0xPackage @0xPackage

Be a creator ## nameless Joined February 2023

Tweets

39
Followers

82
Following

2K
Likes

42

CryptoSoulz @SoulzBTC

2 months ago

The BEST 7 Trading Indicators 0:00 - RSI & SMA 1:17 - Trend Trader 2:36 - Smart Divergences 3:52 - RSI & Envelope Band 4:54 - Buy & Sell Signals 5:49 - Trailing Stop Loss 6:47 - Liquidity

11 130 684 41K 819

View Details

0xPackage @0xPackage

3 months ago

【量子计算到底威胁了什么？一份通向后量子时代的 Crypto 用户指南】 a.foresightnews.pro/article/h5Deta…

0 0 0 22 0

View Details

0xPackage @0xPackage

3 months ago

@wolfyxbt @0xAA_Science 看两下的就退了没啥用

0 0 0 97 0

View Details

0xPackage @0xPackage

4 months ago

#fomogo #ai ai thinking tp/sl

0 0 0 16 0

View Details

0xPackage @0xPackage

4 months ago

#fomo Trading is simple

0 0 0 17 0

View Details

0xPackage @0xPackage

5 months ago

This SMC Pattern Gets You in Before CHoCH youtube.com/shorts/zNsmbf7… 来自 @YouTube

0 0 0 34 0

View Details

0xPackage @0xPackage

5 months ago

Why Your Entries Keep Failing You youtube.com/shorts/ChyK5XQ… 来自 @YouTube

0 0 0 28 0

View Details

0xPackage @0xPackage

6 months ago

大家看看这笔流动性交易怎么样? #ICT #OrderBook #Liquidity #BTC #原创

0 0 0 15 0

View Details

0xPackage @0xPackage

7 months ago

币安实盘盈利4500万美元博主：韭菜，快停止日内交易！ odaily.news/zh-CN/post/520…

0 0 0 21 0

View Details

5-Day AI Agents Intensive Course with Google (3/5) 谷歌这5天的 AI Agents 课程太硬核了，前三天连续发布白皮书，每一本都值得慢慢阅读，理论结合最佳实践，对重新理解和真正把 AI Agents 在企业中落地都很有帮助。再把这三本白皮书发在这里，前几天错过的朋友们，看这一篇： 1. Introduction to Agents kaggle.com/whitepaper-int… 2. Agent Tools & Interoperability with MCP kaggle.com/whitepaper-age… 3. Context Engineering: Sessions & Memory kaggle.com/whitepaper-con… 4-5 天一起期待，这周五全部更新。

6 172 493 55K 645

View Details

苍何 @canghe

8 months ago

重温了一下谷歌发布的 68 页提示词工程书籍，对提示词又再次加深了理解。这本书是完全免费的！你甚至不需要注册。它包含真实的数据、实验和最佳实践。以下是需要了解的最重要的要点： 1，prompt不是比谁更聪明，而是要有系统性。谷歌提供的这本免费书籍对于任何从事LLM的人来说都是一座金矿。 2，提示词是一门科学，不是靠运气： •谷歌的研究表明，结构化提示持续优于随意的表述。 •你的措辞可以让输出质量提升几个数量级。链式思维 > 一次性回答： •把任务拆成步骤（显式把思路写出来）会让模型更准确。 •不要只索要最终结果 → 要引导模型完成推理过程。

8 196 645 65K 764

View Details

Connor Davis @connordavis_ai

8 months ago

If you’re building AI agents right now, you’re probably doing it wrong. Most “agents” break after one task because nobody’s teaching the real framework. Here’s how to build one that actually works ↓

10 29 157 15K 170

View Details

Akshay 🚀 @akshay_pachaar

8 months ago

Here's a common misconception about RAG! Most people think RAG works like this: index a document → retrieve that same document. But indexing ≠ retrieval. What you index doesn't have to be what you feed the LLM. Once you understand this, you can build RAG systems that actually work. Here are 4 indexing strategies that separate good RAG from great RAG: 1) Chunk Indexing ↳ This is the standard approach. Split documents into chunks, embed them, store in a vector database, and retrieve the closest matches. ↳ Simple and effective, but large or noisy chunks will hurt your precision. 2) Sub-chunk Indexing ↳ Break your chunks into smaller sub-chunks for indexing, but retrieve the full chunk for context. ↳ This is powerful when a single section covers multiple concepts. You get better query matching without losing the surrounding context your LLM needs. 3) Query Indexing ↳ Instead of indexing raw text, generate hypothetical questions the chunk could answer. Index those questions instead. ↳ User queries naturally align better with questions than raw document text. This closes the semantic gap between what users ask and what you've stored. ↳ Perfect for QA systems. 4) Summary Indexing ↳ Use an LLM to summarize each chunk. Index the summary, retrieve the full chunk. ↳ This shines with dense, structured data like CSVs and tables where raw text embeddings fall flat. The bottom line: You don't need to retrieve exactly what you indexed. Match your indexing strategy to your data, and your RAG system will perform significantly better. What indexing strategies have worked best for you?

31 150 772 55K 865

View Details

Mia米粒儿🩵 @mia_okx

8 months ago

x.com/i/article/1987…

57 158 823 576K 821

View Details

0xPackage @0xPackage

8 months ago

@kkk__eth @ibitpen @HashKeyGroup @HSKChain @HashKey_Global @HashKeyExchange @HashKey_Capital 纯黑子，回他就没格局了

0 0 0 446 0

View Details

ginobefun @hongming731

8 months ago

我非常推荐阅读上海交通大学和 GAIR 实验室这篇关于《上下文工程 2.0》的论文，它系统性地回答了上下文工程的本质、基础构件和未来发展。读完论文中关于其本质的论述，我很有感触。它指出了一个核心困境：人与机器的交流之所以困难，是因为我们之间存在一道深深的认知鸿沟。我们人类的表达是高熵的。我们说话时充满模糊、暗示，并且严重依赖过往的共识。比如我们对同事说“搞定那个报告”，对方就能理解。而机器，至少在目前，是低熵的。它无法处理这种模糊性，只能理解精确、毫不含糊的指令。「上下文工程」这个概念，正是为了解决这个根本矛盾。论文将这个过程比作「熵减」，这个比喻很贴切。另外，我们也可以这么理解：当前的 AI，比如我们熟悉的 LLM，其技术架构的本质是一个无状态的函数。这意味着，你给它一个输入，它就返回一个输出。它没有真正的记忆，不记得你是谁，也不懂你身处的环境。这和我们人类截然不同。我们是有状态的存在。我们的每一个反应，都建立在长期的记忆和对当下的感知之上。因此，上下文工程的真正本质，是为这个「无状态」的函数，搭建一套庞大而精巧的脚手架。我们用 RAG 和外部数据库来模拟它的长期记忆；我们用多模态输入来模拟它的感官；我们再用复杂的管理系统帮它过滤信息，模拟它的注意力。我们正努力用所有这些外部工具，把一个天生没有记忆、没有感官的「函数」，包装成一个看起来有记忆、能理解环境的智能体。这套脚手架是现阶段的必然产物，它的核心价值是让 AI 从能用变得好用。但它的终极使命，其实是等待一个真正具备原生记忆和感知的，也就是有状态的新架构出现。到那时，这套脚手架也就可以功成身退了。