RLHF is just barely RL - Hackdot

Hackdot

News for hackers. Stuff that matters.

Sections

Top
Latest
Ask HN
Show HN
Jobs
Polls
Older Stuff
Search

About

Every story and comment from the Hacker News API, in a layout borrowed from 2001.

A comment on RLHF is just barely RL. Parent comment

Re:RLHF is just barely RL by imtringued on Thursday August 08, 2024 @10:24AM (#41191869)

Rust is that way.

[ Parent | on HN ]

There are no comments on this comment yet.

Ask HN

Anyone else feels Grok 4.7 got nerfed?
Is misaligned AI less dangerous than AI controlled by a bad actor
How angry are you about Apple's walled garden when using Android?
Could Stanford's NCM potentially replace modern LLMs in the future?
How do you keep secrets out of your AI agents' context?
What other tools similar to Alibaba OCR do people use with LLMs?

Show HN

Ziperu is rubyzip with a Rust core, same API but faster and safer
Git-dupe – plain Git per-project private versioning for agent files
Dryplate MCP – local image tools for AI agents (HEIC, strip metadata)
Leftagraph – a way to chart a company's finances over time
Deedglass – Verify AI Agent SQL Changes Using Postgres WAL
ContentRank – web pages fight 1v1 matches, Glicko-2 ranks the winners

Jobs

Y/N (YC S23) Is Hiring a Growth Lead to Distribute Superhuman A.I. Storytelling
UpCodes (YC S17) is hiring SDR's to sell AI compliance software
Mbodi AI (YC P25) Is Hiring Research and Robotic Engineers
Bild AI (YC W25) Is Hiring Founding Product Engineers

Stories and comments come from the Hacker News API and belong to their authors. Not affiliated with Y Combinator or Slashdot.