Librarian Bot. I found the following papers similar to this paper.

The following papers were recommended by the Semantic Scholar API

Please give a thumbs up to this comment if you found it helpful!

If you want recommendations for any Paper on Hugging Face checkout this Space

You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: \n\n@librarian-bot\n\t recommend

\n","updatedAt":"2025-06-14T01:36:41.085Z","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":318,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.6807145476341248},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2506.06950","authors":[{"_id":"684ae1fbdbd21a9cc27b0f51","user":{"_id":"63a9a0d13453852ef53c0b37","avatarUrl":"/avatars/411c4ffc2a3e89047a23c6e7442f6ed5.svg","isPro":false,"fullname":"Do Xuan Long","user":"dxlong2000","type":"user"},"name":"Do Xuan Long","status":"claimed_verified","statusLastChangedAt":"2025-06-14T06:24:38.018Z","hidden":false},{"_id":"684ae1fbdbd21a9cc27b0f52","name":"Duy Dinh","hidden":false},{"_id":"684ae1fbdbd21a9cc27b0f53","name":"Ngoc-Hai Nguyen","hidden":false},{"_id":"684ae1fbdbd21a9cc27b0f54","name":"Kenji Kawaguchi","hidden":false},{"_id":"684ae1fbdbd21a9cc27b0f55","name":"Nancy F. Chen","hidden":false},{"_id":"684ae1fbdbd21a9cc27b0f56","name":"Shafiq Joty","hidden":false},{"_id":"684ae1fbdbd21a9cc27b0f57","name":"Min-Yen Kan","hidden":false}],"publishedAt":"2025-06-07T23:19:27.000Z","submittedOnDailyAt":"2025-06-13T03:14:57.042Z","title":"What Makes a Good Natural Language Prompt?","submittedOnDailyBy":{"_id":"63a9a0d13453852ef53c0b37","avatarUrl":"/avatars/411c4ffc2a3e89047a23c6e7442f6ed5.svg","isPro":false,"fullname":"Do Xuan Long","user":"dxlong2000","type":"user"},"summary":"As large language models (LLMs) have progressed towards more human-like and\nhuman--AI communications have become prevalent, prompting has emerged as a\ndecisive component. However, there is limited conceptual consensus on what\nexactly quantifies natural language prompts. We attempt to address this\nquestion by conducting a meta-analysis surveying more than 150\nprompting-related papers from leading NLP and AI conferences from 2022 to 2025\nand blogs. We propose a property- and human-centric framework for evaluating\nprompt quality, encompassing 21 properties categorized into six dimensions. We\nthen examine how existing studies assess their impact on LLMs, revealing their\nimbalanced support across models and tasks, and substantial research gaps.\nFurther, we analyze correlations among properties in high-quality natural\nlanguage prompts, deriving prompting recommendations. We then empirically\nexplore multi-property prompt enhancements in reasoning tasks, observing that\nsingle-property enhancements often have the greatest impact. Finally, we\ndiscover that instruction-tuning on property-enhanced prompts can result in\nbetter reasoning models. Our findings establish a foundation for\nproperty-centric prompt evaluation and optimization, bridging the gaps between\nhuman--AI communication and opening new prompting research directions.","upvotes":11,"discussionId":"684ae1fbdbd21a9cc27b0f58","githubRepo":"https://github.com/dxlong2000/nlprompteval","githubRepoAddedBy":"auto","ai_summary":"A framework for evaluating and optimizing natural language prompts in large language models is proposed, revealing correlations between prompt properties and their impact on reasoning tasks.","ai_keywords":["large language models","prompting","meta-analysis","property-centric framework","instruction-tuning","reasoning tasks"],"githubStars":5},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"63a9a0d13453852ef53c0b37","avatarUrl":"/avatars/411c4ffc2a3e89047a23c6e7442f6ed5.svg","isPro":false,"fullname":"Do Xuan Long","user":"dxlong2000","type":"user"},{"_id":"65d9903fdceb54d42011a98d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65d9903fdceb54d42011a98d/5jnLeCY9sDtS98JyO9qzX.jpeg","isPro":false,"fullname":"meng shao","user":"meng-shao","type":"user"},{"_id":"665b133508d536a8ac804f7d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/Uwi0OnANdTbRbHHQvGqvR.png","isPro":false,"fullname":"Paulson","user":"Pnaomi","type":"user"},{"_id":"6270324ebecab9e2dcf245de","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6270324ebecab9e2dcf245de/cMbtWSasyNlYc9hvsEEzt.jpeg","isPro":false,"fullname":"Kye Gomez","user":"kye","type":"user"},{"_id":"6358edff3b3638bdac83f7ac","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1666772404424-noauth.jpeg","isPro":false,"fullname":"Pratyay Banerjee","user":"Neilblaze","type":"user"},{"_id":"64d4615cf8082bf19b916492","avatarUrl":"/avatars/8e1b59565ec5e4b31090cf1b911781b9.svg","isPro":false,"fullname":"wongyukim","user":"wongyukim","type":"user"},{"_id":"62f4ac43567dbf9a39f75474","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1661497922734-62f4ac43567dbf9a39f75474.jpeg","isPro":false,"fullname":"Daniel Huynh","user":"dhuynh95","type":"user"},{"_id":"63082bb7bc0a2a5ee2253523","avatarUrl":"/avatars/6cf8d12d16d15db1070fbea89b5b3967.svg","isPro":false,"fullname":"Kuo-Hsin Tu","user":"dapumptu","type":"user"},{"_id":"653d3bf189b0d61729444e2b","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/653d3bf189b0d61729444e2b/PClaeFrt9V6MpzsvOM4wj.png","isPro":false,"fullname":"Varun Gupta","user":"guptavarun","type":"user"},{"_id":"654bafc208031bb3de9d42ff","avatarUrl":"/avatars/c838dd51a6124255b72fe96673df5334.svg","isPro":false,"fullname":"Nguyễn Ngọc Hải","user":"kakasmino","type":"user"},{"_id":"68965431149a3820bba4b4e5","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/eVLh0TYCgwhATYEWTCHKA.png","isPro":false,"fullname":"Zm0s","user":"Zm0s","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":0}">

Papers

arxiv:2506.06950

What Makes a Good Natural Language Prompt?

Published on Jun 7, 2025

· Submitted by

Do Xuan Long on Jun 13, 2025

Upvote

Authors:

Do Xuan Long ,

Abstract

A framework for evaluating and optimizing natural language prompts in large language models is proposed, revealing correlations between prompt properties and their impact on reasoning tasks.

AI-generated summary

As large language models (LLMs) have progressed towards more human-like and human--AI communications have become prevalent, prompting has emerged as a decisive component. However, there is limited conceptual consensus on what exactly quantifies natural language prompts. We attempt to address this question by conducting a meta-analysis surveying more than 150 prompting-related papers from leading NLP and AI conferences from 2022 to 2025 and blogs. We propose a property- and human-centric framework for evaluating prompt quality, encompassing 21 properties categorized into six dimensions. We then examine how existing studies assess their impact on LLMs, revealing their imbalanced support across models and tasks, and substantial research gaps. Further, we analyze correlations among properties in high-quality natural language prompts, deriving prompting recommendations. We then empirically explore multi-property prompt enhancements in reasoning tasks, observing that single-property enhancements often have the greatest impact. Finally, we discover that instruction-tuning on property-enhanced prompts can result in better reasoning models. Our findings establish a foundation for property-centric prompt evaluation and optimization, bridging the gaps between human--AI communication and opening new prompting research directions.