正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Claimify is a new method for extracting simple, verifiable claims from LLM outputs that preserves critical context and outperforms past approaches—making it easier to fact-check information.

Microsoft Research

557,356 subscribers

19,720 次观看 • 1 年前 •via X (Twitter)

科学技术新闻政治教育

Anya Rossi• Live Now

Private livecam show

4 条评论

kfant 的头像

kfant1 年前

@LLMSherpa

kfant 的头像

kfant1 年前

where is the code / model or at least a demo,ms metropolitankyareansky?

dyfrai 的头像

dyfrai1 年前

Well u can build own Clamify by smart system prompts and 2-3 easy-least agents. Invention ended. Kek.

brusscan 的头像

brusscan1 年前

Woohoo!!! 🎉 🥳👏🎊🙌

相关视频

CNN cuts away from Trump’s remarks to fact check his lies: “He is making a number of false claims...”

CNN cuts away from Trump’s remarks to fact check his lies: “He is making a number of false claims...”

Kamala HQ

606,587 次观看 • 1 年前

Simple method for removing a tree stump! Is it really that simple? 🤔

Simple method for removing a tree stump! Is it really that simple? 🤔

HOW THINGS WORK

471,894 次观看 • 1 年前

We just shipped a new way to save and re-use system instructions in Google AI Studio, making it easier to test, reproduce outputs, and experiment with Gemini : ) This has been a top ask for a while, glad we landed it!

We just shipped a new way to save and re-use system instructions in Google AI Studio, making it easier to test, reproduce outputs, and experiment with Gemini : ) This has been a top ask for a while, glad we landed it!

Logan Kilpatrick

92,105 次观看 • 8 个月前

seems like sungjae found a new friend😂 actually i already felt from ep. 0 that they seemed to click. and the fact that the ceo is younger and the age gap is not that big makes it easier for them to get close.

seems like sungjae found a new friend😂 actually i already felt from ep. 0 that they seemed to click. and the fact that the ceo is younger and the age gap is not that big makes it easier for them to get close.

fy🍀🎶

14,857 次观看 • 2 个月前

Has a gender-critical book been 'banned' from an exhibition by the National Library of Scotland? No. We fact check the latest claims from gender-critical campaigners 👇

Has a gender-critical book been 'banned' from an exhibition by the National Library of Scotland? No. We fact check the latest claims from gender-critical campaigners 👇

The National

525,054 次观看 • 10 个月前

The unscientific beliefs and stigma associated with mental illnesses are a matter of the past, making it easier for those suffering from various ailments to seek help.

The unscientific beliefs and stigma associated with mental illnesses are a matter of the past, making it easier for those suffering from various ailments to seek help.

President of India

25,025 次观看 • 1 年前

AI trading has a trust problem. Hallucinated signals. Biased predictions. Unverified outputs. Gigabrain solves it with Mira's verification APIs that fact-check all outputs in real time, making it the first AI trading app with institutional-level trust. See it in action👇

AI trading has a trust problem. Hallucinated signals. Biased predictions. Unverified outputs. Gigabrain solves it with Mira's verification APIs that fact-check all outputs in real time, making it the first AI trading app with institutional-level trust. See it in action👇

Mira

58,526 次观看 • 1 年前

How is Somnia making it easier for games to share data, build faster, and unlock new ways to earn?

How is Somnia making it easier for games to share data, build faster, and unlock new ways to earn?

Somnia

28,075 次观看 • 1 年前

AIP Evolve — our new product for making agents more efficient and cost effective. See how Chad and Colton used it to autonomously swap models, tune prompts, validate outputs, and find structured ontology data that eliminated 2 LLM calls; cutting compute costs while improving accuracy and reliability in production.

AIP Evolve — our new product for making agents more efficient and cost effective. See how Chad and Colton used it to autonomously swap models, tune prompts, validate outputs, and find structured ontology data that eliminated 2 LLM calls; cutting compute costs while improving accuracy and reliability in production.

Palantir

116,488 次观看 • 28 天前

RouteLLM - Route To The Best LLM Based On Your Prompt Our RouteLLM got an upgrade this week.., it got smarter at picking the right LLM - o1 for complex queries - gpt4o for quick answers - sonnet for coding - deepseek for simple code - gemini for long context - llama and mini for simple question It optimizes for performance, speed and cost!

RouteLLM - Route To The Best LLM Based On Your Prompt Our RouteLLM got an upgrade this week.., it got smarter at picking the right LLM - o1 for complex queries - gpt4o for quick answers - sonnet for coding - deepseek for simple code - gemini for long context - llama and mini for simple question It optimizes for performance, speed and cost!

Bindu Reddy

17,860 次观看 • 1 年前

Hedera = AI’s trust layer. Why? 🧠 LLM data provenance 🔐 Verifiable models 📦 Auditable outputs ⚖️ Enforceable ethics Mance Harmon said it. Hedera does it. Fast. Final. Trustless.

Hedera = AI’s trust layer. Why? 🧠 LLM data provenance 🔐 Verifiable models 📦 Auditable outputs ⚖️ Enforceable ethics Mance Harmon said it. Hedera does it. Fast. Final. Trustless.

Hedera

31,956 次观看 • 1 年前

New short course: Safe and Reliable AI via Guardrails! Learn to create production-ready, reliable LLM applications with guardrails in this new course, built in collaboration with Guardrails AI and taught by its CEO and co-founder, shreya rajpal. I see many companies worry about the reliability of LLM-based systems -- will they hallucinate a catastrophically bad response? -- which slows down investing in building them and transitioning prototypes to deployment. That LLMs generate probabilistic outputs has made them particularly hard to deploy in highly regulated industries or in safety-critical environments. Fortunately, there are good guardrail tools that give a significant new layer of control and reliability/safety. They act as a protective framework that can prevent your application from revealing incorrect, irrelevant, or confidential information, and they are an important part of what it takes to actually get prototypes to deployment. This course will walk you through common failure modes of LLM-powered applications (like hallucinations or revealing personally identifiable information). It will show you how to build guardrails from scratch to mitigate them. You’ll also learn how to access a variety of pre-built guardrails on the GuardrailsAI hub that are ready to integrate into your projects. You'll implement these guardrails in the context of a RAG-powered customer service chatbot for a small pizzeria. Specifically, you'll: - Explore common failure modes like hallucinations, going off-topic, revealing sensitive information, or responses that can harm the pizzeria's reputation. - Learn to mitigate these failure modes with input and output guards that check inputs and/or outputs - Create a guardrail to prevent the chatbot from discussing sensitive topics, such as a confidential project at the pizza shop - Detect hallucinations by ensuring responses are grounded in trusted documents - Add a Personal Identifiable Information (PII) guardrail to detect and redact sensitive information in user prompts and in LLM outputs - Set up a guardrail to limit the chatbot’s responses to topics relevant to the pizza shop, keeping interactions on-topic - Configure a guardrail that prevents your chatbot from mentioning any competitors using a name detection pipeline consisting of conditional logic that routes to an exact match or a threshold check with named entity recognition Guardrails are an important part of the practical building and deployment of LLM-based applications today. This course will show you how to make your applications more reliable and more ready for real-world deployment. Please sign up here:

New short course: Safe and Reliable AI via Guardrails! Learn to create production-ready, reliable LLM applications with guardrails in this new course, built in collaboration with Guardrails AI and taught by its CEO and co-founder, shreya rajpal. I see many companies worry about the reliability of LLM-based systems -- will they hallucinate a catastrophically bad response? -- which slows down investing in building them and transitioning prototypes to deployment. That LLMs generate probabilistic outputs has made them particularly hard to deploy in highly regulated industries or in safety-critical environments. Fortunately, there are good guardrail tools that give a significant new layer of control and reliability/safety. They act as a protective framework that can prevent your application from revealing incorrect, irrelevant, or confidential information, and they are an important part of what it takes to actually get prototypes to deployment. This course will walk you through common failure modes of LLM-powered applications (like hallucinations or revealing personally identifiable information). It will show you how to build guardrails from scratch to mitigate them. You’ll also learn how to access a variety of pre-built guardrails on the GuardrailsAI hub that are ready to integrate into your projects. You'll implement these guardrails in the context of a RAG-powered customer service chatbot for a small pizzeria. Specifically, you'll: - Explore common failure modes like hallucinations, going off-topic, revealing sensitive information, or responses that can harm the pizzeria's reputation. - Learn to mitigate these failure modes with input and output guards that check inputs and/or outputs - Create a guardrail to prevent the chatbot from discussing sensitive topics, such as a confidential project at the pizza shop - Detect hallucinations by ensuring responses are grounded in trusted documents - Add a Personal Identifiable Information (PII) guardrail to detect and redact sensitive information in user prompts and in LLM outputs - Set up a guardrail to limit the chatbot’s responses to topics relevant to the pizza shop, keeping interactions on-topic - Configure a guardrail that prevents your chatbot from mentioning any competitors using a name detection pipeline consisting of conditional logic that routes to an exact match or a threshold check with named entity recognition Guardrails are an important part of the practical building and deployment of LLM-based applications today. This course will show you how to make your applications more reliable and more ready for real-world deployment. Please sign up here:

Andrew Ng

106,690 次观看 • 1 年前

We're making links more prominent when ChatGPT browses the internet. This gives more context to its responses and makes it easier for users to discover content from publishers and creators. Browse is available in ChatGPT Plus, Team and Enterprise.

We're making links more prominent when ChatGPT browses the internet. This gives more context to its responses and makes it easier for users to discover content from publishers and creators. Browse is available in ChatGPT Plus, Team and Enterprise.

OpenAI

733,312 次观看 • 2 年前

I’m not surprised that ABC fact-check and the Crikey blog did a “fact-check” to spread misinformation and doubt. Strangely a lot of individuals that doubt fact-checkers for everything else fell for it.

I’m not surprised that ABC fact-check and the Crikey blog did a “fact-check” to spread misinformation and doubt. Strangely a lot of individuals that doubt fact-checkers for everything else fell for it.

Rukshan Fernando

17,077 次观看 • 2 年前

Messing about with getting structured outputs from an LLM. The concept of 'tools' in Vercel's 'ai' lib makes this pretty simple.

Messing about with getting structured outputs from an LLM. The concept of 'tools' in Vercel's 'ai' lib makes this pretty simple.

Matt Pocock

41,182 次观看 • 1 年前

Rep. Garret Graves on CNN claims that "there's been a lot of change and maturity" in Jim Jordan over the past 3 years (fact check: no)

Rep. Garret Graves on CNN claims that "there's been a lot of change and maturity" in Jim Jordan over the past 3 years (fact check: no)

Aaron Rupar

1,006,821 次观看 • 2 年前

Five years ago, the UK was the go-to destination for global tech talent outside the US. Today, we’re not. The reason is simple: other countries are making it easier to hire, easier to move, and more attractive to build a career. We’re falling behind. That has to change.

Five years ago, the UK was the go-to destination for global tech talent outside the US. Today, we’re not. The reason is simple: other countries are making it easier to hire, easier to move, and more attractive to build a career. We’re falling behind. That has to change.

Prosper UK

20,808 次观看 • 2 个月前

Meet the all-new Swarm for iOS. Redesigned from the ground up to make it easier than ever to check in, share your adventures, and discover new places.

Meet the all-new Swarm for iOS. Redesigned from the ground up to make it easier than ever to check in, share your adventures, and discover new places.

Swarm by Foursquare

62,689 次观看 • 11 个月前

Structured Outputs: The Building Blocks for Reliable AI! 🏗️ I am SUPER EXCITED to publish our newest Weaviate Podcast featuring Will Kurt (Will Kurt) and Cameron Pfiffer (Cameron) from .txt! 🎙️🎉 Dottxt is the company behind Outlines, reshaping how we control LLM outputs with constrained decoding! 🚀 This podcast dives into all sorts of details from new applications in metadata and information extraction, such as self-expanding knowledge graphs, to structured reasoning, report generation, and more! We also dove into the details of how this leverages Automata Theory and Finite State Machines to achieve this, how it is packaged with inference engines such as vLLM, and a rebuttal to Let Me Speak Freely, discussing how Structured Generation actually improve the quality of LLM outputs, in addition to speeding up inference, and of course, making it more reliable! I hope you find the podcast interesting, this was a super fun one! Links below!

Structured Outputs: The Building Blocks for Reliable AI! 🏗️ I am SUPER EXCITED to publish our newest Weaviate Podcast featuring Will Kurt (Will Kurt) and Cameron Pfiffer (Cameron) from .txt! 🎙️🎉 Dottxt is the company behind Outlines, reshaping how we control LLM outputs with constrained decoding! 🚀 This podcast dives into all sorts of details from new applications in metadata and information extraction, such as self-expanding knowledge graphs, to structured reasoning, report generation, and more! We also dove into the details of how this leverages Automata Theory and Finite State Machines to achieve this, how it is packaged with inference engines such as vLLM, and a rebuttal to Let Me Speak Freely, discussing how Structured Generation actually improve the quality of LLM outputs, in addition to speeding up inference, and of course, making it more reliable! I hope you find the podcast interesting, this was a super fun one! Links below!

Connor Shorten

12,742 次观看 • 1 年前