video and text cues, generating rich, layered sound effects that capture every detail—from the main subject to subtle background elements. 🔹High-Fidelity Audio: Using a Representation Alignment (REPA) loss function and a powerful Audio VAE, we've improved generation stability and producing professional-grade audio, free of noise and inconsistencies. HunyuanVideo-Foley achieves SOTA on multiple benchmarks, surpassing all open-source models in audio quality, visual-semantic alignment, and temporal alignment. 👉Try it now: 🌐Project Page: 🔗Code: 📄Technical Report: 🤗Hugging Face:
1:16

Today we're announcing the open-source release of HunyuanVideo-Foley, our new end-to-end Text-Video-to-Audio (TV2A) framework for generating high-fidelity audio.🚀 This tool empowers creators in video production, filmmaking, and game development to generate professional-grade audio that precisely aligns with visual dynamics and semantic context, addressing key challenges in V2A generation.🔊 Key Innovations: 🔹Exceptional Generalization: Trained on a massive 100k-hour multimodal dataset, the model generates contextually-aware soundscapes for a wide range of scenes, from natural landscapes to animated shorts. 🔹Balanced Multimodal Response: Our innovative multimodal diffusion transformer (MMDiT) architecture ensures the model balances video and text cues, generating rich, layered sound effects that capture every detail—from the main subject to subtle background elements. 🔹High-Fidelity Audio: Using a Representation Alignment (REPA) loss function and a powerful Audio VAE, we've improved generation stability and producing professional-grade audio, free of noise and inconsistencies. HunyuanVideo-Foley achieves SOTA on multiple benchmarks, surpassing all open-source models in audio quality, visual-semantic alignment, and temporal alignment. 👉Try it now: 🌐Project Page: 🔗Code: 📄Technical Report: 🤗Hugging Face:

Tencent Hy

122,881 次观看 • 1 年前

Introducing FoundationMotion. A large-scale, video-derived motion annotation dataset & auto-labeling pipeline + advanced models for motion understanding. Fully open-source: code, datasets, and models, free to use and build on. Understanding motion is core to physical reasoning, yet today’s leading models still struggle with simple spatial actions like “turn right” or “move up” or “flip the toast” - mainly due to the lack of large, fine-grained motion datasets. We present FoundationMotion, a fully automated pipeline that: • detects & tracks objects in videos • extracts trajectories • uses LLMs + frames to generate rich motion captions & QA pairs → creating large-scale, high-quality motion datasets at scale. After fine-tuning the open-source models Qwen and NVILA on our annotations, these models now outperform the closed-source Gemini-3-Flash and GPT-5.1 on spatial understanding tasks across autonomous driving, robotics, and everyday scenarios. 📜Paper: 🌐Webpage: 💻 Code: 🕸️Model: 📊 Dataset: 👉 Interactive Demo: Let’s move research forward together. FoundationMotion is also referred to as Wolf V2 🐺, the second chapter in the Wolf series:
1:25

Introducing FoundationMotion. A large-scale, video-derived motion annotation dataset & auto-labeling pipeline + advanced models for motion understanding. Fully open-source: code, datasets, and models, free to use and build on. Understanding motion is core to physical reasoning, yet today’s leading models still struggle with simple spatial actions like “turn right” or “move up” or “flip the toast” - mainly due to the lack of large, fine-grained motion datasets. We present FoundationMotion, a fully automated pipeline that: • detects & tracks objects in videos • extracts trajectories • uses LLMs + frames to generate rich motion captions & QA pairs → creating large-scale, high-quality motion datasets at scale. After fine-tuning the open-source models Qwen and NVILA on our annotations, these models now outperform the closed-source Gemini-3-Flash and GPT-5.1 on spatial understanding tasks across autonomous driving, robotics, and everyday scenarios. 📜Paper: 🌐Webpage: 💻 Code: 🕸️Model: 📊 Dataset: 👉 Interactive Demo: Let’s move research forward together. FoundationMotion is also referred to as Wolf V2 🐺, the second chapter in the Wolf series:

Boyi Li

67,643 次观看 • 9 个月前

“ xAI team is currently working heavily on coding models. Right now, the main focus is training a specialized coding model that will be both fast and smart. I believe we’ll share it with you guys in a few weeks. That’s exciting. Second ,after coding, we all see that the main weakness of Grok 4 is its multimodal capabilities. In fact, it was so bad that Grok was effectively looking at the world while squinting through glass trying to see blurry features and make sense of them. The most immediate improvement we’re stepping on with the next generation pre trained model is huge gains in image understanding, video understanding, and audio. Right now, the model can hear and see the world just like any of you. And with all the tools and other agents it can talk to, we’re going to see a huge unlock for many different application layers once multimodal agents arrive.” — xAI Team
2:01

“ xAI team is currently working heavily on coding models. Right now, the main focus is training a specialized coding model that will be both fast and smart. I believe we’ll share it with you guys in a few weeks. That’s exciting. Second ,after coding, we all see that the main weakness of Grok 4 is its multimodal capabilities. In fact, it was so bad that Grok was effectively looking at the world while squinting through glass trying to see blurry features and make sense of them. The most immediate improvement we’re stepping on with the next generation pre trained model is huge gains in image understanding, video understanding, and audio. Right now, the model can hear and see the world just like any of you. And with all the tools and other agents it can talk to, we’re going to see a huge unlock for many different application layers once multimodal agents arrive.” — xAI Team

Apurv Kochara

88,142 次观看 • 7 个月前

AI + robotics research is starting to pick up steam. We can now instruct Spot using natural language using Language-guided Skill Coordination (LSC). A user provides a natural language instruction: "Bring me the chocolates box, cereal box, and pill bottle, and put them on the bedroom table" and the robot navigates to the location of the target objects and places them on the room table. Current versions of Spot rely on a fixed-set vocabulary that cannot generalize to diverse instructions. In this demo, the researchers presented a method that uses large language models to receive a free-form natural language instruction for object rearrangement, which it then executes. It combines: -A voice-to-text model that processes the instructions into text -An LLM that converts natural language instruction to call a library of skills -A perception module that provides ground truth locations and visual object detection And what your left with is a glimpse into what our future will hold: robots executing tasks from human voice prompts.
0:30

AI + robotics research is starting to pick up steam. We can now instruct Spot using natural language using Language-guided Skill Coordination (LSC). A user provides a natural language instruction: "Bring me the chocolates box, cereal box, and pill bottle, and put them on the bedroom table" and the robot navigates to the location of the target objects and places them on the room table. Current versions of Spot rely on a fixed-set vocabulary that cannot generalize to diverse instructions. In this demo, the researchers presented a method that uses large language models to receive a free-form natural language instruction for object rearrangement, which it then executes. It combines: -A voice-to-text model that processes the instructions into text -An LLM that converts natural language instruction to call a library of skills -A perception module that provides ground truth locations and visual object detection And what your left with is a glimpse into what our future will hold: robots executing tasks from human voice prompts.

AI Breakfast

33,725 次观看 • 3 年前

For the first time in human history, we are teaching a Foundation Model to master the diverse tasks of medicinal chemists, biologists, and computational scientists all in one place. In our latest collaboration with Liquid AI, we are moving away from fragmented, specialized tools toward a single, super-intelligent model. What surprised me most? This model isn't just performing at reasonable levels—it has started outperforming specialist models across physics-based tasks, imaging, and longitudinal data. Why this changes everything: -Synergy over Specialization: Fine-tuning on specific tasks has unlocked unexpected capabilities in synergetic areas, opening a new frontier in multimodal AI research. -Zero-Shot Potential: We are building a model that can perform out-of-scope tasks, moving us closer to an "AI deity" for drug discovery. -Quality First: The goal isn't just to bypass regulations to save time; it’s about using these synergies to develop better, more effective drugs. We are no longer just looking at linear regression or simple text; we are looking at the future of how humanity fights disease. #LiquidAI #InsilicoMedicine #GenerativeAI #DrugDiscovery #DeepTech #BiotechInnovation
1:53

For the first time in human history, we are teaching a Foundation Model to master the diverse tasks of medicinal chemists, biologists, and computational scientists all in one place. In our latest collaboration with Liquid AI, we are moving away from fragmented, specialized tools toward a single, super-intelligent model. What surprised me most? This model isn't just performing at reasonable levels—it has started outperforming specialist models across physics-based tasks, imaging, and longitudinal data. Why this changes everything: -Synergy over Specialization: Fine-tuning on specific tasks has unlocked unexpected capabilities in synergetic areas, opening a new frontier in multimodal AI research. -Zero-Shot Potential: We are building a model that can perform out-of-scope tasks, moving us closer to an "AI deity" for drug discovery. -Quality First: The goal isn't just to bypass regulations to save time; it’s about using these synergies to develop better, more effective drugs. We are no longer just looking at linear regression or simple text; we are looking at the future of how humanity fights disease. #LiquidAI #InsilicoMedicine #GenerativeAI #DrugDiscovery #DeepTech #BiotechInnovation

Alex Zhavoronkov, PhD (aka Aleksandrs Zavoronkovs)

10,544 次观看 • 6 个月前

New short course Multimodal RAG: Chat with Videos, developed with Intel and taught by vasudevlal! In this course, you’ll work with LLaVA (Large Language and Vision Assistant), a Large Vision Language Model (LVLM) that can process both images and text. For example, given an image of a person doing a handstand on a skateboard at the beach, LLaVA doesn't just caption the scene, it’s able to predict possible outcomes, like the person losing balance or falling off. By understanding not just what's in a video frame, but what might happen next, your application can provide more insightful answers to questions about video. You'll build a full multimodal RAG pipeline that can chat about video content: - Use the BridgeTower model to create joint text-image embeddings in a 512-dimensional multimodal semantic space. - Learn video processing techniques to extract keyframes, generate transcripts using Whisper, and create captions. - Use the LanceDB vector database to store and retrieve high-dimensional multimodal embeddings. - Integrate the LLaVA model, combining CLIP's (Contrastive Language Image Pretraining) vision transformer with Llama, for advanced visual-textual reasoning. Your final system will ingest video data, generate embeddings for frames and text, perform similarity searches for relevant content, and use the retrieved multimodal context to inform LVLM-based response generation. The result is a system capable of answering nuanced questions about video content, effectively chatting about the video it has processed. Please sign up here!
2:40

New short course Multimodal RAG: Chat with Videos, developed with Intel and taught by vasudevlal! In this course, you’ll work with LLaVA (Large Language and Vision Assistant), a Large Vision Language Model (LVLM) that can process both images and text. For example, given an image of a person doing a handstand on a skateboard at the beach, LLaVA doesn't just caption the scene, it’s able to predict possible outcomes, like the person losing balance or falling off. By understanding not just what's in a video frame, but what might happen next, your application can provide more insightful answers to questions about video. You'll build a full multimodal RAG pipeline that can chat about video content: - Use the BridgeTower model to create joint text-image embeddings in a 512-dimensional multimodal semantic space. - Learn video processing techniques to extract keyframes, generate transcripts using Whisper, and create captions. - Use the LanceDB vector database to store and retrieve high-dimensional multimodal embeddings. - Integrate the LLaVA model, combining CLIP's (Contrastive Language Image Pretraining) vision transformer with Llama, for advanced visual-textual reasoning. Your final system will ingest video data, generate embeddings for frames and text, perform similarity searches for relevant content, and use the retrieved multimodal context to inform LVLM-based response generation. The result is a system capable of answering nuanced questions about video content, effectively chatting about the video it has processed. Please sign up here!

Andrew Ng

107,967 次观看 • 2 年前

We benchmarked leading multimodal foundation models (GPT-4o, Claude 3.5 Sonnet, Gemini, Llama, etc.) on standard computer vision tasks—from segmentation to surface normal estimation—using standard datasets like COCO and ImageNet. These models have made remarkable progress; however, it is unclear exactly where they stand in terms of understanding vision in detail. Especially when it comes to tasks beyond question-answering. How well do they understand an object's segments or geometry? Our analyses yield an assessment that is quantitatively and qualitatively detailed and is compatible with evaluations developed in the field of computer vision over the past decades. Observed trends: 🔹 The foundation models consistently underperform task-specific SOTA models across all tasks. However, they are respectable generalists, which is remarkable as they are presumably trained primarily on image-text-based tasks. 🔹 They perform semantic tasks notably better than geometric ones. 🔹 GPT-4o performs the best among non-reasoning models, getting the top position in 4 out of 6 tasks. 🔹 Reasoning models, e.g., o3, show improvements in geometric tasks. 🔹 The 'image generation' models, e.g., GPT-40 Image Generation, which have been natively trained multimodally, exhibit quirks. E.g., hallucinated objects, misalignment between the input and output, etc. 🔹 While the prompting techniques affect performance, better models exhibit less sensitivity to variations in prompts. We control for the variance introduced by the prompting methods in our experiments. 🌐 Detailed analyses, visualizations: ⌨️ code: 🧵 1/n
2:36

We benchmarked leading multimodal foundation models (GPT-4o, Claude 3.5 Sonnet, Gemini, Llama, etc.) on standard computer vision tasks—from segmentation to surface normal estimation—using standard datasets like COCO and ImageNet. These models have made remarkable progress; however, it is unclear exactly where they stand in terms of understanding vision in detail. Especially when it comes to tasks beyond question-answering. How well do they understand an object's segments or geometry? Our analyses yield an assessment that is quantitatively and qualitatively detailed and is compatible with evaluations developed in the field of computer vision over the past decades. Observed trends: 🔹 The foundation models consistently underperform task-specific SOTA models across all tasks. However, they are respectable generalists, which is remarkable as they are presumably trained primarily on image-text-based tasks. 🔹 They perform semantic tasks notably better than geometric ones. 🔹 GPT-4o performs the best among non-reasoning models, getting the top position in 4 out of 6 tasks. 🔹 Reasoning models, e.g., o3, show improvements in geometric tasks. 🔹 The 'image generation' models, e.g., GPT-40 Image Generation, which have been natively trained multimodally, exhibit quirks. E.g., hallucinated objects, misalignment between the input and output, etc. 🔹 While the prompting techniques affect performance, better models exhibit less sensitivity to variations in prompts. We control for the variance introduced by the prompting methods in our experiments. 🌐 Detailed analyses, visualizations: ⌨️ code: 🧵 1/n

Amir Zamir

73,434 次观看 • 1 年前

Robots might learn better from video than from language! 📼 Most Vision-Language-Action (VLA) models learn what to do from text, but still struggle with how things move in the real world. That makes them data-hungry and slow to train. mimic video takes a different route. Instead of grounding robot control in text, it grounds it in video, using large pre-trained video models that already capture physical motion and dynamics. The idea is straightforward: let the video model handle “what will happen next,” and let a smaller control model focus only on turning that visual plan into robot actions. The result is big gains in practice. Robots trained this way need 10× less data, converge twice as fast, and perform better on both simulated benchmarks and real bimanual manipulation tasks. If robots can “imagine” motion using video, control becomes a much simpler problem. Shoutout to Jonas Pai, Liam Achenbach, Oier Mees, Elvis Nava and the rest of the team! Here's the project page: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →
0:47

Robots might learn better from video than from language! 📼 Most Vision-Language-Action (VLA) models learn what to do from text, but still struggle with how things move in the real world. That makes them data-hungry and slow to train. mimic video takes a different route. Instead of grounding robot control in text, it grounds it in video, using large pre-trained video models that already capture physical motion and dynamics. The idea is straightforward: let the video model handle “what will happen next,” and let a smaller control model focus only on turning that visual plan into robot actions. The result is big gains in practice. Robots trained this way need 10× less data, converge twice as fast, and perform better on both simulated benchmarks and real bimanual manipulation tasks. If robots can “imagine” motion using video, control becomes a much simpler problem. Shoutout to Jonas Pai, Liam Achenbach, Oier Mees, Elvis Nava and the rest of the team! Here's the project page: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

50,122 次观看 • 9 个月前

LLaVA-3D A Simple yet Effective Pathway to Empowering LMMs with 3D-awareness Recent advancements in Large Multimodal Models (LMMs) have greatly enhanced their proficiency in 2D visual understanding tasks, enabling them to effectively process and understand images and videos. However, the development of LMMs with 3D-awareness for 3D scene understanding has been hindered by the lack of large-scale 3D vision-language datasets and powerful 3D encoders. In this paper, we introduce a simple yet effective framework called LLaVA-3D. Leveraging the strong 2D understanding priors from LLaVA, our LLaVA-3D efficiently adapts LLaVA for 3D scene understanding without compromising 2D understanding capabilities. To achieve this, we employ a simple yet effective representation, 3D Patch, which connects 2D CLIP patch features with their corresponding positions in 3D space. By integrating the 3D Patches into 2D LMMs and employing joint 2D and 3D vision-language instruction tuning, we establish a unified architecture for both 2D image understanding and 3D scene understanding. Experimental results show that LLaVA-3D converges 3.5x faster than existing 3D LMMs when trained on 3D vision-language datasets. Moreover, LLaVA-3D not only achieves state-of-the-art performance across various 3D tasks but also maintains comparable 2D image understanding and vision-language conversation capabilities with LLaVA.
0:31

LLaVA-3D A Simple yet Effective Pathway to Empowering LMMs with 3D-awareness Recent advancements in Large Multimodal Models (LMMs) have greatly enhanced their proficiency in 2D visual understanding tasks, enabling them to effectively process and understand images and videos. However, the development of LMMs with 3D-awareness for 3D scene understanding has been hindered by the lack of large-scale 3D vision-language datasets and powerful 3D encoders. In this paper, we introduce a simple yet effective framework called LLaVA-3D. Leveraging the strong 2D understanding priors from LLaVA, our LLaVA-3D efficiently adapts LLaVA for 3D scene understanding without compromising 2D understanding capabilities. To achieve this, we employ a simple yet effective representation, 3D Patch, which connects 2D CLIP patch features with their corresponding positions in 3D space. By integrating the 3D Patches into 2D LMMs and employing joint 2D and 3D vision-language instruction tuning, we establish a unified architecture for both 2D image understanding and 3D scene understanding. Experimental results show that LLaVA-3D converges 3.5x faster than existing 3D LMMs when trained on 3D vision-language datasets. Moreover, LLaVA-3D not only achieves state-of-the-art performance across various 3D tasks but also maintains comparable 2D image understanding and vision-language conversation capabilities with LLaVA.

AK

41,747 次观看 • 2 年前

⚡️We are excited to announce that our new no-code Enterprise Platform is NOW available in private beta! As RAG apps advance from prototype to production we’ve been overwhelmed by requests for an enterprise grade solution to provide these applications with the data they need. Designed to make it easy to get your data #RAGready, our Platform can preprocess more than 25 file types and soon will be fully #multimodal, also able to ingest audio, video and image files. We ship with a baseline suite of source connectors, including Amazon Web Services S3, Microsoft Azure Blob Storage, OneDrive, SFTP, Databricks Delta Table, Google Drive, Salesforce, Elastic, OpenSearch, and Google Cloud storage with many more fast following. Platform transforms your documents into a standardized JSON schema, broken down into semantically coherent elements allowing you to reconstruct your document in the manner most useful to you. Want only the narrative text but not the headers and footers? This is entirely configurable through the UI. Additionally, we generate more than 30 types of metadata for each element to make it easy to curate the data being written downstream and to support metadata filtering during retrieval. Smart chunking and the ability to choose from a range of embedding models are in from launch, delivering a turnkey solution for chunk and embedding experimentation. As for destination connectors, we've got that covered too, with Amazon Web Services S3, Pinecone, Chroma , Weaviate AI Database, Google Cloud storage, MongoDB, Microsoft Azure cognitive search, PostgreSQL, Elastic, OpenSearch, and Databricks Delta Table. And of course, all of this can be scheduled to keep your data continuously hydrated. The private-beta is live today! Sign-up to get access and come build the future of LLM data foundations with us: 🚀 #ETLforLLMs #AI #DataPreprocessing #DataScience #DataTransformation #LLMs #ETL #ML #PreppingData #MachineLearning #RAG #Engineer #Unstructured #Unstructuredio #RetrievalAugmentedGeneration #multimodal #AIJobs
1:00

⚡️We are excited to announce that our new no-code Enterprise Platform is NOW available in private beta! As RAG apps advance from prototype to production we’ve been overwhelmed by requests for an enterprise grade solution to provide these applications with the data they need. Designed to make it easy to get your data #RAGready, our Platform can preprocess more than 25 file types and soon will be fully #multimodal, also able to ingest audio, video and image files. We ship with a baseline suite of source connectors, including Amazon Web Services S3, Microsoft Azure Blob Storage, OneDrive, SFTP, Databricks Delta Table, Google Drive, Salesforce, Elastic, OpenSearch, and Google Cloud storage with many more fast following. Platform transforms your documents into a standardized JSON schema, broken down into semantically coherent elements allowing you to reconstruct your document in the manner most useful to you. Want only the narrative text but not the headers and footers? This is entirely configurable through the UI. Additionally, we generate more than 30 types of metadata for each element to make it easy to curate the data being written downstream and to support metadata filtering during retrieval. Smart chunking and the ability to choose from a range of embedding models are in from launch, delivering a turnkey solution for chunk and embedding experimentation. As for destination connectors, we've got that covered too, with Amazon Web Services S3, Pinecone, Chroma , Weaviate AI Database, Google Cloud storage, MongoDB, Microsoft Azure cognitive search, PostgreSQL, Elastic, OpenSearch, and Databricks Delta Table. And of course, all of this can be scheduled to keep your data continuously hydrated. The private-beta is live today! Sign-up to get access and come build the future of LLM data foundations with us: 🚀 #ETLforLLMs #AI #DataPreprocessing #DataScience #DataTransformation #LLMs #ETL #ML #PreppingData #MachineLearning #RAG #Engineer #Unstructured #Unstructuredio #RetrievalAugmentedGeneration #multimodal #AIJobs

Unstructured

21,913 次观看 • 2 年前

Voicebox: Text-Guided Multilingual Universal Speech Generation at Scale blog: Large-scale generative models such as GPT and DALL-E have revolutionized natural language processing and computer vision research. These models not only generate high fidelity text or image outputs, but are also generalists which can solve tasks not explicitly taught. In contrast, speech generative models are still primitive in terms of scale and task generalization. In this paper, we present Voicebox, the most versatile text-guided generative model for speech at scale. Voicebox is a non-autoregressive flow-matching model trained to infill speech, given audio context and text, trained on over 50K hours of speech that are neither filtered nor enhanced. Similar to GPT, Voicebox can perform many different tasks through in-context learning, but is more flexible as it can also condition on future context. Voicebox can be used for mono or cross-lingual zero-shot text-to-speech synthesis, noise removal, content editing, style conversion, and diverse sample generation. In particular, Voicebox outperforms the state-of-the-art zero-shot TTS model VALL-E on both intelligibility (5.9% vs 1.9% word error rates) and audio similarity (0.580 vs 0.681) while being up to 20 times faster.
2:43

Voicebox: Text-Guided Multilingual Universal Speech Generation at Scale blog: Large-scale generative models such as GPT and DALL-E have revolutionized natural language processing and computer vision research. These models not only generate high fidelity text or image outputs, but are also generalists which can solve tasks not explicitly taught. In contrast, speech generative models are still primitive in terms of scale and task generalization. In this paper, we present Voicebox, the most versatile text-guided generative model for speech at scale. Voicebox is a non-autoregressive flow-matching model trained to infill speech, given audio context and text, trained on over 50K hours of speech that are neither filtered nor enhanced. Similar to GPT, Voicebox can perform many different tasks through in-context learning, but is more flexible as it can also condition on future context. Voicebox can be used for mono or cross-lingual zero-shot text-to-speech synthesis, noise removal, content editing, style conversion, and diverse sample generation. In particular, Voicebox outperforms the state-of-the-art zero-shot TTS model VALL-E on both intelligibility (5.9% vs 1.9% word error rates) and audio similarity (0.580 vs 0.681) while being up to 20 times faster.

AK

429,355 次观看 • 3 年前

An agent model, plus two multimodal models that can see and operate computers. Nex-AGI releases Nex-N2.5 in mini, Pro, and Max sizes.📜 Apache 2.0. 🤖 🏆 Max leads the published BrowseComp comparison with 92.6. Pro ranks first on OSWorld-G with 87.4. 🖥️ The 35B mini and 397B Pro browse the web, operate software, verify results, and self-correct through visual feedback. 🧠 The 1.6T Max targets complex text-based agent tasks, from coding and tool use to research and knowledge work. It marks Nex-AGI’s first full post-training run at trillion-parameter scale. ⚙️ All three support function calling and direct, adaptive, or full thinking modes.
0:30

An agent model, plus two multimodal models that can see and operate computers. Nex-AGI releases Nex-N2.5 in mini, Pro, and Max sizes.📜 Apache 2.0. 🤖 🏆 Max leads the published BrowseComp comparison with 92.6. Pro ranks first on OSWorld-G with 87.4. 🖥️ The 35B mini and 397B Pro browse the web, operate software, verify results, and self-correct through visual feedback. 🧠 The 1.6T Max targets complex text-based agent tasks, from coding and tool use to research and knowledge work. It marks Nex-AGI’s first full post-training run at trillion-parameter scale. ⚙️ All three support function calling and direct, adaptive, or full thinking modes.

ModelScope

18,509 次观看 • 27 天前