正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Have vision-language models achieved human-level compositional reasoning? Our research suggests: not quite yet. We’re excited to present CREPE – a large-scale Compositional REPresentation Evaluation benchmark for vision-language models – as a 🌟highlight🌟at #CVPR2023. 🧵1/7

Zixian Ma

1,268 subscribers

21,382 次观看 • 3 年前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

7 条评论

Zixian Ma 的头像

Zixian Ma3 年前

🧵2/7 We are the amazing team of @zixianma02, @jhoer100, @momergul_, @monagandhi09, @irena_gao, and @RanjayKrishna. Kudos to @jhoer100 and @amengta for their designs 🫶 Check out our paper and code for more details! Paper: Code:

Zixian Ma 的头像

Zixian Ma3 年前

🧵3/7 CREPE measures two important aspects of compositionality identified in cognitive science literature: systematicity and productivity.

Zixian Ma 的头像

Zixian Ma3 年前

🧵4/7 Systematicity evaluates how well a model generalizes to unseen compositions.

Zixian Ma 的头像

Zixian Ma3 年前

🧵5/7 Productivity tests how well a model comprehends increasingly complex scenes and expressions.

Zixian Ma 的头像

Zixian Ma3 年前

🧵6/7 We devise two separate pipelines to generate test datasets for systematicity and productivity. Our datasets consist of a large quantity of image-text pairs and hard negative captions for image-to-text retrieval evaluation.

Zixian Ma 的头像

Zixian Ma3 年前

🧵7/7 Our evaluation of diverse VLMs shows that model performance degrades when compositions are unseen or when complexity increases, suggesting: 1. SOTA VLMs exhibit little systematicity or productivity; 2. Neither emerges as we scale up the pretraining dataset or model size.

harpreet 的头像

harpreet3 年前

Looking forward to chatting with you all about the paper on the podcast!

相关视频

New research from Meta FAIR: Large Concept Models (LCM) is a fundamentally different paradigm for language modeling that decouples reasoning from language representation, inspired by how humans can plan high-level thoughts to communicate.

New research from Meta FAIR: Large Concept Models (LCM) is a fundamentally different paradigm for language modeling that decouples reasoning from language representation, inspired by how humans can plan high-level thoughts to communicate.

AI at Meta

531,765 次观看 • 1 年前

🤨 Are Multimodal Large Language Models really as 𝐠𝐨𝐨𝐝 at 𝐜𝐡𝐚𝐫𝐭 𝐮𝐧𝐝𝐞𝐫𝐬𝐭𝐚𝐧𝐝𝐢𝐧𝐠 as existing benchmarks such as ChartQA suggest? 🚫 Our ℂ𝕙𝕒𝕣𝕏𝕚𝕧 benchmark suggests NO! 🥇Humans achieve ✨𝟖𝟎+% correctness. 🥈Sonnet 3.5 outperforms GPT-4o by 10+ points, reaching 🌟𝟔𝟎% correctness. 🥉Open-weight models are capped at ⭐𝟑𝟐% correctness. 🪜 Leaderboard: 📜 Preprint: 📊 Charxiv is ✨𝟏𝟎𝟎% handcrafted with rigorous human validation, and it reveals substantial gaps among Multimodal Large Language Models and humans in chart understanding. 🎥👇 80 second video (🎶sound on!). 🧶 1/6

🤨 Are Multimodal Large Language Models really as 𝐠𝐨𝐨𝐝 at 𝐜𝐡𝐚𝐫𝐭 𝐮𝐧𝐝𝐞𝐫𝐬𝐭𝐚𝐧𝐝𝐢𝐧𝐠 as existing benchmarks such as ChartQA suggest? 🚫 Our ℂ𝕙𝕒𝕣𝕏𝕚𝕧 benchmark suggests NO! 🥇Humans achieve ✨𝟖𝟎+% correctness. 🥈Sonnet 3.5 outperforms GPT-4o by 10+ points, reaching 🌟𝟔𝟎% correctness. 🥉Open-weight models are capped at ⭐𝟑𝟐% correctness. 🪜 Leaderboard: 📜 Preprint: 📊 Charxiv is ✨𝟏𝟎𝟎% handcrafted with rigorous human validation, and it reveals substantial gaps among Multimodal Large Language Models and humans in chart understanding. 🎥👇 80 second video (🎶sound on!). 🧶 1/6

Zirui "Colin" Wang

48,373 次观看 • 2 年前

🚀 Excited to announce the first release of a novel open source programming language and platform for language model interaction! Combining prompts, constraints & scripting, LMQL elevates the capabilities of large language models. 🧵1/6 A quick tour.

🚀 Excited to announce the first release of a novel open source programming language and platform for language model interaction! Combining prompts, constraints & scripting, LMQL elevates the capabilities of large language models. 🧵1/6 A quick tour.

LMQL (Language Model Query Language)

199,141 次观看 • 3 年前

Today we’re announcing two new updates in our computer vision work — a new, expanded license for our DINOv2 model and the release of FACET, a comprehensive new benchmark dataset to help evaluate and improve fairness in vision models. More details ➡️ 🧵

Today we’re announcing two new updates in our computer vision work — a new, expanded license for our DINOv2 model and the release of FACET, a comprehensive new benchmark dataset to help evaluate and improve fairness in vision models. More details ➡️ 🧵

AI at Meta

454,110 次观看 • 3 年前

Humans learn and improve from failures. Similarly, foundation models adapt based on human feedback. Can we leverage this failure understanding to enhance robotics systems that use foundation models? Introducing AHA—a vision-language model for detecting and reasoning over failures in robotic manipulation. Project page: 🧵Thread👇 Aha!

Humans learn and improve from failures. Similarly, foundation models adapt based on human feedback. Can we leverage this failure understanding to enhance robotics systems that use foundation models? Introducing AHA—a vision-language model for detecting and reasoning over failures in robotic manipulation. Project page: 🧵Thread👇 Aha!

Jiafei Duan

48,830 次观看 • 2 年前

Ever wondered what robots 🤖 could achieve if they could not just see – but also feel and hear? Introducing FuSe: a recipe for finetuning large vision-language-action (VLA) models with heterogeneous sensory data, such as vision, touch, sound, and more. Details in the thread 👇

Ever wondered what robots 🤖 could achieve if they could not just see – but also feel and hear? Introducing FuSe: a recipe for finetuning large vision-language-action (VLA) models with heterogeneous sensory data, such as vision, touch, sound, and more. Details in the thread 👇

Carlo Sferrazza

46,111 次观看 • 1 年前

Today’s a big day! We have something really exciting to share with the open-source community. We’re launching two open-weight language models: gpt-oss-120b and gpt-oss-20b. They’re incredible models, built for developers, trained for reasoning, efficiency, and real-world use.🧵

Today’s a big day! We have something really exciting to share with the open-source community. We’re launching two open-weight language models: gpt-oss-120b and gpt-oss-20b. They’re incredible models, built for developers, trained for reasoning, efficiency, and real-world use.🧵

Romain Huet

34,219 次观看 • 1 年前

Excited to share our new work: StreamingVLM! 🚀 We tackle a major challenge for Vision-Language Models (VLMs): understanding infinite video streams in real-time without latency blowing up or running out of memory. Paper: Code:

Excited to share our new work: StreamingVLM! 🚀 We tackle a major challenge for Vision-Language Models (VLMs): understanding infinite video streams in real-time without latency blowing up or running out of memory. Paper: Code:

Guangxuan Xiao

92,141 次观看 • 1 年前

🚀 Meta FAIR is releasing several new research artifacts on our road to advanced machine intelligence (AMI). These latest advancements are transforming our understanding of perception. 1️⃣ Meta Perception Encoder: A large-scale vision encoder that excels across several image & video tasks. 2️⃣ Meta Perception Language Model: A fully open & reproducible vision-language model designed to tackle visual recognition tasks. 3️⃣ Meta Locate 3D: An end-to-end model for accurate object localization in 3D environments. 4️⃣ Releasing model weights for our 8B-parameter Dynamic Byte Latent Transformer, an alternative to traditional tokenization methods with the potential to redefine the standards for language model efficiency and reliability. 5️⃣Collaborative Reasoner: A framework for evaluating & improving collaborative reasoning skills in language models. Download the code, datasets, and research papers and learn more about how these artifacts are paving the way for more efficient and accurate AI systems.➡️

🚀 Meta FAIR is releasing several new research artifacts on our road to advanced machine intelligence (AMI). These latest advancements are transforming our understanding of perception. 1️⃣ Meta Perception Encoder: A large-scale vision encoder that excels across several image & video tasks. 2️⃣ Meta Perception Language Model: A fully open & reproducible vision-language model designed to tackle visual recognition tasks. 3️⃣ Meta Locate 3D: An end-to-end model for accurate object localization in 3D environments. 4️⃣ Releasing model weights for our 8B-parameter Dynamic Byte Latent Transformer, an alternative to traditional tokenization methods with the potential to redefine the standards for language model efficiency and reliability. 5️⃣Collaborative Reasoner: A framework for evaluating & improving collaborative reasoning skills in language models. Download the code, datasets, and research papers and learn more about how these artifacts are paving the way for more efficient and accurate AI systems.➡️

AI at Meta

163,602 次观看 • 1 年前

The next frontier of autonomous driving is unlocked by reasoning models. NVIDIA Alpamayo brings together open AI models with reasoning capabilities, closed-loop simulation tools, and massive real-world driving datasets. Alpamayo 1 is a vision–language–action model that explains its own decisions through explicit reasoning traces, enabling trustworthy, humanlike decision-making. Together with NVIDIA’s Physical AI dataset and AlpaSim simulation, Alpamayo provides the tools and scale required to enable level 4 autonomous vehicles. ▶️ Watch now:

The next frontier of autonomous driving is unlocked by reasoning models. NVIDIA Alpamayo brings together open AI models with reasoning capabilities, closed-loop simulation tools, and massive real-world driving datasets. Alpamayo 1 is a vision–language–action model that explains its own decisions through explicit reasoning traces, enabling trustworthy, humanlike decision-making. Together with NVIDIA’s Physical AI dataset and AlpaSim simulation, Alpamayo provides the tools and scale required to enable level 4 autonomous vehicles. ▶️ Watch now:

NVIDIA DRIVE

35,324 次观看 • 9 个月前

DL3DV-10K: A Large-Scale Scene Dataset for Deep Learning-based 3D Vision paper page: We have witnessed significant progress in deep learning-based 3D vision, ranging from neural radiance field (NeRF) based 3D representation learning to applications in novel view synthesis (NVS). However, existing scene-level datasets for deep learning-based 3D vision, limited to either synthetic environments or a narrow selection of real-world scenes, are quite insufficient. This insufficiency not only hinders a comprehensive benchmark of existing methods but also caps what could be explored in deep learning-based 3D analysis. To address this critical gap, we present DL3DV-10K, a large-scale scene dataset, featuring 51.2 million frames from 10,510 videos captured from 65 types of point-of-interest (POI) locations, covering both bounded and unbounded scenes, with different levels of reflection, transparency, and lighting. We conducted a comprehensive benchmark of recent NVS methods on DL3DV-10K, which revealed valuable insights for future research in NVS. In addition, we have obtained encouraging results in a pilot study to learn generalizable NeRF from DL3DV-10K, which manifests the necessity of a large-scale scene-level dataset to forge a path toward a foundation model for learning 3D representation.

DL3DV-10K: A Large-Scale Scene Dataset for Deep Learning-based 3D Vision paper page: We have witnessed significant progress in deep learning-based 3D vision, ranging from neural radiance field (NeRF) based 3D representation learning to applications in novel view synthesis (NVS). However, existing scene-level datasets for deep learning-based 3D vision, limited to either synthetic environments or a narrow selection of real-world scenes, are quite insufficient. This insufficiency not only hinders a comprehensive benchmark of existing methods but also caps what could be explored in deep learning-based 3D analysis. To address this critical gap, we present DL3DV-10K, a large-scale scene dataset, featuring 51.2 million frames from 10,510 videos captured from 65 types of point-of-interest (POI) locations, covering both bounded and unbounded scenes, with different levels of reflection, transparency, and lighting. We conducted a comprehensive benchmark of recent NVS methods on DL3DV-10K, which revealed valuable insights for future research in NVS. In addition, we have obtained encouraging results in a pilot study to learn generalizable NeRF from DL3DV-10K, which manifests the necessity of a large-scale scene-level dataset to forge a path toward a foundation model for learning 3D representation.

AK

49,929 次观看 • 2 年前

1/ Introducing ᴏᴘᴇɴꜱᴄʜᴏʟᴀʀ: a retrieval-augmented LM to help scientists synthesize knowledge 📚 UW NLP Ai2 With open models & 45M-paper datastores, it outperforms proprietary systems & match human experts. Try out our demo! We also introduce ꜱᴄʜᴏʟᴀʀQᴀʙᴇɴᴄʜ, a new large-scale multi-domain benchmark for scientific research synthesis, covering CS, Bio and Physics.

1/ Introducing ᴏᴘᴇɴꜱᴄʜᴏʟᴀʀ: a retrieval-augmented LM to help scientists synthesize knowledge 📚 UW NLP Ai2 With open models & 45M-paper datastores, it outperforms proprietary systems & match human experts. Try out our demo! We also introduce ꜱᴄʜᴏʟᴀʀQᴀʙᴇɴᴄʜ, a new large-scale multi-domain benchmark for scientific research synthesis, covering CS, Bio and Physics.

Akari Asai

249,765 次观看 • 1 年前

Day 2 of 3 MLX Releases: Introducing Local Computer-Use 🚀🔥 A powerful tool built with MLX that uses Vision Language models and Voice models to control your Mac through visual understanding, planning and reasoning. Features ⚡️Automate your workflow with natural language 😎 Control your computer “hands-free” This project now supports both: 🤖 Level 1 (GUI Agent) 🧠 Level 2 (Autonomous GUI Agent) Get started: > pip install -U mlx-vlm mlx-audio mlx-whisper Please leave us a star and send a PR :)

Day 2 of 3 MLX Releases: Introducing Local Computer-Use 🚀🔥 A powerful tool built with MLX that uses Vision Language models and Voice models to control your Mac through visual understanding, planning and reasoning. Features ⚡️Automate your workflow with natural language 😎 Control your computer “hands-free” This project now supports both: 🤖 Level 1 (GUI Agent) 🧠 Level 2 (Autonomous GUI Agent) Get started: > pip install -U mlx-vlm mlx-audio mlx-whisper Please leave us a star and send a PR :)

Prince Canuma

45,867 次观看 • 1 年前

LP-MusicCaps: LLM-Based Pseudo Music Captioning paper page: Automatic music captioning, which generates natural language descriptions for given music tracks, holds significant potential for enhancing the understanding and organization of large volumes of musical data. Despite its importance, researchers face challenges due to the costly and time-consuming collection process of existing music-language datasets, which are limited in size. To address this data scarcity issue, we propose the use of large language models (LLMs) to artificially generate the description sentences from large-scale tag datasets. This results in approximately 2.2M captions paired with 0.5M audio clips. We term it Large Language Model based Pseudo music caption dataset, shortly, LP-MusicCaps. We conduct a systemic evaluation of the large-scale music captioning dataset with various quantitative evaluation metrics used in the field of natural language processing as well as human evaluation. In addition, we trained a transformer-based music captioning model with the dataset and evaluated it under zero-shot and transfer-learning settings. The results demonstrate that our proposed approach outperforms the supervised baseline model.

LP-MusicCaps: LLM-Based Pseudo Music Captioning paper page: Automatic music captioning, which generates natural language descriptions for given music tracks, holds significant potential for enhancing the understanding and organization of large volumes of musical data. Despite its importance, researchers face challenges due to the costly and time-consuming collection process of existing music-language datasets, which are limited in size. To address this data scarcity issue, we propose the use of large language models (LLMs) to artificially generate the description sentences from large-scale tag datasets. This results in approximately 2.2M captions paired with 0.5M audio clips. We term it Large Language Model based Pseudo music caption dataset, shortly, LP-MusicCaps. We conduct a systemic evaluation of the large-scale music captioning dataset with various quantitative evaluation metrics used in the field of natural language processing as well as human evaluation. In addition, we trained a transformer-based music captioning model with the dataset and evaluated it under zero-shot and transfer-learning settings. The results demonstrate that our proposed approach outperforms the supervised baseline model.

AK

78,794 次观看 • 3 年前

Video Game Bench introduce a research preview of VideoGameBench, a benchmark which challenges vision-language models to complete, in real-time, a suite of 20 different popular video games from both hand-held consoles and PC GPT-4o, Claude Sonnet 3.7, Gemini 2.5 Pro, and Gemini 2.0 Flash playing Doom II (default difficulty) on VideoGameBench-Lite with the same input prompt! Models achieve varying levels of success but none are able to pass even the first level.

Video Game Bench introduce a research preview of VideoGameBench, a benchmark which challenges vision-language models to complete, in real-time, a suite of 20 different popular video games from both hand-held consoles and PC GPT-4o, Claude Sonnet 3.7, Gemini 2.5 Pro, and Gemini 2.0 Flash playing Doom II (default difficulty) on VideoGameBench-Lite with the same input prompt! Models achieve varying levels of success but none are able to pass even the first level.

AK

154,281 次观看 • 1 年前

Super excited to share 🧠MLGym 🦾 – the first Gym environment for AI Research Agents 🤖🔬 We introduce MLGym and MLGym-Bench, a new framework and benchmark for evaluating and developing LLM agents on AI research tasks. The key contributions of our work are: 🕹️ Enables the exploration of different training algorithms for AI Research Agents such as RL 🛠️ Provides a flexible evaluation framework that can accommodate different artifacts such as models, algorithms, or predictions 🤖 Allows researchers to evaluate any model without the need to develop a custom agentic harness 🎯 Introduces 13 diverse open-ended AI Research tasks for evaluating AI Research Agents on a wide range of domains such as computer vision, natural language processing, reinforcement learning, game theory, and logical reasoning. 📈 Proposes a new evaluation metric for AI Research Agents MLGym makes it easy to: 1) Add new tasks 2) Evaluate new models 3) Integrate new agents Check out a video of the MLGym Agent to see how it performs the full pipeline of idea generation💡, implementation 👩‍💻, experimentation 👩‍🔬, and iteration 🔄 to improve on ML tasks. Huge thanks to the exceptionally talented Deepak Nathani who led this work and to all the other amazing collaborators who made this possible 🙏🫶🚀

Super excited to share 🧠MLGym 🦾 – the first Gym environment for AI Research Agents 🤖🔬 We introduce MLGym and MLGym-Bench, a new framework and benchmark for evaluating and developing LLM agents on AI research tasks. The key contributions of our work are: 🕹️ Enables the exploration of different training algorithms for AI Research Agents such as RL 🛠️ Provides a flexible evaluation framework that can accommodate different artifacts such as models, algorithms, or predictions 🤖 Allows researchers to evaluate any model without the need to develop a custom agentic harness 🎯 Introduces 13 diverse open-ended AI Research tasks for evaluating AI Research Agents on a wide range of domains such as computer vision, natural language processing, reinforcement learning, game theory, and logical reasoning. 📈 Proposes a new evaluation metric for AI Research Agents MLGym makes it easy to: 1) Add new tasks 2) Evaluate new models 3) Integrate new agents Check out a video of the MLGym Agent to see how it performs the full pipeline of idea generation💡, implementation 👩‍💻, experimentation 👩‍🔬, and iteration 🔄 to improve on ML tasks. Huge thanks to the exceptionally talented Deepak Nathani who led this work and to all the other amazing collaborators who made this possible 🙏🫶🚀

Roberta Raileanu

105,211 次观看 • 1 年前

MotionGPT: Human Motion as a Foreign Language paper page: Though the advancement of pre-trained large language models unfolds, the exploration of building a unified model for language and other multi-modal data, such as motion, remains challenging and untouched so far. Fortunately, human motion displays a semantic coupling akin to human language, often perceived as a form of body language. By fusing language data with large-scale motion models, motion-language pre-training that can enhance the performance of motion-related tasks becomes feasible. Driven by this insight, we propose MotionGPT, a unified, versatile, and user-friendly motion-language model to handle multiple motion-relevant tasks. Specifically, we employ the discrete vector quantization for human motion and transfer 3D motion into motion tokens, similar to the generation process of word tokens. Building upon this "motion vocabulary", we perform language modeling on both motion and text in a unified manner, treating human motion as a specific language. Moreover, inspired by prompt learning, we pre-train MotionGPT with a mixture of motion-language data and fine-tune it on prompt-based question-and-answer tasks. Extensive experiments demonstrate that MotionGPT achieves state-of-the-art performances on multiple motion tasks including text-driven motion generation, motion captioning, motion prediction, and motion in-between.

MotionGPT: Human Motion as a Foreign Language paper page: Though the advancement of pre-trained large language models unfolds, the exploration of building a unified model for language and other multi-modal data, such as motion, remains challenging and untouched so far. Fortunately, human motion displays a semantic coupling akin to human language, often perceived as a form of body language. By fusing language data with large-scale motion models, motion-language pre-training that can enhance the performance of motion-related tasks becomes feasible. Driven by this insight, we propose MotionGPT, a unified, versatile, and user-friendly motion-language model to handle multiple motion-relevant tasks. Specifically, we employ the discrete vector quantization for human motion and transfer 3D motion into motion tokens, similar to the generation process of word tokens. Building upon this "motion vocabulary", we perform language modeling on both motion and text in a unified manner, treating human motion as a specific language. Moreover, inspired by prompt learning, we pre-train MotionGPT with a mixture of motion-language data and fine-tune it on prompt-based question-and-answer tasks. Extensive experiments demonstrate that MotionGPT achieves state-of-the-art performances on multiple motion tasks including text-driven motion generation, motion captioning, motion prediction, and motion in-between.

AK

125,353 次观看 • 3 年前

I’m incredibly excited to share a sneak peak of Wayve’s GAIA-1, a major step forward in action and language-conditioned world models for #AutonomousDriving. Here’s a glimpse of what GAIA-1 can do - more in the 🧵. (These are not real videos!) But what's a world model?

I’m incredibly excited to share a sneak peak of Wayve’s GAIA-1, a major step forward in action and language-conditioned world models for #AutonomousDriving. Here’s a glimpse of what GAIA-1 can do - more in the 🧵. (These are not real videos!) But what's a world model?

Jamie Shotton

198,922 次观看 • 3 年前

Boston Dynamics has released a new video of their next-generation humanoid robot, Atlas. "We are excited to share some of our progress on developing Large Behavior Models for Atlas. This work is part of a collaboration between AI research teams at Toyota Research Institute (TRI) and Boston Dynamics. We have been building end-to-end language-conditioned policies that enable Atlas to accomplish long-horizon manipulation tasks."

Boston Dynamics has released a new video of their next-generation humanoid robot, Atlas. "We are excited to share some of our progress on developing Large Behavior Models for Atlas. This work is part of a collaboration between AI research teams at Toyota Research Institute (TRI) and Boston Dynamics. We have been building end-to-end language-conditioned policies that enable Atlas to accomplish long-horizon manipulation tasks."

Sawyer Merritt

200,744 次观看 • 1 年前