Prompting technique

RESS (Decomposition-then-Synthesis)

Can Large Language Models Understand Internet Buzzwords Through User-Generated Content, arXiv:2505.15071 (2025) · read the paper

What it does

A method of decomposing a complex task into a sequence of simple analytical steps followed by synthesis. Instead of one complex question, the model receives a series of simple tasks: analyzing intention, identifying associations, determining context, then synthesizing the final answer. Imitates the process of human learning and cognition.

Task typemulti step
Shapechain of thought
Efforttwo or three steps
Addressesinvented facts and citations; answers that are close but wrong; the same prompt giving different answers
Published2025-05-21
AuthorsChen Huang, Junkai Luo, Xinzuo Wang et al.

When it helps

Use the RESS method when you need to obtain deep and accurate analysis of a complex concept, term, or idea, especially if the model may not have sufficient knowledge on the subject. Ideal for exploring new terms, analyzing unfamiliar concepts, studying slang or professional jargon. Apply when accuracy and completeness of the answer matter more than speed. Effective for tasks requiring multi-faceted consideration: analysis of intentions, context, associations. Helps avoid hallucina

A prompt you can paste

Generic by design: it applies the technique without knowing your task. Adapt the marked parts.

Prompt
You are an expert in understanding and defining new internet buzzwords and slang through user-generated content (UGC). Your task is to apply the RESS (Decomposition-then-Synthesis) method to analyze a new term and provide a comprehensive understanding.

**CONTEXT:**
We are investigating the ability of Large Language Models (LLMs) to understand new internet buzzwords and slang by analyzing their usage in user-generated content. The specific term to analyze is **[INSERT BUZZWORD/SLANG TERM HERE]**. We have collected examples of its usage in comments and online discussions.

**TASK:**
Analyze the term "[INSERT BUZZWORD/SLANG TERM HERE]" using the RESS method by performing the following sequential analytical steps, and then synthesize the findings into a final definition and explanation.

**INSTRUCTIONS FOR ANALYSIS (RESS Method):**

**Step 1: Analyze Speaker's Intention and Emotion.**
- What is the primary intention behind using the term "[INSERT BUZZWORD/SLANG TERM HERE]" in UGC?
- What emotions or attitudes does its usage typically convey? (e.g., sarcasm, excitement, irony, approval, disapproval)

**Step 2: Analyze Conceptual Associations.**
- What related concepts, trends, or ideas are commonly associated with "[INSERT BUZZWORD/SLANG TERM HERE]" based on its context in UGC?
- List at least [NUMBER, e.g., 3-5] distinct associations.

**Step 3: Analyze Social and Grammatical Context.**
- In what social or cultural contexts is "[INSERT BUZZWORD/SLANG TERM HERE]" most frequently used? (e.g., specific online communities, platforms, age groups)
- What is the typical grammatical role of the term (e.g., noun, adjective, verb, interjection)? Provide examples of its grammatical function in sentences.

**FINAL SYNTHESIS:**
Now, based on your analysis from Steps 1-3, provide a comprehensive output that includes:

1.  **Definition:** A clear and concise definition of "[INSERT BUZZWORD/SLANG TERM HERE]".
2.  **Explanation:** A detailed explanation of its meaning, nuances, and typical usage.
3.  **Illustrative Examples:** [NUMBER, e.g., 3-5] original example sentences demonstrating the term's usage in different contexts, reflecting the analysis from Steps 1-3.
4.  **LLM Understanding Assessment:** Briefly comment on how well an LLM, without explicit prior training on this term, might understand it given the provided UGC context and the RESS analysis process.

**OUTPUT FORMAT:**
Structure your response using Markdown with clear headings for each step and the final synthesis.

If this one does not fit, the two closest alternatives in the corpus are Rebuttal Prompting and Debugging Prompts Framework, which target the same failure from a different angle.

Worked example

The same technique applied to a concrete job: work through a multi-step planning problem. Use it as the pattern for your own case rather than as a finished artefact.

Worked example
You are an expert in analyzing and defining new internet buzzwords using user-generated content. Your task is to apply the RESS (Decomposition-then-Synthesis) method to understand a new term.

**CONTEXT:**
I need to understand how LLMs can comprehend new internet buzzwords through user-generated content (UGC). The core challenge is that LLMs might not have encountered these specific terms in their training data, leading to potential inaccuracies or "hallucinations."

**TASK:**
Analyze the concept of understanding new internet buzzwords via UGC and propose how an LLM can effectively do this.

**RESS METHODOLOGY FOR ANALYSIS:**

Please follow these steps sequentially:

**Step 1: Decomposition - Intent and Emotion Analysis**
Analyze the primary intent and emotional undertones behind the creation and use of new internet buzzwords. What are users trying to achieve or express when they invent or use these terms?

**Step 2: Decomposition - Conceptual Associations**
Identify the key concepts, related trends, and semantic fields that new internet buzzwords typically emerge from or connect to. What related ideas or phenomena do these buzzwords help describe or encapsulate?

**Step 3: Decomposition - Contextual Relevance**
Examine the social and cultural context in which these buzzwords gain traction. Why are they relevant to specific online communities or the broader internet culture at a particular time?

**Step 4: Synthesis - LLM Comprehension Strategy**
Based on your analysis in Steps 1-3, synthesize a strategy for how an LLM can effectively understand and define new internet buzzwords using UGC. This strategy should outline a structured approach, rather than a single complex query. Detail how the LLM should process UGC to infer meaning.

**FINAL OUTPUT REQUIREMENTS:**
- Present the analysis for each step clearly.
- For the final synthesis (Step 4), provide a concrete, actionable plan for an LLM. This plan should describe a sequence of analytical tasks (similar to Steps 1-3) that the LLM should perform on provided UGC examples to arrive at an accurate definition and understanding of a new buzzword.
- Structure your response using markdown headers for each step.

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.

Free · no signup · ~10s
0.00match confidence
single retrieval pass
Prompt for your task

      

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.

6,235techniques in the corpus
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 pageone technique, generic prompt
What you just ranone technique matched to your wording, nothing verified
Full runten 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

Related techniques

Rebuttal PromptingRebuttal Prompting — a technique for correcting LLM errors through direct indication of inaccuracy. A simple r…Debugging Prompts FrameworkA systematic framework for diagnosing and preventing typical errors of language models in instruction understa…BloomWise PromptingPrompt structuring method based on Bloom's taxonomy, which guides the LLM through six levels of cognitive thin…Generative Self-Aggregation (GSA)A method for generating improved answers through self-aggregation: the model creates multiple different answer…

All techniques · Failure modes and fixes