saliagato · HackerTrans

Submissions

200B Weights of Responsibility: The Stress of Working in Modern AI [pdf]

gwern.net

1 pointsby saliagato2 lata temu0 comments

Inference Wholesale

inference.net

1 pointsby saliagato2 lata temu0 comments

AI the Product vs. AI the Feature [video]

youtube.com

3 pointsby saliagato2 lata temu0 comments

How the voices for ChatGPT were chosen

openai.com

54 pointsby saliagato2 lata temu71 comments

A Message to Our Customers (2016)

apple.com

1 pointsby saliagato2 lata temu0 comments

Defending LLMs against Jailbreaking Attacks via Backtranslation

arxiv.org

67 pointsby saliagato2 lata temu48 comments

How Do Humans Write Code? Large Models Do It the Same Way Too

arxiv.org

1 pointsby saliagato2 lata temu0 comments

Reserved Capacity

openai.com

1 pointsby saliagato2 lata temu0 comments

People will be able to work in their sleep

independent.co.uk

1 pointsby saliagato2 lata temu2 comments

Exploring OCR Capabilities of GPT-4V

arxiv.org

2 pointsby saliagato2 lata temu0 comments

If you jailbreak this GPT you win $100

chat.openai.com

2 pointsby saliagato3 lata temu0 comments

Wearable AI raises $1.9M – Aims to replace God

fastcompany.com

2 pointsby saliagato3 lata temu1 comments

Language Models May Not Be Few-Shot Anymore

arxiv.org

13 pointsby saliagato3 lata temu0 comments

Will scaling work?

dwarkeshpatel.com

242 pointsby saliagato3 lata temu283 comments

Airbnb is deploying AI to block New Year's Eve bookings that could be parties

news.airbnb.com

18 pointsby saliagato3 lata temu64 comments

GPT-4 Can Almost Perfectly Handle Unnatural Scrambled Text

arxiv.org

202 pointsby saliagato3 lata temu142 comments

OpenAI’s November 6th DevDay has been leaked

the-decoder.com

5 pointsby saliagato3 lata temu3 comments

Ask HN: Is OpenAI exposing user data via Google Search?

3 pointsby saliagato3 lata temu3 comments

Ask HN: How does GPT-4V work?

4 pointsby saliagato3 lata temu0 comments

LLMs trained on “A is B” fail to learn “B is A”

paperswithcode.com

120 pointsby saliagato3 lata temu158 comments

comments

saliagato

2 lata temudiscuss

[flagged]

saliagato

2 lata temudiscuss

Can't comment on the cost (varies depending on who you are and what you do in your life) but the benefit is really high.

source: I started from 0 and the very first bounty I submitted (related to CEX) was valued to $100k+

saliagato

2 lata temudiscuss

Speed of development, developer experience, ready to go templates. You should check it out, it’s really cool

saliagato

2 lata temudiscuss

gpt-4 was indeed trained on gpt-3 instruct series (davinci, specifically). gpt-4 was never a newly trained model

saliagato

2 lata temudiscuss

worked 100% of the time for me

saliagato

2 lata temudiscuss

This sounds so fictionalized that may be fake

saliagato

2 lata temudiscuss

[flagged]

saliagato

2 lata temudiscuss

This post is still the first in the feed

saliagato

2 lata temudiscuss

basically, yes. Pinecone? Dead. Azure AI Search? Dead. Quadrant? Dead.

saliagato

3 lata temudiscuss

Hard to believe OpenAI uses Quadrant when they are backed by Microsoft, thus having Azure Cognitive Search (now "AI" Search)

saliagato

3 lata temudiscuss

saliagato

3 lata temudiscuss

I wonder how many were left out

saliagato

3 lata temudiscuss

Same. Tip: if you put a random number it will get you through regardless

saliagato

3 lata temudiscuss

this is so cool lol

saliagato

3 lata temudiscuss

Why do you (dang) always write a comment specifying that people can read more and even providing some links when it's clear that when you reach the bottom of the page you have to click "read more" to indeed read more. Isn't it a bit useless?

saliagato

3 lata temudiscuss

same topic different tweets

saliagato

3 lata temudiscuss

Hopefully they don’t

saliagato

3 lata temudiscuss

We all know what. HN moderators are deleting all related comments.

Edit: dang is right, sorry y’all

saliagato

3 lata temudiscuss

I think they are running their RAG on Microsoft's Azure Cognitive Search. Satya mentioned something about it in the yesterday Microsoft Ignite keynote

saliagato

3 lata temudiscuss

It can’t count because what the LLM sees is a bunch of tokens not words