7*24

Hot topic

Market

Activities

On May 26, ECB Executive Board Member Isabel Schnabel stated that a rate hike in June is inevitable; we can no longer ignore the surge in inflation. The latest data indicates that there are upward risks to inflation and downward risks to economic growth. The inflation shock is gradually spreading to other parts of the consumer basket, with increasingly evident signs.

[ECB Executive Schnabel: Rate Hike in June Inevitable]On May 26, ECB Executive Board Member Isabel Schnabel stated that a rate hike in June is inevitable; we can no longer ignore the surge in inflation. The latest data indicates that there are upward risks to inflation and downward risks to economic growth. The inflation shock is gradually spreading to other parts of the consumer basket, with increasingly evident signs.

ECB Executive Schnabel: Rate Hike in June Inevitable

According to Onchain Lens monitoring, Whale Evaded (@ICanPlug) opened a short position of 525.34 BTC with 40x leverage one hour ago, valued at approximately $40.26 million. Additionally, two days ago, they opened a long position of 53,500 ZEC with 5x leverage, valued at around $33.3 million, which is currently facing a floating loss of over $1.48 million.

[Whale Opens $40.26 Million BTC Short Position While Holding $33.3 Million ZEC Long Position]According to Onchain Lens monitoring, Whale Evaded (@ICanPlug) opened a short position of 525.34 BTC with 40x leverage one hour ago, valued at approximately $40.26 million. Additionally, two days ago, they opened a long position of 53,500 ZEC with 5x leverage, valued at around $33.3 million, which is currently facing a floating loss of over $1.48 million.

Whale Opens $40.26 Million BTC Short Position While Holding $33.3 Million ZEC Long Position

On May 26, DeFi yield protocol Pendle Finance announced that it will concentrate co-incentives on the Limit Orders (LO) mechanism to enhance the platform's liquidity depth and trading execution efficiency. Pendle stated that since the full LO incentive system was launched about two months ago, the proportion of limit orders in the platform's total swap trading volume has increased from 44% to 71%, with monthly LO trading volume nearly doubling, making it the core driving force of Pendle's trading activity. Data shows that Pendle currently allocates about 6,500 PENDLE incentives weekly to limit orders, supporting approximately $400 million in nominal order book depth. The platform claims that on an annualized basis, every $1 in incentives can generate about $800 in liquidity, achieving a capital efficiency of approximately 800 times. Pendle also announced new co-incentive rules: if project parties provide incentives in PENDLE, they will receive an additional $0.22 in PENDLE for every $1 invested; if incentives are provided in other tokens, they will receive $0.15 in PENDLE for every $1. If overall demand exceeds the weekly incentive cap of 9,000 PENDLE, rewards will be distributed proportionally. Pendle stated that LOs have proven to be the most effective tool for improving market quality and liquidity depth, and future co-incentive resources will prioritize the limit order ecosystem.

[Pendle Shifts Co-Incentives to Limit Orders, Reports 71% Trading Volume Contribution from LOs]On May 26, DeFi yield protocol Pendle Finance announced that it will concentrate co-incentives on the Limit Orders (LO) mechanism to enhance the platform's liquidity depth and trading execution efficiency. Pendle stated that since the full LO incentive system was launched about two months ago, the proportion of limit orders in the platform's total swap trading volume has increased from 44% to 71%, with monthly LO trading volume nearly doubling, making it the core driving force of Pendle's trading activity. Data shows that Pendle currently allocates about 6,500 PENDLE incentives weekly to limit orders, supporting approximately $400 million in nominal order book depth. The platform claims that on an annualized basis, every $1 in incentives can generate about $800 in liquidity, achieving a capital efficiency of approximately 800 times. Pendle also announced new co-incentive rules: if project parties provide incentives in PENDLE, they will receive an additional $0.22 in PENDLE for every $1 invested; if incentives are provided in other tokens, they will receive $0.15 in PENDLE for every $1. If overall demand exceeds the weekly incentive cap of 9,000 PENDLE, rewards will be distributed proportionally. Pendle stated that LOs have proven to be the most effective tool for improving market quality and liquidity depth, and future co-incentive resources will prioritize the limit order ecosystem.

Pendle Shifts Co-Incentives to Limit Orders, Reports 71% Trading Volume Contribution from LOs

On May 26, the South Korean government announced that it will monitor the risks associated with overseas private debt investments totaling approximately $37 billion made by Korean investors, following a series of risk events in the global private debt sector. In a joint statement released on Tuesday, South Korea's financial regulatory authorities and the Ministry of Finance stated that the government will 'continuously pay attention to the risk exposure of Korean investors while considering market conditions' and will 'maintain a close cooperation system among relevant departments to respond to changes in the situation.' According to the statement, as of the end of February, the overall risk exposure of the South Korean financial sector to overseas private debt investments was 30.5 trillion won (approximately $20.2 billion), with pension fund-related investments amounting to about 25.4 trillion won (approximately $16.8 billion). Reports have indicated that the Financial Supervisory Service of South Korea began reviewing the overseas private credit risk exposure across all industries under its regulatory scope last month. The South Korean government stated: 'Considering the scale of overseas private debt investments relative to total assets, the overall investment risk in the financial sector remains at a controllable level.'

[South Korea to Monitor $37 Billion Overseas Private Debt Investment Risks]On May 26, the South Korean government announced that it will monitor the risks associated with overseas private debt investments totaling approximately $37 billion made by Korean investors, following a series of risk events in the global private debt sector. In a joint statement released on Tuesday, South Korea's financial regulatory authorities and the Ministry of Finance stated that the government will 'continuously pay attention to the risk exposure of Korean investors while considering market conditions' and will 'maintain a close cooperation system among relevant departments to respond to changes in the situation.' According to the statement, as of the end of February, the overall risk exposure of the South Korean financial sector to overseas private debt investments was 30.5 trillion won (approximately $20.2 billion), with pension fund-related investments amounting to about 25.4 trillion won (approximately $16.8 billion). Reports have indicated that the Financial Supervisory Service of South Korea began reviewing the overseas private credit risk exposure across all industries under its regulatory scope last month. The South Korean government stated: 'Considering the scale of overseas private debt investments relative to total assets, the overall investment risk in the financial sector remains at a controllable level.'

South Korea to Monitor $37 Billion Overseas Private Debt Investment Risks

According to monitoring by Beating, xAI has announced that its terminal programming AI, Grok Build (Beta), is now officially available to all SuperGrok and X Premium+ subscribers. Developers can now use the official command line interface to debug, modify, and refactor code in natural language on their local terminals. Previously, testing access was limited to flagship SuperGrok Heavy subscribers paying $300 per month. Designed as a terminal tool specifically for professional software engineering, Grok Build introduces a Plan Mode workflow that analyzes the overall project and generates a detailed implementation plan for user review before executing any code changes. The tool natively supports Model Context Protocol (MCP) and multi-agent parallel collaboration, allowing it to run in a headless mode without a graphical user interface. Developers can also directly invoke Imagine's image and video generation capabilities in the CLI interface to build automated tasks or orchestrate workflows.

[xAI Officially Launches Grok Build for SuperGrok and X Premium+ Subscribers]According to monitoring by Beating, xAI has announced that its terminal programming AI, Grok Build (Beta), is now officially available to all SuperGrok and X Premium+ subscribers. Developers can now use the official command line interface to debug, modify, and refactor code in natural language on their local terminals. Previously, testing access was limited to flagship SuperGrok Heavy subscribers paying $300 per month. Designed as a terminal tool specifically for professional software engineering, Grok Build introduces a Plan Mode workflow that analyzes the overall project and generates a detailed implementation plan for user review before executing any code changes. The tool natively supports Model Context Protocol (MCP) and multi-agent parallel collaboration, allowing it to run in a headless mode without a graphical user interface. Developers can also directly invoke Imagine's image and video generation capabilities in the CLI interface to build automated tasks or orchestrate workflows.

xAI Officially Launches Grok Build for SuperGrok and X Premium+ Subscribers

On May 26, Yu Xian, the founder of Slow Mist, posted on social media to interpret the Squid security incident. He stated that samples showed the relevant Safe wallets were all single-signature, with different owners, but the issue was not with the private keys. The problem lies in the module used by these Safe addresses (SquidRouterModule), which has vulnerabilities that allow attackers to forge messages and easily bypass related verifications, initiating subsequent exchange operations to transfer funds from the target Safe wallet. Additionally, Yu Xian disclosed information about the addresses where the attackers' profits were deposited.

[Slow Mist's Yu Xian: Squid Security Incident Not Due to Private Key, Vulnerability in Safe Wallet's Module]On May 26, Yu Xian, the founder of Slow Mist, posted on social media to interpret the Squid security incident. He stated that samples showed the relevant Safe wallets were all single-signature, with different owners, but the issue was not with the private keys. The problem lies in the module used by these Safe addresses (SquidRouterModule), which has vulnerabilities that allow attackers to forge messages and easily bypass related verifications, initiating subsequent exchange operations to transfer funds from the target Safe wallet. Additionally, Yu Xian disclosed information about the addresses where the attackers' profits were deposited.

Slow Mist's Yu Xian: Squid Security Incident Not Due to Private Key, Vulnerability in Safe Wallet's Module

On May 26, the DeFi digital lending protocol Saturn Foundation announced via Twitter that it has blacklisted the wallets associated with the attackers of the SquidRouterModule and frozen the stolen funds. The Saturn contracts or infrastructure were not affected. Previously, according to monitoring by Blockaid, the SquidRouterModule on Ethereum and Base was exploited, resulting in the clearing of 86 Gnosis Safes within approximately 2 hours, with losses amounting to around 3 million dollars.

[Saturn Foundation Blacklists Attackers' Wallets and Freezes Stolen Funds]On May 26, the DeFi digital lending protocol Saturn Foundation announced via Twitter that it has blacklisted the wallets associated with the attackers of the SquidRouterModule and frozen the stolen funds. The Saturn contracts or infrastructure were not affected. Previously, according to monitoring by Blockaid, the SquidRouterModule on Ethereum and Base was exploited, resulting in the clearing of 86 Gnosis Safes within approximately 2 hours, with losses amounting to around 3 million dollars.

Saturn Foundation Blacklists Attackers' Wallets and Freezes Stolen Funds

According to monitoring by Dongcha Beating, Xiaomi Auto has officially released the new framework for the Xiaomi EV World Model, which integrates a driving assistance world model. This marks the first time that deep coupling of 3D reconstruction and video generation modules has been achieved internally. In autonomous driving simulations, traditional technologies often separate reconstruction from generation. The reconstruction module can restore scenes but cannot predict changes, while the generation module can predict the future but tends to suffer from distortion and drift over long sequences. The team proposed the JointWM architecture, which uses a 3D geometric structure as a physical skeleton to anchor the scene, and then uses the generation module to complete visual details and predict unobserved areas, setting multiple best performance records in mainstream benchmarks such as Waymo and nuScenes. Specifically, the reconstruction module WorldRec abandons the traditional pixel-by-pixel paradigm, opting for sparse 3D query points to represent scenes, incrementally fusing into a cross-view 4D Gaussian spatial skeleton, achieving rapid reconstruction of a 10-second video in just 10 seconds. Based on the geometric priors provided by the reconstruction module, the generation module WorldGen is constrained by the physical boundaries of the skeleton and is only responsible for generating reasonable lighting and textures. For content beyond the boundaries, such as future frames and blind spots, the generation module performs physical predictions through a two-stage temporal training and distribution matching distillation mechanism. The entire architecture achieves generation speeds of 0.19 seconds for a single view and 0.46 seconds for three views on an H20 GPU, and supports video generation of up to 1 minute long. This solution achieved a PSNR score of 28.48 in the Waymo reconstruction accuracy test and maintained a lead in zero-shot generalization in nuScenes. In terms of generation efficiency, the solution is 5.6 times faster than the autoregressive baseline Epona, with spatiotemporal coherence ranking among the best in similar algorithms. Currently, the research results have been implemented in three major scenarios for Xiaomi Auto, including delivering over 100,000 segments of high-quality synthetic data for perception model training, constructing a highly realistic closed-loop simulation environment to replicate long-tail traffic conditions, and launching a driving assistance classroom to guide user operations through generative videos.

[Xiaomi Unveils Integrated World Model Framework, Sets New Benchmark Performance Records]According to monitoring by Dongcha Beating, Xiaomi Auto has officially released the new framework for the Xiaomi EV World Model, which integrates a driving assistance world model. This marks the first time that deep coupling of 3D reconstruction and video generation modules has been achieved internally. In autonomous driving simulations, traditional technologies often separate reconstruction from generation. The reconstruction module can restore scenes but cannot predict changes, while the generation module can predict the future but tends to suffer from distortion and drift over long sequences. The team proposed the JointWM architecture, which uses a 3D geometric structure as a physical skeleton to anchor the scene, and then uses the generation module to complete visual details and predict unobserved areas, setting multiple best performance records in mainstream benchmarks such as Waymo and nuScenes. Specifically, the reconstruction module WorldRec abandons the traditional pixel-by-pixel paradigm, opting for sparse 3D query points to represent scenes, incrementally fusing into a cross-view 4D Gaussian spatial skeleton, achieving rapid reconstruction of a 10-second video in just 10 seconds. Based on the geometric priors provided by the reconstruction module, the generation module WorldGen is constrained by the physical boundaries of the skeleton and is only responsible for generating reasonable lighting and textures. For content beyond the boundaries, such as future frames and blind spots, the generation module performs physical predictions through a two-stage temporal training and distribution matching distillation mechanism. The entire architecture achieves generation speeds of 0.19 seconds for a single view and 0.46 seconds for three views on an H20 GPU, and supports video generation of up to 1 minute long. This solution achieved a PSNR score of 28.48 in the Waymo reconstruction accuracy test and maintained a lead in zero-shot generalization in nuScenes. In terms of generation efficiency, the solution is 5.6 times faster than the autoregressive baseline Epona, with spatiotemporal coherence ranking among the best in similar algorithms. Currently, the research results have been implemented in three major scenarios for Xiaomi Auto, including delivering over 100,000 segments of high-quality synthetic data for perception model training, constructing a highly realistic closed-loop simulation environment to replicate long-tail traffic conditions, and launching a driving assistance classroom to guide user operations through generative videos.

Xiaomi Unveils Integrated World Model Framework, Sets New Benchmark Performance Records

On May 26, the "15th Five-Year Plan Outline for National Economic and Social Development of Shenzhen" was released. It proposes to deeply implement the "AI+" initiative, promoting the comprehensive integration of artificial intelligence in industries, technology, people's livelihoods, and urban governance through greater scene openness. The plan aims to drive the integration of "AI+ advanced manufacturing," constructing a new generation of manufacturing system technology base that is highly autonomous, deeply flexible, and globally efficient, creating a future manufacturing model where "design equals manufacturing, manufacturing equals intelligence, and manufacturing equals service."

[Shenzhen's 15th Five-Year Plan: Promoting AI Empowerment Across Industries]On May 26, the "15th Five-Year Plan Outline for National Economic and Social Development of Shenzhen" was released. It proposes to deeply implement the "AI+" initiative, promoting the comprehensive integration of artificial intelligence in industries, technology, people's livelihoods, and urban governance through greater scene openness. The plan aims to drive the integration of "AI+ advanced manufacturing," constructing a new generation of manufacturing system technology base that is highly autonomous, deeply flexible, and globally efficient, creating a future manufacturing model where "design equals manufacturing, manufacturing equals intelligence, and manufacturing equals service."

Shenzhen's 15th Five-Year Plan: Promoting AI Empowerment Across Industries

Web3 bounty platform Layer3 released product updates, launched a new task platform, and officially ended beta testing.

Detail

【Web3 bounty platform Layer3 launched a new task platform】Web3 bounty platform Layer3 released product updates, launched a new task platform, and officially ended beta testing.

Web3 bounty platform Layer3 launched a new task platform

XRP Ledger Hard Fork In 8 Days? Upgrade Deadline Sparks Network Split Debate

As Presale Funding Nears $7.2 Million, Ozak AI’s Early Pricing Window Begins to Narrow

Market Expert Updates XRP Roadmap To $300 With New Data

$50 Million Funding Ignites Airdrop Anticipation, Variational Becomes New Focus of Perp DEX

Musk Posted a Recruitment Ad for SpaceX, and After Reading the Comments Section, I Understood

Cloudflare CEO: How I Decided Which Employees to Replace with AI?

Secures Over $60 Million in Funding from Dragonfly, Sequoia, and Others. An In-Depth Look at the On-Chain Derivatives Protocol Variational | CryptoSeed

XRP Ledger Hits No. 4 In RWA Rankings In Less Than A Year

General Tensor and Talisman Joining Forces To Build Financial Rails for Decentralized AI

Warsh's First Day in Office, Markets Deliver a 'Wake-up Call': Rate Hike Expected This Year

Article

Shanghai and Shenzhen Markets Total Trading Volume Reaches 32,057.3 Billion Yuan, SMIC Leads in Trading Volume

Web3 bounty platform Layer3 launched a new task platform

Isenção de responsabilidade：O conteúdo acima não representa as posições da HTX.，A HTX não faz quaisquer recomendações de negociação.。

Artigos relacionados

Todos os comentários0Mais recentePopular

Tópicos em destaque

Artigos relacionados