MMLU: Public Relations

Accuracy on MMLU's public relations questions, one of 57 subject tests of academic and professional knowledge.

Also known as: public_relations

unassessed

This page is a discovery lead. Nobody has yet assessed it against the catalogue contract, so it carries no disposition. Absence of evidence here is not evidence of staleness.
Categoryknowledge
SubcategorySocial Sciences
Page statusactive
Metricaccuracy
Directionhigher_is_better
Unit%
Dataset size110
Dataset licenceMIT
PublisherUC Berkeley

What it measures

Public-relations principles and practice: media relations, crisis communication, reputation management, and publicity campaigns. Framed as four-option multiple-choice questions and scored zero-shot or few-shot by exact match against the labelled option, as one of the 57 subject subsets that make up the MMLU benchmark.

Task format

Four-option multiple-choice question answering (A-D), one correct answer, zero-shot or few-shot.

Models reporting this benchmark

These figures come from the model cards, which carry one collection date per card and no per-score attribution. They are shown as reported, not as verified evidence.
ModelProviderScoreCard as of
Mixtral 8x22B Instruct v0.1Mistral AI77.32026-04
Yi 1.5 34B01.AI76.42026-04
Yi 1.5 34B 32K01.AI75.52024-07
Meta Llama 3 70BMeta74.52024-07
Meta Llama 3 70B InstructMeta74.52026-04
Meta Llama 3 70B InstructNous Research74.52026-04
Yi 9B01.AI74.52024-07
Phi 3 mini 4K instructMicrosoft73.62024-07
Yi 1.5 9B 32K01.AI73.62024-07
Yi 6B01.AI73.62024-07
Yi 6B Chat01.AI73.62024-07
Yi 1.5 34B Chat01.AI72.72026-04
Yi 1.5 34B Chat 16K01.AI72.72024-07
Yi 1.5 9B Chat 16K01.AI72.72024-07
Yi 34B 200K01.AI72.72024-07
Yi 34B Chat01.AI72.72024-07
Meta Llama 3 8BMeta71.82024-07
Meta Llama 3 8BNous Research71.82024-07
Nous Hermes 2 Yi 34BNous Research71.82024-07
gemma 7B itGoogle DeepMind70.92024-07
Mistral 7B Instruct v0.2Mistral AI70.92024-07
Mixtral 8x7B Instruct v0.1Mistral AI70.92026-04
Yi 1.5 9B01.AI70.92024-07
Mixtral 8x7B v0.1Mistral AI70.02026-04
Nous Hermes 2 SOLAR 10.7BNous Research70.02024-07
Nous Hermes 2 Mixtral 8x7B DPONous Research69.12024-07
Yi 1.5 6B01.AI69.12024-07
Yi 1.5 9B Chat01.AI69.12024-07
Phi 3 mini 128K instructMicrosoft67.32024-07
Hermes 2 Theta Llama 3 8BNous Research65.52024-07
Meta Llama 3 8B InstructMeta64.52024-07
Meta Llama 3 8B InstructNous Research64.52024-07
Mistral 7B v0.3Mistral AI64.52024-07
mistral 7B v0.3 bnb 4bitUnsloth64.52024-07
Yi 1.5 6B Chat01.AI64.52024-07
falcon 40BTII62.72024-07
phi 2Microsoft62.72024-07
Hermes 2 Pro Llama 3 8BNous Research61.82024-07
Qwen2 1.5B InstructAlibaba / Qwen Team60.02024-07
deepseek llm 7B baseDeepSeek59.12024-07
deepseek llm 7B chatDeepSeek59.12024-07
chatglm2 6BZhipu AI58.22024-07
deepseek coder 6.7B instructDeepSeek52.72024-07
deepseek coder 6.7B baseDeepSeek50.92024-07
Qwen2 0.5B InstructAlibaba / Qwen Team50.02024-07
gemma 2BGoogle DeepMind47.32024-07
gemma 2B itGoogle DeepMind41.82024-07
deepseek coder 1.3B baseDeepSeek33.62024-07
deepseek coder 1.3B instructDeepSeek33.62024-07
OLMo 1B hfAllen AI24.52024-07

Data

This page as JSON · Edit on GitHub