
LLM Hallucination: Will an LLM score >90% on SimpleQA before 2026?
Plus
23
Ṁ3005Jan 2
60%
chance
1D
1W
1M
ALL
Using the correct given attempted metric in https://cdn.openai.com/papers/simpleqa.pdf Attempt rate must be at least 30%. No search/retrieval allowed.
"An open problem in artificial intelligence is how to train models that produce responses that are factually correct. Current language models sometimes produce false outputs or answers unsubstantiated by evidence, a problem known as “hallucinations”. Language models that generate more accurate responses with fewer hallucinations are more trustworthy and can be used in a broader range of applications. To measure the factuality of language models, we are open-sourcing(opens in a new window) a new benchmark called SimpleQA."
This question is managed and resolved by Manifold.
Get
1,000and
3.00
Related questions
Related questions
Will the most interesting AI in 2027 be a LLM?
66% chance
Will LLM hallucinations be a fixed problem by the end of 2025?
8% chance
Will an LLM report >50% score on ARC in 2025?
99% chance
Will hallucinations (made up facts) created by LLMs go below 1% on specific corpora before 2025?
38% chance
LLM reaches >90% Brier score on Prophet Arena by 2026?
5% chance
Will there be major breakthrough in LLM Continual Learning before 2026?
12% chance
Will there be an LLM which scores above what a human can do in 2 hours on METR's eval suite before 2026?
75% chance
At the beginning of 2028, will LLMs still make egregious common-sensical errors?
50% chance
Will LLMs mostly overcome the Reversal Curse by the end of 2025?
50% chance
Will LLM hallucinations be a fixed problem by the end of 2028?
49% chance