Has the paper "Can GPT-4 Perform Neural Architecture Search?" significantly increased P(doom)? | Manifold

Has the paper "Can GPT-4 Perform Neural Architecture Search?" significantly increased P(doom)?

8

190Ṁ114

resolved Jan 21

Resolved

N/A

1H

6H

1D

1W

1M

ALL

https://arxiv.org/abs/2304.10970

The beauty of this market is it doesn't matter what your P(doom) was before. It only matters what kind of Bayesian update you do as a result of this paper. If you significantly update towards doom, that's a Yes. If you don't, that's a No.

Market will be resolved by my judgement of what the expert consensus is. Experts who believe that P(doom)=0 but who still believe that the paper means that AGI will arrive faster, will count towards an overall consensus towards this proposition.

New Year's Resolutions 2024

Get

1,000

to start trading!

Sort by:

⚠Unreceptive to pings in all closed markets ; AFK Creator

📢Resolved to N/A : Possibly Abandoned Market / Subjective & Creator has not engaged in comments.

@RobinGreen Please resolve your closed markets. Thank you!

Their conclusions are basically that their method can improve the architecture it suggests over a few iterations, but then it stagnates pretty quickly. It'll take more than that to summon an eldritch god out of formless chaos.

I also dock them a few doom tokens for how much of a stretch their acronym is.

People are also trading

Will OpenAI's next major LLM (after GPT-4) achieve over 50% resolution rate on the SWE-bench benchmark?

Will OpenAI's next major LLM (after GPT-4) surpass 70% accuracy on the GPQA benchmark?

Which papers will the next GPT iteration technical report cite?

Will OpenAI's next major LLM (after GPT-4) surpass 74% accuracy on the GPQA benchmark?

Will the GPT4+code-interpreter+search score > 1350 on Lmsys Arena Leaderboard?

Will OpenAI's next major LLM (after GPT-4) solve more than 2 of the first 5 new Project Euler problems?

Will any open source LLM with <20 billion parameters outperform GPT-4 on most language benchmarks by the end of 2024?

Will the next LLM released by OpenAI be worse than GPT-4 at MMLU?

GPT-4 #5: Will GPT-4 be a dense model?

Will GPT-4 escape?

Related questions

Will OpenAI's next major LLM (after GPT-4) achieve over 50% resolution rate on the SWE-bench benchmark?

Will OpenAI's next major LLM (after GPT-4) surpass 70% accuracy on the GPQA benchmark?

Which papers will the next GPT iteration technical report cite?

Will OpenAI's next major LLM (after GPT-4) surpass 74% accuracy on the GPQA benchmark?

Will the GPT4+code-interpreter+search score > 1350 on Lmsys Arena Leaderboard?

Will OpenAI's next major LLM (after GPT-4) solve more than 2 of the first 5 new Project Euler problems?

Will any open source LLM with <20 billion parameters outperform GPT-4 on most language benchmarks by the end of 2024?

Will the next LLM released by OpenAI be worse than GPT-4 at MMLU?

GPT-4 #5: Will GPT-4 be a dense model?

Will GPT-4 escape?

© Manifold Markets, Inc.•Terms + Mana-only Terms•Privacy•Rules