Craig Wright Witnesses Face Questions About Their Memories in COPA Trial

CoinDeskPolicy發佈於 2024-02-14更新於 2024-02-15

文章摘要

Lawyers for the Crypto Open Patent Alliance declared that the recollection of the witnesses today was "hazy" and "confused."

  • Craig Wright witnesses Ignatius Pang, Robert Jenkins and Shoaib Yousef took to the stand on Thursday.
  • Lawyers for the Crypto Open Patent Alliance questioned the soundness of the three witnesses' memories.
  • The trial on whether or not Wright is Satoshi Nakamoto, bitcoin's creator, is still ongoing.
10

Lawyers for the Crypto Open Patent Alliance (COPA) questioned the soundness of Craig Wright's witnesses' memories on Thursday in an ongoing trial probing whether or not he is – as Wright claims – Bitcoin creator Satoshi Nakamoto.

Three witnesses testified virtually, including Ignatius Pang, who has known Wright since 2007. His statement centered around a memory of Wright asking him to build a LEGO blockchain in 2008, which Wright likened to a Chinese recursive chain puzzle.

Another witness, Robert Jenkins, testified that he had conversations with Wright around 2009 or 2010 about concepts tied to bitcoin and blockchains.

Advertisement
Advertisement

Shoaib Yousef, the last witness to testify Thursday, said he discussed digital currency with Wright in the late 2000s.

Jonathan Moss, an attorney representing COPA, questioned whether the witnesses could accurately recall events from "16 years ago." Both Jenkins and Yousef agreed it may be difficult to recount events from that far back.

"I would suggest to you that you have confused these dates and details as you are looking back with the hindsight of what you now know," Moss told Jenkins.

Jonathan Hough, another COPA attorney, made a similar statement to Pang: "All I am saying to you is this hazy recollection is not a reliable recollection."

Wright was taken to court by COPA in the beginning of February in an effort to prove he is not Satoshi Nakamoto, the pseudonymous creator of bitcoin (BTC). Wright himself took the stand last week and faced a week’s worth of questioning, which came to a close on Wednesday.

The trial is set to continue on Friday, when three additional witnesses will take the stand.

Edited by Parikshit Mishra and Nikhilesh De.

你可能也喜歡

文生图Scaling新变量,被字节Seed团队发现了

字节Seed团队研究发现,文生图模型的性能提升不仅依赖于模型、数据和算力的扩展,更关键的是训练图像所配文本描述(Caption)中携带的、可被模型有效利用的视觉信息量。传统上增加描述长度并不能持续提升模型表现,因为新增文本常为解释或改写,并未增加明确的视觉监督信息。 团队提出两个衡量Caption信息量的指标:Grounded Perplexity Gain(GPG,衡量图像对文本预测能力的提升)和Effective Detailness(ED,衡量文本对图像属性的准确覆盖)。实验表明,在固定训练设置下,扩散模型的最终训练损失与Caption的GPG和ED值呈现高度相关的规律性,而非与文本长度相关。 基于此,团队设计了结构化提示(Structured Prompt, SP),以JSON格式明确组织图像的全局信息、元素细节及其相互关系,减少了自然语言描述中的歧义,提升了文本条件向模型传递视觉信息的能力(即Diffusability)。同时,他们通过训练大型语言模型(LLM)作为Prompter,使其能将用户简短指令转化为高质量的结构化提示(即提升Promptability),并辅以迭代修正循环来优化生成结果。 最终,在相同骨干模型(Qwen-Image)和训练资源下,采用结构化提示接口的系统在复杂组合、空间关系和世界知识等生成任务上显著优于仅使用自然语言描述的训练方式。这项工作表明,提升文本条件的信息含量和组织明确性,是推动文生图模型能力前进的一个重要新维度。

marsbit17 分鐘前

文生图Scaling新变量,被字节Seed团队发现了

marsbit17 分鐘前

加密市场降温,Neutrl 如何将代币化股票变成新「收益矿场」?

加密市场降温时,稳定币需求增加,但合成美元(如Ethena等)的收益往往因依赖加密投机活动(资金费率、基差等)而收窄。Neutrl协议试图解决这一结构性错配,将其市场中性(delta-neutral)策略延伸至代币化股票和上市前股份领域。 核心逻辑在于:股票锁仓(如IPO前股份、员工解锁等)的供给节奏独立于加密市场周期,且存在折价交易机会。随着股票永续合约市场快速发展(日成交量从8400万美元飙升至55亿美元),为这些锁仓股份提供了流动性对冲工具,使得折价套利成为可能。 一笔典型交易示例:以约15%的折价买入锁仓股票(锁定期约6个月),同时做空等额的股票永续合约对冲方向性风险。持有期间,空头仓位可获得资金费率收益(观测期内年化约10.9%)。六个月持有期整体收益约20.5个百分点(15%来自入场折价,5.5%来自资金费率),且不依赖股价涨跌。 当前,符合条件的股票锁仓市场规模巨大(估计达482亿至1.74万亿美元),远大于加密解锁市场,且股票永续市场尚处早期,套利机会丰厚。未来,这一套利框架可扩展至其他能生长出流动性做空工具的市场(如商品、利率等),随着资产代币化进程持续获得收益。Neutrl的策略旨在在加密市场平静期,为稳定币持有者提供独立于加密周期的收益来源。

marsbit41 分鐘前

加密市场降温,Neutrl 如何将代币化股票变成新「收益矿场」?

marsbit41 分鐘前

交易

現貨
活动图片