A major ML paper demonstrates symbolic-enhanced transformer successor outperforming standard transformers by March 2025

17

Ṁ1kṀ4.1k

resolved Apr 2

Resolved

NO

1H

6H

1D

1W

1M

ALL

Will a published machine learning paper demonstrate a new architecture that combines transformers with symbolic methods (category theory, programming language theory, or logic theory) and achieves superior performance on standard benchmarks compared to traditional transformer-only architectures?

Resolution criteria: Paper must be published on arXiv, and very much preferably a major ML conference/journal, and show statistically significant improvements over baseline transformers on multiple standard tasks.

Market context

Technical AI Timelines

Mechanistic interpretability

Get

1,000

to start trading!

🏅 Top traders

#	Trader	Total profit
1		Ṁ458
2		Ṁ214
3		Ṁ124
4		Ṁ116
5		Ṁ80

Sort by:

rip to me

I feel like I'm gonna be off with the timing of such a result by like a couple of days with this one aren't I lmao

People are also trading

Will an AI achieve >85% performance on the FrontierMath benchmark before 2028?

On January 1, 2027, a Transformer-like model will continue to hold the state-of-the-art position in most benchmark

Will superposition in transformers be mostly solved by 2026?

If OpenAI makes a transformer sized advancement in the next 5 years, will they publish an accompanying paper?

Will transformer architectures lose their dominant position in deep learning before 2028?

Will the most capable, public multimodal model at the end of 2027 in my judgement use a transformer-like architecture?

Are Transformers “the last technology to do all of ML” (before AGI)?

Will Transformer-Based LLMs Make Up ≥75% of Parameters in the Top General AI by 2030?

Will there be a significant advancement in frontier AI model architecture by end of year 2026?

Will an AI achieve >85% performance on the FrontierMath benchmark before 2027?

Related questions

Will an AI achieve >85% performance on the FrontierMath benchmark before 2028?

On January 1, 2027, a Transformer-like model will continue to hold the state-of-the-art position in most benchmark

Will superposition in transformers be mostly solved by 2026?

If OpenAI makes a transformer sized advancement in the next 5 years, will they publish an accompanying paper?

Will transformer architectures lose their dominant position in deep learning before 2028?

Will the most capable, public multimodal model at the end of 2027 in my judgement use a transformer-like architecture?

Are Transformers “the last technology to do all of ML” (before AGI)?

Will Transformer-Based LLMs Make Up ≥75% of Parameters in the Top General AI by 2030?

Will there be a significant advancement in frontier AI model architecture by end of year 2026?

Will an AI achieve >85% performance on the FrontierMath benchmark before 2027?

© Manifold Markets, Inc.•Terms•Privacy