Question

What will be the best perplexity score by a language model on the Penn Treebank (Word Level) by the end of 2024?

3 comments3

Total Forecasters15

Community Prediction10.9

Make a Prediction

My Prediction

community

lower 25%

median

upper 75%

...

8.21

10.9

16.6

Authors:

Matthew_Barnett

Opened:

Closes:

Scheduled resolution:

AI Progress Essay Contest

AI Technical Benchmarks

Computing and Math

Artificial Intelligence

When will a language model be developed that, when tested, yields approximately human-level output?

2024-06

What will be the state-of-the-art language modelling performance (in perplexity) on WikiText-103 by the following dates?

Human-Level Language Models

Comments

? comments

Authors:

Matthew_Barnett

Opened:

Closes:

Scheduled resolution:

AI Progress Essay Contest

AI Technical Benchmarks

Computing and Math

Artificial Intelligence

SoftBank to Invest $500M in OpenAI for the First Time—Is It Late to the Party?

Observer (NY)•Oct 1, 2024

The Improvisational Nature of LLMs and Human Thought

Psychology Today•Oct 1, 2024

AI and the Human Mind: Uncovering the Machine "Unconscious"

Psychology Today•Oct 1, 2024

Learn more about Metaculus NewsMatch

When will a language model be developed that, when tested, yields approximately human-level output?

2024-06

MetaculusM

🚀Metaculus Open-Source Rewrite is Live!

What will be the best perplexity score by a language model on the Penn Treebank (Word Level) by the end of 2024?

Make a Prediction

When will a language model be developed that, when tested, yields approximately human-level output?

What will be the state-of-the-art language modelling performance (in perplexity) on WikiText-103 by the following dates?

Human-Level Language Models

Comments

When will a language model be developed that, when tested, yields approximately human-level output?

What will be the state-of-the-art language modelling performance (in perplexity) on WikiText-103 by the following dates?

Human-Level Language Models

When will a language model be developed that, when tested, yields approximately human-level output?

What will be the state-of-the-art language modelling performance (in perplexity) on WikiText-103 by the following dates?

Human-Level Language Models

When will a language model be developed that, when tested, yields approximately human-level output?

What will be the state-of-the-art language modelling performance (in perplexity) on WikiText-103 by the following dates?

Human-Level Language Models