7*24

Hot topic

Market

Activities

On May 12, the Depository Trust & Clearing Corporation (DTCC) in the United States announced that it will utilize Chainlink infrastructure in its blockchain-based collateral management platform, further extending their previous collaboration to one of Wall Street's core risk management functions. DTCC stated that its Collateral AppChain will adopt Chainlink's Runtime Environment (CRE) and data standards to support functionalities such as pricing, valuation, margin management, collateral optimization, and settlement. This AppChain is a Besu-based blockchain platform designed to facilitate asset tokenization and enable real-time collateral management around the clock (24/7).

[DTCC Expands Collateral System with Chainlink Integration]On May 12, the Depository Trust & Clearing Corporation (DTCC) in the United States announced that it will utilize Chainlink infrastructure in its blockchain-based collateral management platform, further extending their previous collaboration to one of Wall Street's core risk management functions. DTCC stated that its Collateral AppChain will adopt Chainlink's Runtime Environment (CRE) and data standards to support functionalities such as pricing, valuation, margin management, collateral optimization, and settlement. This AppChain is a Besu-based blockchain platform designed to facilitate asset tokenization and enable real-time collateral management around the clock (24/7).

DTCC Expands Collateral System with Chainlink Integration

On-chain detective ZachXBT has revealed that American hacker Dritan Kapllani Jr. is suspected of participating in a social engineering theft targeting crypto users, amounting to approximately $19 million. ZachXBT stated that Dritan has long flaunted luxury cars, high-end watches, private jets, and nightlife on social media. During a 'Band 4 Band (B4B)' voice chat on Discord on April 23, 2026, he publicly showcased an Exodus wallet containing $3.68 million in assets to prove he was wealthier than another hacker. ZachXBT tracked the address and found it linked to a social engineering theft on March 14, 2026, involving 185 BTC (approximately $13 million). The following day, Dritan's Exodus wallet received about $5.3 million from that theft. As of the B4B call six weeks later, approximately $1.6 million had been spent or laundered. ZachXBT noted that Dritan has been active in the hacker community 'The Com' and had previously not faced formal charges due to his minor status, but he is now over 18 years old.

[ZachXBT Exposes American Hacker Dritan Kapllani Jr. for Alleged Involvement in $19 Million Crypto Social Engineering Theft]On-chain detective ZachXBT has revealed that American hacker Dritan Kapllani Jr. is suspected of participating in a social engineering theft targeting crypto users, amounting to approximately $19 million. ZachXBT stated that Dritan has long flaunted luxury cars, high-end watches, private jets, and nightlife on social media. During a 'Band 4 Band (B4B)' voice chat on Discord on April 23, 2026, he publicly showcased an Exodus wallet containing $3.68 million in assets to prove he was wealthier than another hacker. ZachXBT tracked the address and found it linked to a social engineering theft on March 14, 2026, involving 185 BTC (approximately $13 million). The following day, Dritan's Exodus wallet received about $5.3 million from that theft. As of the B4B call six weeks later, approximately $1.6 million had been spent or laundered. ZachXBT noted that Dritan has been active in the hacker community 'The Com' and had previously not faced formal charges due to his minor status, but he is now over 18 years old.

ZachXBT Exposes American Hacker Dritan Kapllani Jr. for Alleged Involvement in $19 Million Crypto Social Engineering Theft

On May 12, ZK intelligent verifiable computing platform Brevis officially announced the launch of Pico Prism 2.0, which brings a full-stack upgrade to its Ethereum real-time block proof system, now officially live on the Ethereum mainnet (with a 60 million Gas limit). According to official disclosures, the average proof generation time per block for Pico Prism 2.0 is approximately 6.1 seconds, with 99.9% of blocks confirmed within a 12-second block interval. The system operates on a dual-machine hardware configuration consisting of 16 RTX 5090 GPUs across two machines, with a total cost of about $100,000. Under the same testing conditions, the computational efficiency has improved by approximately 5.3 times compared to the previous generation Pico Prism 1.0. This upgrade meets the real-time proof performance targets set by the Ethereum Foundation, including an average latency of under 10 seconds and local deployment hardware costs kept below $100,000. This milestone also indicates that real-time ZK proofs on the Ethereum chain are now feasible. Previously, Brevis was selected for the Ethereum Foundation's 'On-Prem Proving Initiative', which is set to launch in May 2026, aiming to test whether ZK proofs can scale as decentralized infrastructure without relying on a few cloud service providers, marking the closest rehearsal to actual integration of L1 zkEVM to date.

[Brevis Launches Pico Prism 2.0, Enhancing Ethereum Real-Time Block Proof Efficiency by 5.3 Times]On May 12, ZK intelligent verifiable computing platform Brevis officially announced the launch of Pico Prism 2.0, which brings a full-stack upgrade to its Ethereum real-time block proof system, now officially live on the Ethereum mainnet (with a 60 million Gas limit). According to official disclosures, the average proof generation time per block for Pico Prism 2.0 is approximately 6.1 seconds, with 99.9% of blocks confirmed within a 12-second block interval. The system operates on a dual-machine hardware configuration consisting of 16 RTX 5090 GPUs across two machines, with a total cost of about $100,000. Under the same testing conditions, the computational efficiency has improved by approximately 5.3 times compared to the previous generation Pico Prism 1.0. This upgrade meets the real-time proof performance targets set by the Ethereum Foundation, including an average latency of under 10 seconds and local deployment hardware costs kept below $100,000. This milestone also indicates that real-time ZK proofs on the Ethereum chain are now feasible. Previously, Brevis was selected for the Ethereum Foundation's 'On-Prem Proving Initiative', which is set to launch in May 2026, aiming to test whether ZK proofs can scale as decentralized infrastructure without relying on a few cloud service providers, marking the closest rehearsal to actual integration of L1 zkEVM to date.

Brevis Launches Pico Prism 2.0, Enhancing Ethereum Real-Time Block Proof Efficiency by 5.3 Times

According to monitoring by Dongcha Beating and reported by TechCrunch, during last year's holiday season, Amazon's Ring evaluated over 40 companies when selecting an AI voice provider due to a surge in customer support calls, ultimately choosing the startup Vapi to handle its incoming call operations. Currently, Ring has integrated 100% of its incoming calls into the Vapi platform. Sources indicate that this deployment helped Vapi secure $50 million in Series B funding led by Peak XV Partners, with a post-money valuation of approximately $500 million. Vapi CEO Jordan Dearsley stated that Ring began engaging with Vapi in the middle of the fourth quarter last year, weighing whether to expand call center capacity, rely more on traditional automated phone systems, or deploy AI intelligent agents that can respond to customers in a more natural manner. Dearsley believes that Ring chose Vapi because the platform allows Ring engineers to have fine-grained control over the behavior of AI agents in real customer interactions.

[AI Voice Startup Vapi Secures $50 Million in Series B Funding Led by Peak XV Partners]According to monitoring by Dongcha Beating and reported by TechCrunch, during last year's holiday season, Amazon's Ring evaluated over 40 companies when selecting an AI voice provider due to a surge in customer support calls, ultimately choosing the startup Vapi to handle its incoming call operations. Currently, Ring has integrated 100% of its incoming calls into the Vapi platform. Sources indicate that this deployment helped Vapi secure $50 million in Series B funding led by Peak XV Partners, with a post-money valuation of approximately $500 million. Vapi CEO Jordan Dearsley stated that Ring began engaging with Vapi in the middle of the fourth quarter last year, weighing whether to expand call center capacity, rely more on traditional automated phone systems, or deploy AI intelligent agents that can respond to customers in a more natural manner. Dearsley believes that Ring chose Vapi because the platform allows Ring engineers to have fine-grained control over the behavior of AI agents in real customer interactions.

AI Voice Startup Vapi Secures $50 Million in Series B Funding Led by Peak XV Partners

On May 12, Michael Saylor tweeted that the CLARITY Act, being advanced by the U.S. Congress, will unlock the potential of the next phase of the 'digital capital, digital credit, and digital equity' market, and will bring stronger institutional legitimacy and adoption frameworks for BTC.

[Michael Saylor: The CLARITY Act Will Drive Institutional Adoption of Digital Capital and Bitcoin]On May 12, Michael Saylor tweeted that the CLARITY Act, being advanced by the U.S. Congress, will unlock the potential of the next phase of the 'digital capital, digital credit, and digital equity' market, and will bring stronger institutional legitimacy and adoption frameworks for BTC.

Michael Saylor: The CLARITY Act Will Drive Institutional Adoption of Digital Capital and Bitcoin

On May 12, Wells Fargo raised the target price for Nvidia (NVDA.O) from $265 to $315.

[Wells Fargo Raises Nvidia Target Price from $265 to $315]On May 12, Wells Fargo raised the target price for Nvidia (NVDA.O) from $265 to $315.

Wells Fargo Raises Nvidia Target Price from $265 to $315

According to monitoring by Dongcha Beating, Dylan Zhang, a PhD student in computer science at the University of Illinois, conducted a series of memory experiments with agents, leading to an unusual conclusion: repeatedly summarizing experiences may actually degrade the model's memory. One particularly striking result came from ARC-AGI: researchers selected 19 questions that GPT-5.4 could answer correctly without memory, then provided the model with the true solutions to these questions while it simultaneously wrote 'experience summaries.' Logically, this is akin to open-book review; however, after multiple rounds of memory compression, the same model's accuracy plummeted from 100% to 54%. The original trajectory was not incorrect; the real issue was the model rewriting the correct trajectory into generalized experiences. Worse still, this memory degradation is not an isolated case. In the WebShop online shopping task, the AWM memory method scored 0.64 after ingesting 8 expert trajectories, but when the number of trajectories increased to 128, the score dropped to 0.20, exactly returning to the no-memory baseline. This means that as memory accumulates, the benefits are effectively nullified by the model itself. The problem lies not in 'too little experience,' but in 'too frequent summarization.' The experiences written by large models are not objective logs; each summary is a form of regeneration. By the end, specific premises are discarded, rules for different tasks are blended together, and details that could guide actions transform into seemingly correct yet practically useless statements like 'prioritize the most direct action' or 'use the correct tools.' An extreme example presented in the original text shows that 50 structured memories were merged into one, compressing multiple task differences into a single generalized process, resulting in the loss of 6 to 13 successful samples in the next round of evaluation. The author's suggestion is quite restrained: do not rush to have the agent write a 'mistake book' in every round. A more stable approach is to retain filtered original operation trajectories and only abstract summaries when truly necessary. In the experiments, keeping the original episodes and disabling abstract summarization led to performance that matched or exceeded tested compressed memory methods across multiple agent benchmarks. For developers, this conclusion is straightforward: showing the model what it has actually done is usually more useful than having it memorize a bunch of abstract rules.

[AI Review Reversal: GPT-5.4's Perfect Score Solutions Drop to 54%]According to monitoring by Dongcha Beating, Dylan Zhang, a PhD student in computer science at the University of Illinois, conducted a series of memory experiments with agents, leading to an unusual conclusion: repeatedly summarizing experiences may actually degrade the model's memory. One particularly striking result came from ARC-AGI: researchers selected 19 questions that GPT-5.4 could answer correctly without memory, then provided the model with the true solutions to these questions while it simultaneously wrote 'experience summaries.' Logically, this is akin to open-book review; however, after multiple rounds of memory compression, the same model's accuracy plummeted from 100% to 54%. The original trajectory was not incorrect; the real issue was the model rewriting the correct trajectory into generalized experiences. Worse still, this memory degradation is not an isolated case. In the WebShop online shopping task, the AWM memory method scored 0.64 after ingesting 8 expert trajectories, but when the number of trajectories increased to 128, the score dropped to 0.20, exactly returning to the no-memory baseline. This means that as memory accumulates, the benefits are effectively nullified by the model itself. The problem lies not in 'too little experience,' but in 'too frequent summarization.' The experiences written by large models are not objective logs; each summary is a form of regeneration. By the end, specific premises are discarded, rules for different tasks are blended together, and details that could guide actions transform into seemingly correct yet practically useless statements like 'prioritize the most direct action' or 'use the correct tools.' An extreme example presented in the original text shows that 50 structured memories were merged into one, compressing multiple task differences into a single generalized process, resulting in the loss of 6 to 13 successful samples in the next round of evaluation. The author's suggestion is quite restrained: do not rush to have the agent write a 'mistake book' in every round. A more stable approach is to retain filtered original operation trajectories and only abstract summaries when truly necessary. In the experiments, keeping the original episodes and disabling abstract summarization led to performance that matched or exceeded tested compressed memory methods across multiple agent benchmarks. For developers, this conclusion is straightforward: showing the model what it has actually done is usually more useful than having it memorize a bunch of abstract rules.

AI Review Reversal: GPT-5.4's Perfect Score Solutions Drop to 54%

Slow Mist issued a security warning stating that Aurellion was attacked, resulting in a loss of approximately 455,003 USDC (about $455,000). Analysis indicates that the root of the vulnerability may lie in the lack of effective protection in the initialize(address) function of the SafeOwnable Facet. The Diamond contract failed to update the _initialized version slot correctly because it did not go through the initialize path when setting the owner, allowing the attacker to reinitialize the contract and override owner permissions. Slow Mist reported that the attacker has seized ownership of the Diamond contract and transferred USDC from multiple authorized addresses.

[Slow Mist: Aurellion Attacked Due to Initialization Vulnerability, Approximately $455,000 USDC Stolen]Slow Mist issued a security warning stating that Aurellion was attacked, resulting in a loss of approximately 455,003 USDC (about $455,000). Analysis indicates that the root of the vulnerability may lie in the lack of effective protection in the initialize(address) function of the SafeOwnable Facet. The Diamond contract failed to update the _initialized version slot correctly because it did not go through the initialize path when setting the owner, allowing the attacker to reinitialize the contract and override owner permissions. Slow Mist reported that the attacker has seized ownership of the Diamond contract and transferred USDC from multiple authorized addresses.

Slow Mist: Aurellion Attacked Due to Initialization Vulnerability, Approximately $455,000 USDC Stolen

According to monitoring by Dongcha Beating, NVIDIA and the MIT research team have released a new post-training framework for large language models called Lightning OPD (Offline Policy Distillation). This technology enhances training efficiency by 4 times by pre-computing the log-probabilities of the teacher model offline, eliminating the need for real-time teacher services that must remain online throughout traditional distillation training. Previously, standard Online Policy Distillation (OPD) required both student and teacher models to run simultaneously on a single machine, which often led to out-of-memory (OOM) errors as models grew larger. Lightning OPD frees up all GPU resources for the student model. In tests using 8 H100 GPUs on a single node, Lightning OPD successfully distilled the Qwen3-30B-A3B-Base (a large MoE model with a total of 30 billion parameters), achieving a score of 71.0 in the AIME 2024 test; in contrast, standard OPD encountered OOM errors under the same hardware configuration. For the smaller Qwen3-8B model, this framework achieved a score of 69.9 in just 30 GPU hours. The research team pointed out a hidden prerequisite for achieving offline distillation in their paper: 'teacher consistency.' The student model must use the same teacher model during supervised fine-tuning (SFT) and subsequent distillation phases. Failing to adhere to this principle can lead to biased gradient directions, ultimately degrading the model's performance.

[MIT and NVIDIA Launch Lightning OPD: A 4x Faster Offline Distillation Framework]According to monitoring by Dongcha Beating, NVIDIA and the MIT research team have released a new post-training framework for large language models called Lightning OPD (Offline Policy Distillation). This technology enhances training efficiency by 4 times by pre-computing the log-probabilities of the teacher model offline, eliminating the need for real-time teacher services that must remain online throughout traditional distillation training. Previously, standard Online Policy Distillation (OPD) required both student and teacher models to run simultaneously on a single machine, which often led to out-of-memory (OOM) errors as models grew larger. Lightning OPD frees up all GPU resources for the student model. In tests using 8 H100 GPUs on a single node, Lightning OPD successfully distilled the Qwen3-30B-A3B-Base (a large MoE model with a total of 30 billion parameters), achieving a score of 71.0 in the AIME 2024 test; in contrast, standard OPD encountered OOM errors under the same hardware configuration. For the smaller Qwen3-8B model, this framework achieved a score of 69.9 in just 30 GPU hours. The research team pointed out a hidden prerequisite for achieving offline distillation in their paper: 'teacher consistency.' The student model must use the same teacher model during supervised fine-tuning (SFT) and subsequent distillation phases. Failing to adhere to this principle can lead to biased gradient directions, ultimately degrading the model's performance.

MIT and NVIDIA Launch Lightning OPD: A 4x Faster Offline Distillation Framework

Meson.fi, the stablecoin cross-chain protocol, has announced its launch on the Aptos mainnet. Meson becomes the first stablecoin cross-chain protocol to run on the Aptos Autumn mainnet and Aptos becomes the 17th blockchain network supported by Meson.
Aptos is a blockchain project founded by former Libra core members and uses the Move development language. meson provides a cross-chain path to mainstream stablecoins on public chains such as Aptos and ethereum.
Meson says it is actively seeking to collaborate with the Aptos eco-project to jointly promote the use of mainstream stablecoins such as USDT/USDC on Aptos and provide users with a convenient and smooth cross-chain experience across different blockchain networks.

Detail

【Meson, the stablecoin cross-chain protocol, goes live on the Aptos mainnet】Meson.fi, the stablecoin cross-chain protocol, has announced its launch on the Aptos mainnet. Meson becomes the first stablecoin cross-chain protocol to run on the Aptos Autumn mainnet and Aptos becomes the 17th blockchain network supported by Meson.
Aptos is a blockchain project founded by former Libra core members and uses the Move development language. meson provides a cross-chain path to mainstream stablecoins on public chains such as Aptos and ethereum.
Meson says it is actively seeking to collaborate with the Aptos eco-project to jointly promote the use of mainstream stablecoins such as USDT/USDC on Aptos and provide users with a convenient and smooth cross-chain experience across different blockchain networks.

Meson, the stablecoin cross-chain protocol, goes live on the Aptos mainnet

Top Ethereum Holder Makes Bold Move Says ETH Is Close to Generational Run, Here’s When

Coinbase裁员14%，熊市和AI哪个才是主因？

David Schwartz Says Ripple Stock Is His Real Bet, Not XRP

租来的轨道：这一波稳定币 FX 热钱，到底在为什么买单？

MegaETH 如何让 TVL 在 TGE 一周内达到 700m？拆解包装手法

Bulls Approach the Ceiling

The Bitcoin Vector #54

Roobet Launches Prediction Markets on May 6, The First Major Crypto Casino to Integrate the Format

Pre-ETF Era Bitcoin Whales Surface To Cash In On $80,000 Rally

Article

ACA fluctuated over 50% in 24 hours!

Meson, the stablecoin cross-chain protocol, goes live on the Aptos mainnet

Clause de non-responsabilité：Le contenu ci-dessus ne représente pas les positions de HTX.，HTX ne fournit pas de recommandations de trading.。

Articles connexes

Tous les commentaires0RécentPopulaire

Sujets d'actualité

Articles connexes