Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

To Adapt or Not to Adapt? Real-Time Adaptation for Semantic Segmentation paper page: The goal of Online Domain Adaptation for semantic segmentation is to handle unforeseeable domain changes that occur during deployment, like sudden weather events. However, the high computational costs associated with brute-force adaptation make this paradigm unfeasible... for real-world applications. In this paper we propose HAMLET, a Hardware-Aware Modular Least Expensive Training framework for real-time domain adaptation. Our approach includes a hardware-aware back-propagation orchestration agent (HAMT) and a dedicated domain-shift detector that enables active control over when and how the model is adapted (LT). Thanks to these advancements, our approach is capable of performing semantic segmentation while simultaneously adapting at more than 29FPS on a single consumer-grade GPU. Our framework's encouraging accuracy and speed trade-off is demonstrated on OnDA and SHIFT benchmarks through experimental results.show more

AK

482,159 subscribers

18,871 Aufrufe • vor 3 Jahren •via X (Twitter)

Wissenschaft & Technologie Bildung

Anya Rossi• Live Now

Private livecam show

1 Kommentare

Profilbild von Pier Luigi Dovesi

Pier Luigi Dovesivor 3 Jahren

💀 Project page: 📜 Arxiv: 🧑🏻‍💻 Code: 📽️ Video:

Ähnliche Videos

Tracking Anything with Decoupled Video Segmentation paper page: Training data for video segmentation are expensive to annotate. This impedes extensions of end-to-end algorithms to new video segmentation tasks, especially in large-vocabulary settings. To 'track anything' without training on video data for every individual task, we develop a decoupled video segmentation approach (DEVA), composed of task-specific image-level segmentation and class/task-agnostic bi-directional temporal propagation. Due to this design, we only need an image-level model for the target task (which is cheaper to train) and a universal temporal propagation model which is trained once and generalizes across tasks. To effectively combine these two modules, we use bi-directional propagation for (semi-)online fusion of segmentation hypotheses from different frames to generate a coherent segmentation. We show that this decoupled formulation compares favorably to end-to-end approaches in several data-scarce tasks including large-vocabulary video panoptic segmentation, open-world video segmentation, referring video segmentation, and unsupervised video object segmentation.

Tracking Anything with Decoupled Video Segmentation paper page: Training data for video segmentation are expensive to annotate. This impedes extensions of end-to-end algorithms to new video segmentation tasks, especially in large-vocabulary settings. To 'track anything' without training on video data for every individual task, we develop a decoupled video segmentation approach (DEVA), composed of task-specific image-level segmentation and class/task-agnostic bi-directional temporal propagation. Due to this design, we only need an image-level model for the target task (which is cheaper to train) and a universal temporal propagation model which is trained once and generalizes across tasks. To effectively combine these two modules, we use bi-directional propagation for (semi-)online fusion of segmentation hypotheses from different frames to generate a coherent segmentation. We show that this decoupled formulation compares favorably to end-to-end approaches in several data-scarce tasks including large-vocabulary video panoptic segmentation, open-world video segmentation, referring video segmentation, and unsupervised video object segmentation.

AK

305,633 Aufrufe • vor 2 Jahren

Tencent announces AppAgent Multimodal Agents as Smartphone Users paper page: Recent advancements in large language models (LLMs) have led to the creation of intelligent agents capable of performing complex tasks. This paper introduces a novel LLM-based multimodal agent framework designed to operate smartphone applications. Our framework enables the agent to operate smartphone applications through a simplified action space, mimicking human-like interactions such as tapping and swiping. This novel approach bypasses the need for system back-end access, thereby broadening its applicability across diverse apps. Central to our agent's functionality is its innovative learning method. The agent learns to navigate and use new apps either through autonomous exploration or by observing human demonstrations. This process generates a knowledge base that the agent refers to for executing complex tasks across different applications. To demonstrate the practicality of our agent, we conducted extensive testing over 50 tasks in 10 different applications, including social media, email, maps, shopping, and sophisticated image editing tools. The results affirm our agent's proficiency in handling a diverse array of high-level tasks.

Tencent announces AppAgent Multimodal Agents as Smartphone Users paper page: Recent advancements in large language models (LLMs) have led to the creation of intelligent agents capable of performing complex tasks. This paper introduces a novel LLM-based multimodal agent framework designed to operate smartphone applications. Our framework enables the agent to operate smartphone applications through a simplified action space, mimicking human-like interactions such as tapping and swiping. This novel approach bypasses the need for system back-end access, thereby broadening its applicability across diverse apps. Central to our agent's functionality is its innovative learning method. The agent learns to navigate and use new apps either through autonomous exploration or by observing human demonstrations. This process generates a knowledge base that the agent refers to for executing complex tasks across different applications. To demonstrate the practicality of our agent, we conducted extensive testing over 50 tasks in 10 different applications, including social media, email, maps, shopping, and sophisticated image editing tools. The results affirm our agent's proficiency in handling a diverse array of high-level tasks.

AK

343,834 Aufrufe • vor 2 Jahren

VideoRF: Rendering Dynamic Radiance Fields as 2D Feature Video Streams paper page: Neural Radiance Fields (NeRFs) excel in photorealistically rendering static scenes. However, rendering dynamic, long-duration radiance fields on ubiquitous devices remains challenging, due to data storage and computational constraints. In this paper, we introduce VideoRF, the first approach to enable real-time streaming and rendering of dynamic radiance fields on mobile platforms. At the core is a serialized 2D feature image stream representing the 4D radiance field all in one. We introduce a tailored training scheme directly applied to this 2D domain to impose the temporal and spatial redundancy of the feature image stream. By leveraging the redundancy, we show that the feature image stream can be efficiently compressed by 2D video codecs, which allows us to exploit video hardware accelerators to achieve real-time decoding. On the other hand, based on the feature image stream, we propose a novel rendering pipeline for VideoRF, which has specialized space mappings to query radiance properties efficiently. Paired with a deferred shading model, VideoRF has the capability of real-time rendering on mobile devices thanks to its efficiency. We have developed a real-time interactive player that enables online streaming and rendering of dynamic scenes, offering a seamless and immersive free-viewpoint experience across a range of devices, from desktops to mobile phones.

VideoRF: Rendering Dynamic Radiance Fields as 2D Feature Video Streams paper page: Neural Radiance Fields (NeRFs) excel in photorealistically rendering static scenes. However, rendering dynamic, long-duration radiance fields on ubiquitous devices remains challenging, due to data storage and computational constraints. In this paper, we introduce VideoRF, the first approach to enable real-time streaming and rendering of dynamic radiance fields on mobile platforms. At the core is a serialized 2D feature image stream representing the 4D radiance field all in one. We introduce a tailored training scheme directly applied to this 2D domain to impose the temporal and spatial redundancy of the feature image stream. By leveraging the redundancy, we show that the feature image stream can be efficiently compressed by 2D video codecs, which allows us to exploit video hardware accelerators to achieve real-time decoding. On the other hand, based on the feature image stream, we propose a novel rendering pipeline for VideoRF, which has specialized space mappings to query radiance properties efficiently. Paired with a deferred shading model, VideoRF has the capability of real-time rendering on mobile devices thanks to its efficiency. We have developed a real-time interactive player that enables online streaming and rendering of dynamic scenes, offering a seamless and immersive free-viewpoint experience across a range of devices, from desktops to mobile phones.

AK

38,686 Aufrufe • vor 2 Jahren

Introducing ALE-Bench, ALE-Agent! Towards Automating Long-Horizon Algorithm Engineering for Hard Optimization Problems Blog: Paper: ALE-Bench is a coding benchmark primarily focused on hard optimization (NP-hard) problems. We developed this benchmark with AtCoder Inc., a leading coding contest platform company. What makes ALE-Bench unique is its focus on hard optimization problems that demand long-horizon and creative reasoning. It’s open-ended, in the sense that true optima are out of reach (NP-hard) and scores can continuously improve. We believe this benchmark has the potential to become one of the key benchmarks for reasoning and coding in the next generation. ALE-Agent is our end-to-end agent that we specifically designed for this challenging domain. In fact, our ALE-Agent has already built an impressive track record in the wild! In May 2025, our agent participated in a live AtCoder Heuristic Competition (AHC), alongside 1,000 other participants in real-time. AHC is considered to be one of the most challenging coding competitions in this domain. Our ALE-Agent achieved an impressive ranking of 21st out of 1,000 human participants in the competition (top 2%), marking a turning point for AI discovery of solutions to hard optimization problems with a wide spectrum of important real world applications such as logistics, routing, packing, factory production planning, power-grid balancing. We look forward to applying this technology to real industrial optimization opportunities. Building on the insights from this study, Sakana AI will continue to tackle the challenge of developing AI with even greater algorithm engineering capabilities. ALE-Bench Dataset: ALE-Bench Code: This research was conducted in collaboration with AtCoder Inc. (AtCoder). We are deeply grateful for their outstanding expertise and contributions in optimization and algorithms, which were invaluable in providing data, analyzing results, and enabling our AI agent’s participation in their contests.

Introducing ALE-Bench, ALE-Agent! Towards Automating Long-Horizon Algorithm Engineering for Hard Optimization Problems Blog: Paper: ALE-Bench is a coding benchmark primarily focused on hard optimization (NP-hard) problems. We developed this benchmark with AtCoder Inc., a leading coding contest platform company. What makes ALE-Bench unique is its focus on hard optimization problems that demand long-horizon and creative reasoning. It’s open-ended, in the sense that true optima are out of reach (NP-hard) and scores can continuously improve. We believe this benchmark has the potential to become one of the key benchmarks for reasoning and coding in the next generation. ALE-Agent is our end-to-end agent that we specifically designed for this challenging domain. In fact, our ALE-Agent has already built an impressive track record in the wild! In May 2025, our agent participated in a live AtCoder Heuristic Competition (AHC), alongside 1,000 other participants in real-time. AHC is considered to be one of the most challenging coding competitions in this domain. Our ALE-Agent achieved an impressive ranking of 21st out of 1,000 human participants in the competition (top 2%), marking a turning point for AI discovery of solutions to hard optimization problems with a wide spectrum of important real world applications such as logistics, routing, packing, factory production planning, power-grid balancing. We look forward to applying this technology to real industrial optimization opportunities. Building on the insights from this study, Sakana AI will continue to tackle the challenge of developing AI with even greater algorithm engineering capabilities. ALE-Bench Dataset: ALE-Bench Code: This research was conducted in collaboration with AtCoder Inc. (AtCoder). We are deeply grateful for their outstanding expertise and contributions in optimization and algorithms, which were invaluable in providing data, analyzing results, and enabling our AI agent’s participation in their contests.

Sakana AI

237,195 Aufrufe • vor 1 Jahr

$FAU Erlangen-Nürnberg presents TRIPS Trilinear Point Splatting for Real-Time Radiance Field Rendering paper page: Point-based radiance field rendering has demonstrated impressive results for novel view synthesis, offering a compelling blend of rendering quality and computational efficiency. However, also latest approaches in this domain are not without their shortcomings. 3D Gaussian Splatting [Kerbl and Kopanas et al. 2023] struggles when tasked with rendering highly detailed scenes, due to blurring and cloudy artifacts. On the other hand, ADOP [R\"uckert et al. 2022] can accommodate crisper images, but the neural reconstruction network decreases performance, it grapples with temporal instability and it is unable to effectively address large gaps in the point cloud. In this paper, we present TRIPS (Trilinear Point Splatting), an approach that combines ideas from both Gaussian Splatting and ADOP. The fundamental concept behind our novel technique involves rasterizing points into a screen-space image pyramid, with the selection of the pyramid layer determined by the projected point size. This approach allows rendering arbitrarily large points using a single trilinear write. A lightweight neural network is then used to reconstruct a hole-free image including detail beyond splat resolution. Importantly, our render pipeline is entirely differentiable, allowing for automatic optimization of both point sizes and positions. Our evaluation demonstrate that TRIPS surpasses existing state-of-the-art methods in terms of rendering quality while maintaining a real-time frame rate of 60 frames per second on readily available hardware. This performance extends to challenging scenarios, such as scenes featuring intricate geometry, expansive landscapes, and auto-exposed footage.$

FAU Erlangen-Nürnberg presents TRIPS Trilinear Point Splatting for Real-Time Radiance Field Rendering paper page: Point-based radiance field rendering has demonstrated impressive results for novel view synthesis, offering a compelling blend of rendering quality and computational efficiency. However, also latest approaches in this domain are not without their shortcomings. 3D Gaussian Splatting [Kerbl and Kopanas et al. 2023] struggles when tasked with rendering highly detailed scenes, due to blurring and cloudy artifacts. On the other hand, ADOP [R\"uckert et al. 2022] can accommodate crisper images, but the neural reconstruction network decreases performance, it grapples with temporal instability and it is unable to effectively address large gaps in the point cloud. In this paper, we present TRIPS (Trilinear Point Splatting), an approach that combines ideas from both Gaussian Splatting and ADOP. The fundamental concept behind our novel technique involves rasterizing points into a screen-space image pyramid, with the selection of the pyramid layer determined by the projected point size. This approach allows rendering arbitrarily large points using a single trilinear write. A lightweight neural network is then used to reconstruct a hole-free image including detail beyond splat resolution. Importantly, our render pipeline is entirely differentiable, allowing for automatic optimization of both point sizes and positions. Our evaluation demonstrate that TRIPS surpasses existing state-of-the-art methods in terms of rendering quality while maintaining a real-time frame rate of 60 frames per second on readily available hardware. This performance extends to challenging scenarios, such as scenes featuring intricate geometry, expansive landscapes, and auto-exposed footage.

AK

45,459 Aufrufe • vor 2 Jahren

“Vegas is Dying” is more than a social media trend, it’s a reflection of real decline in travel and tourism to the greatest city on Earth. While this is temporarily challenging for our businesses, especially small local places, it presents a real opportunity for visitors looking to vacation on a budget. The reality is, Vegas is far from a ghost town, but even a 10% decline forces our resorts to offer value, deals and more affordable options and this shift will represent more of a buyers market that may help correct the luxury pricing market to a more consumer friendly experience. #vegas #lasvegas #vegasstarfish #travel #tourism #vegaslocal #budgettravel #creatorsearchinsights

“Vegas is Dying” is more than a social media trend, it’s a reflection of real decline in travel and tourism to the greatest city on Earth. While this is temporarily challenging for our businesses, especially small local places, it presents a real opportunity for visitors looking to vacation on a budget. The reality is, Vegas is far from a ghost town, but even a 10% decline forces our resorts to offer value, deals and more affordable options and this shift will represent more of a buyers market that may help correct the luxury pricing market to a more consumer friendly experience. #vegas #lasvegas #vegasstarfish #travel #tourism #vegaslocal #budgettravel #creatorsearchinsights

Jen G.

51,864 Aufrufe • vor 11 Monaten

#WATCH | Delhi: At the joint press conference with US Secretary of State Marco Rubio, EAM Dr S Jaishankar says, "Where defence and security cooperation is concerned, you're all aware that the 10-year major defence partnership framework agreement was recently renewed. A comprehensive underwater domain awareness roadmap was also signed. We discussed the importance of taking into account the Make in India approach and lessons drawn from recent conflicts while going forward in the defence domain. On the economic front, we spoke about the value of concluding at an early date the final text of the interim agreement regarding reciprocal and mutually beneficial trade. This will be an important step towards a comprehensive bilateral trade agreement, which was envisaged during Prime Minister Modi's visit in February 2025. We had a team recently in Washington, and our expectation is that an American team will be visiting India soon for that purpose. We spent some time today discussing energy issues, and again, you're all aware that our government's fundamental responsibility is to address the needs of 1.4 billion people. Ensuring the accessibility and affordability of energy for them is our prime objective. Secretary and I therefore welcome the expansion in our energy trade in recent months. Diversified supplies are at the heart of energy security for India. We devoted some time to nuclear energy cooperation as well. The passage of the Shanti Act has opened up new possibilities. An American delegation was recently in India. We hope to realise the potential of cooperation in the nuclear domain. I also raised with the Secretary some regulatory issues that we have on the American side..."

#WATCH | Delhi: At the joint press conference with US Secretary of State Marco Rubio, EAM Dr S Jaishankar says, "Where defence and security cooperation is concerned, you're all aware that the 10-year major defence partnership framework agreement was recently renewed. A comprehensive underwater domain awareness roadmap was also signed. We discussed the importance of taking into account the Make in India approach and lessons drawn from recent conflicts while going forward in the defence domain. On the economic front, we spoke about the value of concluding at an early date the final text of the interim agreement regarding reciprocal and mutually beneficial trade. This will be an important step towards a comprehensive bilateral trade agreement, which was envisaged during Prime Minister Modi's visit in February 2025. We had a team recently in Washington, and our expectation is that an American team will be visiting India soon for that purpose. We spent some time today discussing energy issues, and again, you're all aware that our government's fundamental responsibility is to address the needs of 1.4 billion people. Ensuring the accessibility and affordability of energy for them is our prime objective. Secretary and I therefore welcome the expansion in our energy trade in recent months. Diversified supplies are at the heart of energy security for India. We devoted some time to nuclear energy cooperation as well. The passage of the Shanti Act has opened up new possibilities. An American delegation was recently in India. We hope to realise the potential of cooperation in the nuclear domain. I also raised with the Secretary some regulatory issues that we have on the American side..."

ANI

26,917 Aufrufe • vor 1 Monat

At the panel session on “Rising Seas, Resilient Communities: Climate Adaptation Strategies in West Africa” at Harvard Business School earlier today, I had the opportunity to share Lagos State’s realities as a coastal megacity. Lagos sits on the continental shelf. Our land mass has grown from 3,577 to 4,050 square kilometres, largely through reclamation of wetlands and lagoon bodies. While this expansion has created space for real estate and infrastructure, it also comes with risks. This is why we insist that no reclamation takes place without Environmental Impact Assessments and drainage masterplans. These checks are necessary because when reclamation is done indiscriminately, it blocks stormwater pathways, destroys natural buffers like mangroves, and exposes millions of our people to flooding and erosion. As a city of over 22 million residents, we face additional pressures in managing daily waste. Lagos has long relied on a linear waste system with landfills, many of which are now decommissioned or about to be decommissioned . Today, we are deliberately shifting to a circular model that sees waste as a resource for wealth and energy. We have entered partnerships across the sub-region and beyond , including with a Ghanaian company that is helping us turn thousands of tonnes of waste into compost and reusable materials, while also setting up transfer loading stations to reduce what ends up in our landfills . We also recognize that Lagos is not alone in these challenges. The entire Gulf of Guinea shares similar threats. What happens in Togo, with its 42-mile coastline, or in Ghana, can affect us directly. That is why we have reached out to our neighbors and built regional partnerships on coastal protection, solid waste management, and climate adaptation. Our commitment is clear: to protect our environment, prosecute those who damage it, communicate with our citizens through advocacy on safe practices, and work with regional and international partners to build a Lagos that is not just surviving climate change but leading in adaptation strategies. These moves and developments are set to put Lagos on a #NewHorizon.

At the panel session on “Rising Seas, Resilient Communities: Climate Adaptation Strategies in West Africa” at Harvard Business School earlier today, I had the opportunity to share Lagos State’s realities as a coastal megacity. Lagos sits on the continental shelf. Our land mass has grown from 3,577 to 4,050 square kilometres, largely through reclamation of wetlands and lagoon bodies. While this expansion has created space for real estate and infrastructure, it also comes with risks. This is why we insist that no reclamation takes place without Environmental Impact Assessments and drainage masterplans. These checks are necessary because when reclamation is done indiscriminately, it blocks stormwater pathways, destroys natural buffers like mangroves, and exposes millions of our people to flooding and erosion. As a city of over 22 million residents, we face additional pressures in managing daily waste. Lagos has long relied on a linear waste system with landfills, many of which are now decommissioned or about to be decommissioned . Today, we are deliberately shifting to a circular model that sees waste as a resource for wealth and energy. We have entered partnerships across the sub-region and beyond , including with a Ghanaian company that is helping us turn thousands of tonnes of waste into compost and reusable materials, while also setting up transfer loading stations to reduce what ends up in our landfills . We also recognize that Lagos is not alone in these challenges. The entire Gulf of Guinea shares similar threats. What happens in Togo, with its 42-mile coastline, or in Ghana, can affect us directly. That is why we have reached out to our neighbors and built regional partnerships on coastal protection, solid waste management, and climate adaptation. Our commitment is clear: to protect our environment, prosecute those who damage it, communicate with our citizens through advocacy on safe practices, and work with regional and international partners to build a Lagos that is not just surviving climate change but leading in adaptation strategies. These moves and developments are set to put Lagos on a #NewHorizon.

Tokunbo Wahab

10,221 Aufrufe • vor 10 Monaten

Dominion energy is trying to use eminent domain to take more homes in Virginia They also put in for a proposal for a high-voltage power pole that's taller than the Statue of Liberty that would run straight over the homes of an extremely nice neighborhood The other proposal would see power lines run though an elementary school and high school Apparently it’s hard to use eminent domain and or run lines through schools because it would require school board approval Homeowners in the neighbored are fighting back by actually donating some of their land the new lines would run through to the school, so that it would be harder to take with eminent domain Imagine having to donator your own land to a school so a private company has a harder time uses eminent domain to steal your land…. This is where we are as a country… There are multiple proposed routes, one could impact 69 homes, another could impact 182 homes

Dominion energy is trying to use eminent domain to take more homes in Virginia They also put in for a proposal for a high-voltage power pole that's taller than the Statue of Liberty that would run straight over the homes of an extremely nice neighborhood The other proposal would see power lines run though an elementary school and high school Apparently it’s hard to use eminent domain and or run lines through schools because it would require school board approval Homeowners in the neighbored are fighting back by actually donating some of their land the new lines would run through to the school, so that it would be harder to take with eminent domain Imagine having to donator your own land to a school so a private company has a harder time uses eminent domain to steal your land…. This is where we are as a country… There are multiple proposed routes, one could impact 69 homes, another could impact 182 homes

Wall Street Apes

42,719 Aufrufe • vor 4 Tagen

GDN: The $KNDX Solution No more hardware bottlenecks Kondux GDN offers streamlined solutions to the problems users may face these days, such as limited access to state-of-the-art hardware & GPU power. We solve this by offloading heavy computational tasks to GDN while streaming 3D assets in real time to any device, regardless of local device capabilities. This means users can easily perform complex activities like high performance gaming, 3D VFX rendering & digital twin simulations for manufacturing without reliance on local GPU power & hardware. For creators, developers & enterprises, this eliminates massive roadblocks for their creative work flow. Instant collaboration through blockchain security In addition, our smart contracts enables sharing digital assets instant and secure, allowing for easy collaboration across multiple platforms. By utilizing our DNA technology, viewing a 3D asset doesn’t automatically grant ownership, and ensures creators 3D assets are protected from unauthorized use. Our kNFT tech brings life, personality & evolution to digital assets. Part of our mission is facilitating the shift of conversation from only traditional static NFTs, towards dynamic high-fidelity 3D assets with real world utility. Advanced interactivity & immersion in real time for all.

GDN: The $KNDX Solution No more hardware bottlenecks Kondux GDN offers streamlined solutions to the problems users may face these days, such as limited access to state-of-the-art hardware & GPU power. We solve this by offloading heavy computational tasks to GDN while streaming 3D assets in real time to any device, regardless of local device capabilities. This means users can easily perform complex activities like high performance gaming, 3D VFX rendering & digital twin simulations for manufacturing without reliance on local GPU power & hardware. For creators, developers & enterprises, this eliminates massive roadblocks for their creative work flow. Instant collaboration through blockchain security In addition, our smart contracts enables sharing digital assets instant and secure, allowing for easy collaboration across multiple platforms. By utilizing our DNA technology, viewing a 3D asset doesn’t automatically grant ownership, and ensures creators 3D assets are protected from unauthorized use. Our kNFT tech brings life, personality & evolution to digital assets. Part of our mission is facilitating the shift of conversation from only traditional static NFTs, towards dynamic high-fidelity 3D assets with real world utility. Advanced interactivity & immersion in real time for all.

Kondux

50,669 Aufrufe • vor 1 Jahr

So ai.com has a countdown timer hitting zero at Super Bowl kickoff tonight Here's why this is insane: → Kris bought the AI domain for $70M in crypto. The largest publicly disclosed domain sale in history → The domain previously redirected to ChatGPT, then xAI/Grok, then DeepSeek. Now it's launching a consumer AI agent platform → ~128M people will watch the Super Bowl tonight (in 16 hours). 18% of viewers say the ads are the main reason they tune in → A 30-second Super Bowl ad costs ~$8M this year, with some going for $10M+ Tech ad spending is 2x what it was during the 2022 "Crypto Bowl" when FTX, Crypto.com and others flooded the broadcast right before the market imploded → The product: a personal AI agent that can message, trade stocks, build projects, with a "decentralised network of billions of agents" vision This could be the mass onboarding event for personal AI Agents - the alpha will be getting your handle before it’s not available → Marszalek is running the exact same playbook as CryptoDotCom: buy a killer domain, splash on marketing, ride mainstream adoption. In 2021, he spent $700M on Staples Center naming rights, for one of the most valuable sports venue naming rights in history. Built a $1.5B revenue business → He paid for the domain entirely in crypto: ai.com traffic is about to go absolutely parabolic One of the most recognisable URLs on the internet + 128M eyeballs + a Super Bowl ad = the biggest single-day domain launch in history? This is going to provide a huge uptick in visibility to personal agents Yes we’ve seen the same playbook flop for crypto, but this time it’s far more interesting + useful to the average persons daily lives Compound this with the continued trajectory of OpenClaw developments and I strongly believe we are about to hit an inflection point in adoption

So ai.com has a countdown timer hitting zero at Super Bowl kickoff tonight Here's why this is insane: → Kris bought the AI domain for $70M in crypto. The largest publicly disclosed domain sale in history → The domain previously redirected to ChatGPT, then xAI/Grok, then DeepSeek. Now it's launching a consumer AI agent platform → ~128M people will watch the Super Bowl tonight (in 16 hours). 18% of viewers say the ads are the main reason they tune in → A 30-second Super Bowl ad costs ~$8M this year, with some going for $10M+ Tech ad spending is 2x what it was during the 2022 "Crypto Bowl" when FTX, Crypto.com and others flooded the broadcast right before the market imploded → The product: a personal AI agent that can message, trade stocks, build projects, with a "decentralised network of billions of agents" vision This could be the mass onboarding event for personal AI Agents - the alpha will be getting your handle before it’s not available → Marszalek is running the exact same playbook as CryptoDotCom: buy a killer domain, splash on marketing, ride mainstream adoption. In 2021, he spent $700M on Staples Center naming rights, for one of the most valuable sports venue naming rights in history. Built a $1.5B revenue business → He paid for the domain entirely in crypto: ai.com traffic is about to go absolutely parabolic One of the most recognisable URLs on the internet + 128M eyeballs + a Super Bowl ad = the biggest single-day domain launch in history? This is going to provide a huge uptick in visibility to personal agents Yes we’ve seen the same playbook flop for crypto, but this time it’s far more interesting + useful to the average persons daily lives Compound this with the continued trajectory of OpenClaw developments and I strongly believe we are about to hit an inflection point in adoption

0xSammy

91,296 Aufrufe • vor 5 Monaten

Introducing SoftMatcha 2: A Fast and Soft Pattern Matcher for Trillion-Scale Pre-Training Corpora What lies within a trillion-scale pre-training corpus? Can you truly guarantee your benchmarks are uncontaminated simply because there are no exact string matches? Alongside several research institutions in Japan, Sakana AI is proud to have collaborated in the development of SoftMatcha 2, an ultra-fast and flexible search tool that enables search over trillion-scale natural language corpora in under 0.3 seconds, even while handling semantic variations (substitution, insertion, and deletion). No existing tool meets all these criteria, including infini-gram-mini (EMNLP’25 Best Paper) or the original SoftMatcha (ICLR’25). Our approach employs string matching based on suffix arrays that scales well with corpus size. To mitigate the combinatorial explosion induced by the semantic relaxation of queries, our method is built on two key algorithmic ideas: fast exact lookup enabled by a disk-aware design, and dynamic corpus-aware pruning. As a practical application, we demonstrate that SoftMatcha 2 identifies potential benchmark contamination in pre-training corpora that existing exact-match approaches miss. You can try searching through a 100B-scale corpus via our online demo. The system remains blazingly fast even on trillion-token corpora, so we encourage you to host it yourself for larger scales. Demo: Paper: Code: This work is a collaboration with researchers from the University of Tokyo, NII, Kyoto University, SOKENDAI, NINJAL, Tohoku University, and RIKEN.

Introducing SoftMatcha 2: A Fast and Soft Pattern Matcher for Trillion-Scale Pre-Training Corpora What lies within a trillion-scale pre-training corpus? Can you truly guarantee your benchmarks are uncontaminated simply because there are no exact string matches? Alongside several research institutions in Japan, Sakana AI is proud to have collaborated in the development of SoftMatcha 2, an ultra-fast and flexible search tool that enables search over trillion-scale natural language corpora in under 0.3 seconds, even while handling semantic variations (substitution, insertion, and deletion). No existing tool meets all these criteria, including infini-gram-mini (EMNLP’25 Best Paper) or the original SoftMatcha (ICLR’25). Our approach employs string matching based on suffix arrays that scales well with corpus size. To mitigate the combinatorial explosion induced by the semantic relaxation of queries, our method is built on two key algorithmic ideas: fast exact lookup enabled by a disk-aware design, and dynamic corpus-aware pruning. As a practical application, we demonstrate that SoftMatcha 2 identifies potential benchmark contamination in pre-training corpora that existing exact-match approaches miss. You can try searching through a 100B-scale corpus via our online demo. The system remains blazingly fast even on trillion-token corpora, so we encourage you to host it yourself for larger scales. Demo: Paper: Code: This work is a collaboration with researchers from the University of Tokyo, NII, Kyoto University, SOKENDAI, NINJAL, Tohoku University, and RIKEN.

Sakana AI

101,614 Aufrufe • vor 5 Monaten

OpenAI just launched Operator. This got me thinking about the potential impact of AI agents on how buyers will discover domain names in the future, so I decided to give Operator a test run. The results were eye-opening. I started with a simple prompt: Help me find a great domain for my pet food brand from It should be a cute name and ideally under $5000." Operator took it from there. It accessed searched for pet food-related domains, applied the appropriate pricing filter, and shortlisted names that exuded cuteness. It didn’t stop there—it even assisted me in negotiating a better price through live chat. The implications of this technology are huge. For marketplaces, it will become critical to optimize the way domains are merchandised to ensure AI agents can easily discover relevant options. This means having in-depth filters, a highly effective search, and real time communication tools will matter more than ever. It is time to start adapting for this shift

Darpan

11,133 Aufrufe • vor 1 Jahr

Right now, many may think that in modern warfare the main thing is simply to have the tools – interceptors, different types of drones, electronic warfare systems. But the reality of modern war is that military technology changes very quickly. It is not enough just to have the tools – you also need real experience using them and the same speed of adaptation as the threat itself. Right now, only Ukraine has this experience, only our soldiers do. And that is why it is important not to waste time or money, but to build joint agreements and joint results. This should also be reflected in the SAFE program, which gives Europe the opportunity to rearm now and adapt our European defense production to modern challenges. “Together” is the key word for Europe so that everyone is protected. From an intervention at the Bucharest Nine Summit. (3/5)

Right now, many may think that in modern warfare the main thing is simply to have the tools – interceptors, different types of drones, electronic warfare systems. But the reality of modern war is that military technology changes very quickly. It is not enough just to have the tools – you also need real experience using them and the same speed of adaptation as the threat itself. Right now, only Ukraine has this experience, only our soldiers do. And that is why it is important not to waste time or money, but to build joint agreements and joint results. This should also be reflected in the SAFE program, which gives Europe the opportunity to rearm now and adapt our European defense production to modern challenges. “Together” is the key word for Europe so that everyone is protected. From an intervention at the Bucharest Nine Summit. (3/5)

Volodymyr Zelenskyy / Володимир Зеленський

154,431 Aufrufe • vor 2 Monaten

We are thrilled to announce the launch of Hoppy Racing, a new HOPPY DIVISION dedicated to bringing the adrenaline of racing and motorsport events to the meme Space. This Strategic move aims to push HOPPY by sponsoring real-world racing events and partnerships with motorsport teams to ensure Hoppy Racing becomes a global household name, showcasing the power of $HOPPY to the world. Through our IRL Marketing, This will bring more visibility and legitimacy to new investors who may have previously been unaware of the HOPPY token. Our first supercar event will take place this Quarter in one of the world’s most iconic cities Dubai as it is known for its crypto adaptation, luxury, and global appeal, which is the perfect stage to launch Hoppy Racing. Stay tuned for more details and exclusive updates.

We are thrilled to announce the launch of Hoppy Racing, a new HOPPY DIVISION dedicated to bringing the adrenaline of racing and motorsport events to the meme Space. This Strategic move aims to push HOPPY by sponsoring real-world racing events and partnerships with motorsport teams to ensure Hoppy Racing becomes a global household name, showcasing the power of $HOPPY to the world. Through our IRL Marketing, This will bring more visibility and legitimacy to new investors who may have previously been unaware of the HOPPY token. Our first supercar event will take place this Quarter in one of the world’s most iconic cities Dubai as it is known for its crypto adaptation, luxury, and global appeal, which is the perfect stage to launch Hoppy Racing. Stay tuned for more details and exclusive updates.

hoppy

43,919 Aufrufe • vor 1 Jahr

First, we are capable of producing at least 2,000 effective and combat-proven interceptors every day. We can produce more – it depends on investment. We need about 1,000 interceptors a day, and we can supply at least another 1,000 a day to our allies. Second, we know how to build radar and acoustic coverage to respond to how “shaheds” and other drones approach. Third, we have software that allows radars to keep working even under electronic warfare jamming. In real time, we analyze enemy frequencies and respond to them. And because of this system, we understand how effective our defense is against almost every attack drone, and we can move our positions and air defense to get better results. This iPad gives full control of the situation because we have a system for using our defense tools. If a “shahed” needs to be stopped in the Emirates – we can do it. If it needs to be stopped in Europe or the United Kingdom – we can do it. It is a matter of technology, investment, and cooperation. And the fact that we got through this winter – which Russia tried to make deadly for all our families – shows that our solutions work. From my address to the Parliament of the United Kingdom (4/7)

First, we are capable of producing at least 2,000 effective and combat-proven interceptors every day. We can produce more – it depends on investment. We need about 1,000 interceptors a day, and we can supply at least another 1,000 a day to our allies. Second, we know how to build radar and acoustic coverage to respond to how “shaheds” and other drones approach. Third, we have software that allows radars to keep working even under electronic warfare jamming. In real time, we analyze enemy frequencies and respond to them. And because of this system, we understand how effective our defense is against almost every attack drone, and we can move our positions and air defense to get better results. This iPad gives full control of the situation because we have a system for using our defense tools. If a “shahed” needs to be stopped in the Emirates – we can do it. If it needs to be stopped in Europe or the United Kingdom – we can do it. It is a matter of technology, investment, and cooperation. And the fact that we got through this winter – which Russia tried to make deadly for all our families – shows that our solutions work. From my address to the Parliament of the United Kingdom (4/7)

Volodymyr Zelenskyy / Володимир Зеленський

697,102 Aufrufe • vor 4 Monaten

During the last two days, the 32 Allies of #NATO met in the Chiefs of Defence format in Brussels, to discuss our unified military approach to the security developments of our time. Through SACEUR’s assessment of readiness, deterrence and defence, and SACT’s views on near and far future, we communicate to our citizens and the rest of the world that we are ready for the present and we are prepared for whatever lies on the horizon. Our focus on partnerships, from across the globe, also demonstrates our understanding that our security is global and interconnected, and #NATO must reflect on this reality. And our support for #Ukraine through initiatives like NSATU, JATEC and PURL, continues to grow, showing that our unity and determination are more than words. From our discussions, three main conclusions stand out: We are #united. We are #resilient. We are #dynamic, as we #adapt succesfully, while we #deter effectively! #StrongerTogether #SlavaUkraïni

During the last two days, the 32 Allies of #NATO met in the Chiefs of Defence format in Brussels, to discuss our unified military approach to the security developments of our time. Through SACEUR’s assessment of readiness, deterrence and defence, and SACT’s views on near and far future, we communicate to our citizens and the rest of the world that we are ready for the present and we are prepared for whatever lies on the horizon. Our focus on partnerships, from across the globe, also demonstrates our understanding that our security is global and interconnected, and #NATO must reflect on this reality. And our support for #Ukraine through initiatives like NSATU, JATEC and PURL, continues to grow, showing that our unity and determination are more than words. From our discussions, three main conclusions stand out: We are #united. We are #resilient. We are #dynamic, as we #adapt succesfully, while we #deter effectively! #StrongerTogether #SlavaUkraïni

Admiral Giuseppe CAVO DRAGONE - NATO CMC

36,518 Aufrufe • vor 5 Monaten

We are thrilled to share that our first paper from my new lab, Spateo ( for spatiotemporal modeling of molecular holograms, is now online in Cell: Spateo is a comprehensive analytical framework for 3D whole-embryo spatiotemporal modeling. Its advanced features include: • 3D alignment and reconstruction at the whole-mouse-embryo scale (see the animation). • 3D spatial domain digitization and cell-cell communication analysis to understand spatial gene expression gradients and both inter- and intracellular communication. • 3D morphometric and volumetric analyses along with 3D morphogenesis vector field modeling to quantify dynamics such as surface area, volume, and cell density across organs, and to dissect the interplay between morphogenesis factors and cell migration. • A “Google Earth”-like browser, Spateo-viewer ( and for interactive and intuitive exploration of 3D spatial data. • Additional features, such as RNA signal-based single-cell segmentation. We are also honored that Nature “News and Views” has highlighted this work as well: This is really an amazing outcome after two years' heroic revision process that rewrite the entire paper using a new data ( for whole mouse embryos.

We are thrilled to share that our first paper from my new lab, Spateo ( for spatiotemporal modeling of molecular holograms, is now online in Cell: Spateo is a comprehensive analytical framework for 3D whole-embryo spatiotemporal modeling. Its advanced features include: • 3D alignment and reconstruction at the whole-mouse-embryo scale (see the animation). • 3D spatial domain digitization and cell-cell communication analysis to understand spatial gene expression gradients and both inter- and intracellular communication. • 3D morphometric and volumetric analyses along with 3D morphogenesis vector field modeling to quantify dynamics such as surface area, volume, and cell density across organs, and to dissect the interplay between morphogenesis factors and cell migration. • A “Google Earth”-like browser, Spateo-viewer ( and for interactive and intuitive exploration of 3D spatial data. • Additional features, such as RNA signal-based single-cell segmentation. We are also honored that Nature “News and Views” has highlighted this work as well: This is really an amazing outcome after two years' heroic revision process that rewrite the entire paper using a new data ( for whole mouse embryos.

evo-devo

58,343 Aufrufe • vor 1 Jahr

🚀 The Segment Anything Model (SAM) has been upgraded to SAM2, featuring an efficient image encoder for segmenting images and videos. But does SAM2 outperform SAM1 in medical image and video segmentation? We're thrilled to present our paper "Segment Anything in Medical Images and Videos: Benchmark and Deployment"! We comprehensively benchmark SAM2 across 11 medical image modalities and videos. 📄 Paper: 💻 Code: **Highlights:** 1. SAM2 doesn’t always outperform SAM1 in 2D medical images, but excels in video segmentation, making it more accurate and efficient for 3D images, such as CT and MR scans. 2. MedSAM still outperforms SAM2 on most 2D modalities, but SAM2 surpasses MedSAM for 3D image segmentation in a slice-by-slice approach. 3. Segmentation performance varies with model size; sometimes the smallest model outperforms larger ones. 4. Fine-tuning SAM2 significantly boosts its performance for medical image segmentation. While SAM2 may struggle with challenging objects that have unclear boundaries or low contrast, it excels in generating good initial segmentation masks for common medical images and videos. However, the official interface doesn’t support medical data formats and has limitations on video length. To address this, we've developed a 3D Slicer Plugin and Gradio API for efficient 3D medical image and video segmentation. We invite you to try them out and provide feedback! 🔧 Deployment: - 3D Slicer Plugin: - Gradio API: (Note: Due to GPU limitations, the online API is available for only 12 hours and may be slow. We highly recommend deploying the Gradio API with your own computing resources: A big shoutout to Jun Ma (JunMa) who recently joined our UHN AI hub (UHN AI Hub) as Machine Learning Lead, and kudos to all co-authors: Sumin Kim, Feifei Li, Mohammed Baharoon (Mohammed Baharoon), Reza Asakereh, and Hongwei Lyu! This is true teamwork! Looking forward to collaborating with the community to advance 3D medical image and video segmentation foundation models! University Health Network U of T Department of Computer Science Department of Laboratory Medicine & Pathobiology Temerty Centre for AI in Medicine (T-CAIREM) Vector Institute #MedTech #AIinHealthcare #DeepLearning #MedicalImaging #SAM2 #MedSAM #AIResearch

🚀 The Segment Anything Model (SAM) has been upgraded to SAM2, featuring an efficient image encoder for segmenting images and videos. But does SAM2 outperform SAM1 in medical image and video segmentation? We're thrilled to present our paper "Segment Anything in Medical Images and Videos: Benchmark and Deployment"! We comprehensively benchmark SAM2 across 11 medical image modalities and videos. 📄 Paper: 💻 Code: Highlights: 1. SAM2 doesn’t always outperform SAM1 in 2D medical images, but excels in video segmentation, making it more accurate and efficient for 3D images, such as CT and MR scans. 2. MedSAM still outperforms SAM2 on most 2D modalities, but SAM2 surpasses MedSAM for 3D image segmentation in a slice-by-slice approach. 3. Segmentation performance varies with model size; sometimes the smallest model outperforms larger ones. 4. Fine-tuning SAM2 significantly boosts its performance for medical image segmentation. While SAM2 may struggle with challenging objects that have unclear boundaries or low contrast, it excels in generating good initial segmentation masks for common medical images and videos. However, the official interface doesn’t support medical data formats and has limitations on video length. To address this, we've developed a 3D Slicer Plugin and Gradio API for efficient 3D medical image and video segmentation. We invite you to try them out and provide feedback! 🔧 Deployment: - 3D Slicer Plugin: - Gradio API: (Note: Due to GPU limitations, the online API is available for only 12 hours and may be slow. We highly recommend deploying the Gradio API with your own computing resources: A big shoutout to Jun Ma (JunMa) who recently joined our UHN AI hub (UHN AI Hub) as Machine Learning Lead, and kudos to all co-authors: Sumin Kim, Feifei Li, Mohammed Baharoon (Mohammed Baharoon), Reza Asakereh, and Hongwei Lyu! This is true teamwork! Looking forward to collaborating with the community to advance 3D medical image and video segmentation foundation models! University Health Network U of T Department of Computer Science Department of Laboratory Medicine & Pathobiology Temerty Centre for AI in Medicine (T-CAIREM) Vector Institute #MedTech #AIinHealthcare #DeepLearning #MedicalImaging #SAM2 #MedSAM #AIResearch

Bo Wang

178,481 Aufrufe • vor 1 Jahr

This work is not about a new technique. GMT (General Motion Tracking) shows good engineering practices that you can actually train a single unified whole-body control policy for all agile motion, and it works in the real world, directly with sim2real without adaptation. This is different from many existing works that trains/adapts one individual policy for each specific motion. Instead of just showing the beautiful motions/dances, this is actually practical. Because now you can convert all types of commands into real actions in real time without tuning anything. Visit to check out how we handle the following challenges: ✅ Partial Observability. ✅ Hardware Limitations. ✅ Unbalanced Data Distribution. ✅ Model Expressiveness. Work led by Zixuan Chen Mazeyu Ji

This work is not about a new technique. GMT (General Motion Tracking) shows good engineering practices that you can actually train a single unified whole-body control policy for all agile motion, and it works in the real world, directly with sim2real without adaptation. This is different from many existing works that trains/adapts one individual policy for each specific motion. Instead of just showing the beautiful motions/dances, this is actually practical. Because now you can convert all types of commands into real actions in real time without tuning anything. Visit to check out how we handle the following challenges: ✅ Partial Observability. ✅ Hardware Limitations. ✅ Unbalanced Data Distribution. ✅ Model Expressiveness. Work led by Zixuan Chen Mazeyu Ji

Xiaolong Wang

83,261 Aufrufe • vor 1 Jahr