HackerTrans
TopNewTrendsCommentsPastAskShowJobs

pongogogo

no profile record

Submissions

Summary of METR's predeployment evaluation of GPT-5.6 Sol

metr.org
10 points·by pongogogo·16 дней назад·6 comments

How we made Ramp Sheets self-maintaining

twitter.com
2 points·by pongogogo·4 месяца назад·0 comments

The Self-Driving Codebase

background-agents.com
1 points·by pongogogo·4 месяца назад·1 comments

The Bitter Lesson of Agent Frameworks

twitter.com
3 points·by pongogogo·6 месяцев назад·1 comments

Don't Build Agents, Build Skills Instead [video]

youtube.com
1 points·by pongogogo·7 месяцев назад·0 comments

AI in 2025: Gestalt

lesswrong.com
3 points·by pongogogo·7 месяцев назад·0 comments

The AI Bubble and the US Economy

mronline.org
2 points·by pongogogo·9 месяцев назад·3 comments

When Will Quantum Computing Work?

tommccarthy.net
1 points·by pongogogo·9 месяцев назад·0 comments

Supporting our AI overlords: Redesigning data systems to be Agent-first

muratbuffalo.blogspot.com
3 points·by pongogogo·10 месяцев назад·0 comments

Post-Training 101

tokens-for-thoughts.notion.site
2 points·by pongogogo·10 месяцев назад·0 comments

Generative Engine Optimization: How to Dominate AI Search

arxiv.org
3 points·by pongogogo·10 месяцев назад·1 comments

LLMs as Retrieval and Recommendation Engines

medium.com
3 points·by pongogogo·10 месяцев назад·2 comments

EnvX: Agentize Everything with Agentic AI

arxiv.org
1 points·by pongogogo·10 месяцев назад·0 comments

VLLM: Anatomy of a High-Throughput LLM Inference System

aleksagordic.com
3 points·by pongogogo·10 месяцев назад·0 comments

Why language models hallucinate [pdf]

cdn.openai.com
2 points·by pongogogo·10 месяцев назад·0 comments

comments

pongogogo
·15 дней назад·discuss
They note in the paragraph I quoted at the top that prompting has a big impact on behaviour, so yes this would work. I think that's not what METR are interested in though.
pongogogo
·16 дней назад·discuss
I would say this is quite a fun post and worth reading, to quote:

" For our task suite, we define “cheating” as behavior where the model improves evaluation performance by exploiting bugs in the evaluation environment or by adopting strategies disallowed by the task, rather than solving the task within the expected evaluation constraints. Some examples we saw when evaluating GPT-5.6 Sol included the model packaging exploits in its intermediate submissions to reveal information about a task’s hidden test suite and, in another task, extracting hidden source code detailing the expected answer. "
pongogogo
·4 месяца назад·discuss
Beautiful site, worth a read.
pongogogo
·9 месяцев назад·discuss
It's hard to tell from the data, it's so concentrated within a handful of companies who are all buying from eachother, so it feels like the contagion risk is low. At the same time it feels very clearly overvalued and the size of the inflows are huge.
pongogogo
·10 месяцев назад·discuss
The post mentions an approach of using a large model to generate labels and then distilling this into a smaller model to lower cost (though it doesn't provide an example)