Skip to content

AI Dynamics

Global AI News Aggregator

Rechercher

PROMPT ENGINEERING

Improving LLM Reasoning Through Multi-Option Comparison

By

@godofprompt

–

07 February 2025 10h35

4/ Contrast Multiple Approaches

Want the best solution? Have deepseek-R1 compare multiple options.

Example:
“List 10 ways to achieve X. Then determine the most effective approach.”

This increases its reasoning time and improves decision-making. pic.twitter.com/v8fpgGaizk
— God of Prompt (@godofprompt) 7 février 2025

4/ Contrast Multiple Approaches Want the best solution? Have deepseek-R1 compare multiple options. Example:
“List 10 ways to achieve X. Then determine the most effective approach.” This increases its reasoning time and improves decision-making.

→ View original post on X — @godofprompt

7 February 2025
Improving AI Response Accuracy via Self-Reflection Prompting

By

@godofprompt

–

07 February 2025 10h35

3/ Encourage Self-Reflection

The model is great at checking its own work—if you ask it to.

Prompt it with:
“Identify potential flaws in your approach and suggest solutions.”

You’ll get a more refined and accurate response. pic.twitter.com/WrHq6DtAA9
— God of Prompt (@godofprompt) 7 février 2025

3/ Encourage Self-Reflection The model is great at checking its own work—if you ask it to. Prompt it with:
“Identify potential flaws in your approach and suggest solutions.” You’ll get a more refined and accurate response.

→ View original post on X — @godofprompt

7 February 2025
Optimizing Prompting Strategies for Deepseek-R1 Reasoning

By

@godofprompt

–

07 February 2025 10h35

2/ Encourage Step-wise Reasoning
Deepseek-R1 performs significantly better when it works step by step.

Use phrases like:
“Break down the problem step by step and solve it.”

This forces deeper thinking and better results. pic.twitter.com/kzihKoeRYI
— God of Prompt (@godofprompt) 7 février 2025

2/ Encourage Step-wise Reasoning
Deepseek-R1 performs significantly better when it works step by step. Use phrases like:
“Break down the problem step by step and solve it.” This forces deeper thinking and better results.

→ View original post on X — @godofprompt

7 February 2025
Improving DeepSeek-R1 Output Quality via Role Specification

By

@godofprompt

–

07 February 2025 10h35

1/ Role Specification
Always assign a relevant role to deepseek-R1. Example:
“You are a senior software engineer specializing in AI infrastructure.” This sets context and improves output quality.

→ View original post on X — @godofprompt

7 February 2025
Optimizing Prompting Techniques for DeepSeek-R1 Performance

By

@godofprompt

–

07 February 2025 10h35

Many people struggle to get the best responses from deepseek-R1. They assume it works like ChatGPT. But with the right prompting techniques, you can unlock its full potential. Here’s how to improve accuracy and encourage deep thinking:

→ View original post on X — @godofprompt

7 February 2025
Cursor Transforms Into Devin Through Advanced Prompt Engineering

By

@localghost

–

07 February 2025 6h15

Cursor turns into Devin with just a few prompts.

→ View original post on X — @localghost

7 February 2025
Language Priors Steering Thoughts in AI Models

By

@karpathy

–

07 February 2025 4h25

Cute idea, reminds me of “let’s think step by step” trick. Both lean on the language prior to steer the thoughts.

→ View original post on X — @karpathy

7 February 2025
Pydantic AI: Agent Engineering with Graphs and Tools

By

@latentspacepod

–

07 February 2025 0h12

Agent Engineering with Pydantic + Graphs https://
latent.space/p/pydantic with @samuelcolvin
! We've heard from @jxnlco 3 years running that @Pydantic Is All We Need, and this is even more true today with the launch of Pydantic AI! With a little @marimo_io love at the end 🙂

→ View original post on X — @latentspacepod

7 February 2025
Gratitude for Most-Used GPT Among Nearly 100 Models

By

@thatroblennon

–

06 February 2025 17h11

thanks so much Dieter. Of all my GPTs, and I have almost 100, I use this one the most.

→ View original post on X — @thatroblennon

6 February 2025
Agentic Object Detection: AI Vision Without Training Data

By

@andrewyng

–

06 February 2025 17h07

Introducing Agentic Object Detection!

Given a text prompt like “unripe strawberries” or “Kellogg’s branded cereal” and an image, we use an agentic workflow to reason at length and detect the specified objects. No need to label any training data. Watch the video for details. pic.twitter.com/ceNMx92pEu
— Andrew Ng (@AndrewYNg) 6 février 2025

Introducing Agentic Object Detection! Given a text prompt like “unripe strawberries” or “Kellogg’s branded cereal” and an image, we use an agentic workflow to reason at length and detect the specified objects. No need to label any training data. Watch the video for details.

→ View original post on X — @andrewyng

6 February 2025

←Previous Page

1 … 601 602 603 604 605 … 1,161

INNOVATION GENERATIVE AI RESEARCH LLMS TOOLS MACHINE LEARNING CODE MARKET TRENDS TECHNOLOGY BUSINESS BIG TECH ETHICS ENTERPRISE AI SOFTWARE AGENTS AUTOMATION APPS COMPUTING DATA POLICY OPEN SOURCE MULTIMODAL AI REGULATION CULTURE CREATIVE AI PROMPT ENGINEERING SOCIETY ECONOMY SAFETY EDUCATION INVESTMENT AI HARDWARE AGI HARDWARE JOBS STARTUPS INDUSTRY ROBOTICS WORKFORCE SECURITY CYBERSECURITY HEALTHCARE AI SYSTEMS SUSTAINABILITY WEB3 DECENTRALIZED AI

AI Dynamics

Global AI News Aggregator

About
Archives
Contact

Rechercher