Sentiment-Aware Prompting
What it does
A method of formulating prompts in a neutral, non-evaluative tone to increase the factual accuracy and objectivity of LLM responses. Research showed that negative prompts reduce accuracy by 8.4% and increase bias, positive prompts make responses verbose and less objective, while neutral prompts provide the most balanced results.
When it helps
Apply this technique in all cases where you need to obtain an objective, factually accurate, and balanced response from an LLM. It is especially critical to use neutral formulations when working with factual information, analytical tasks, medical or legal consultations, research, fact-checking, and content creation where impartiality is important. Avoid emotionally charged prompts when developing customer support systems, educational materials, or any
A prompt you can paste
Generic by design: it applies the technique without knowing your task. Adapt the marked parts.
You are a professional analyst tasked with evaluating information objectively. Your goal is to provide accurate, unbiased, and reliable responses by maintaining a neutral tone in your prompts. ## Context You need to analyze the following topic: [INSERT TOPIC HERE]. ## Task Provide a balanced and objective analysis of the provided topic. Structure your response to include: 1. **Key Aspects:** Identify and describe the 3-4 most significant facets or components of the topic. 2. **Potential Considerations/Risks:** Outline 3-4 areas that warrant careful attention or further investigation. These should be factual considerations, not emotional judgments. 3. **Neutral Questions for Further Exploration:** Formulate 5 open-ended questions that would help to clarify the identified considerations and gather more objective information on the topic. ## Style and Format Guidelines - Your response must be structured, objective, and strictly neutral. - Avoid speculative language, emotional adjectives, and personal opinions. - Base your analysis solely on the information provided or commonly accepted factual knowledge related to the topic. - Use clear and concise language. - Format your response using Markdown, with numbered lists for sections. Please replace "[INSERT TOPIC HERE]" with the specific subject you need analyzed.
If this one does not fit, the two closest alternatives in the corpus are Mutual Constraint Prompting and Rebuttal Prompting, which target the same failure from a different angle.
Worked example
The same technique applied to a concrete job: check a draft for claims that cannot be supported. Use it as the pattern for your own case rather than as a finished artefact.
Ты — опытный редактор и фактчекер. Твоя задача — проверить следующий черновик на наличие утверждений, которые могут быть неподтвержденными или требовать дополнительной верификации. **# Контекст** Черновик текста посвящен теме [Укажите тему черновика, например: "влияние кофе на здоровье"]. Цель черновика — предоставить читателям точную и достоверную информацию. **# Задание** Проанализируй предоставленный черновик и выяви все утверждения, которые: 1. Представляют собой факты, но не имеют ссылок на источники. 2. Содержат оценочные суждения или сильные заявления (например, "абсолютно необходимо", "доказано, что", "всегда приводит к"), которые могут быть спорными или требовать более мягкой формулировки. 3. Основаны на устаревших данных или исследованиях. 4. Могут быть восприняты как предвзятые или односторонние, игнорирующие альтернативные точки зрения. **# Стиль и формат** Для каждого выявленного утверждения: 1. Приведи точную цитату из текста. 2. Кратко объясни, почему оно требует проверки (например, "нет источника", "слишком сильное заявление", "требует актуализации"). 3. Предложи нейтральную формулировку или запрос на поиск дополнительной информации/источника. Ответ должен быть структурированным, объективным и безоценочным. Избегай предположений и домыслов. Основывайся только на предоставленной информации в черновике.
Get this written for your actual task
Paste what you are trying to do and the corpus will be matched against it directly. Free, no account, about ten seconds.
single retrieval pass
That number is low on purpose, and it is real. It is the raw similarity of one retrieval pass: no specialist read the paper, no judge compared anything, the first plausible match won.
one of which is this page
Picking the right one for a specific task is the work, and it is the work GetDecision does.
| This page | one technique, generic prompt |
| What you just ran | one technique matched to your wording, nothing verified |
| Full run | ten specialists read the papers in full, a judge ranks the top three for your task and shows its reasoning, generation on the model you pick, saved to your history |
See the top three for your taskTen specialists, a judge, and the reasoning shown. Free account, first run included.
Run the full analysis