Sistemas de feedback com participação humana
Projete e implemente sistemas que integrem feedback humano para melhorar continuamente o desempenho dos LLMs e corrigir erros.
Sistemas de feedback com participação humana é uma aula grátis de Prompt Engineering & LLM Optimization for Developers no CoddyKit. Esta é a aula 2 de 4. Você pode ler a aula completa abaixo gratuitamente — depois pratica ao vivo no navegador com um editor de código integrado e um tutor de IA 24/7. Faz parte do caminho de aprendizado de Prompt Engineering & LLM Optimization for Developers, e seu progresso é sincronizado entre a web e o app CoddyKit. O curso de Prompt Engineering & LLM Optimization for Developers inclui 4 aulas no total.
Partes desta aula ainda não foram traduzidas e aparecem em inglês.
What is Human-in-the-Loop?
When working with Large Language Models (LLMs), sometimes their outputs aren't quite right. This is where Human-in-the-Loop (HITL) systems come in.
HITL means humans actively review, correct, or provide feedback on an LLM's output. This feedback then helps the model learn and improve over time, making it smarter and more reliable.
Why Humans Help LLMs
Even powerful LLMs have limitations. They can:
- Hallucinate: Make up facts.
- Be biased: Reflect biases from their training data.
- Lack up-to-date info: Not know about recent events.
- Misunderstand complex requests: Struggle with nuanced instructions.
Human oversight helps catch and fix these issues, ensuring higher quality and safer outputs.
How HITL Systems Work
A typical HITL system involves a few key parts:
- LLM Generation: The model creates an output.
- Human Review: A person evaluates the output.
- Feedback Collection: The human's judgment is captured.
- Model Improvement: This feedback is used to refine the LLM, either by adjusting prompts or fine-tuning the model itself.
It's a continuous cycle of creation, evaluation, and learning.
Two Ways to Give Feedback
Human feedback can be broadly categorized into two types:
- Explicit Feedback: Direct ratings or corrections from users or annotators. Think "thumbs up/down" or editing a generated text.
- Implicit Feedback: Inferred from user behavior without direct input. For example, how long a user spends on a response or if they click on certain links.
Both types are valuable for different reasons!
Getting Direct Feedback
Explicit feedback is clear and intentional. Common ways to collect it include:
- Thumbs Up/Down: Simple binary feedback on response quality.
- Star Ratings: A more granular scale (e.g., 1-5 stars).
- Direct Edits: Users correcting or rewriting parts of the LLM's output.
- Annotator Labels: Expert human annotators tagging specific issues like factual errors or toxicity.
This data directly tells you what worked and what didn't.
Learning from User Behavior
Implicit feedback is less direct but still powerful. It's about observing how users interact with LLM outputs:
- Session Duration: Users spending more time on a response might indicate engagement.
- Click-Through Rates: If an LLM suggests links, clicks show relevance.
- Search Refinements: If a user immediately rephrases their query, the initial response might have been poor.
- Copying Text: Users copying output might find it useful.
Analyzing these behaviors can reveal subtle preferences and issues.
Capturing Feedback in Code
Let's imagine a basic way to capture explicit feedback (like a "thumbs up" or "thumbs down") after an LLM generates a response. This simple Python code shows how you might structure the collection of feedback.
It doesn't call an LLM, but focuses on the feedback part.
def collect_feedback(llm_output: str, user_rating: str):
"""
Simulates collecting feedback for an LLM output.
In a real system, this would store to a database.
"""
feedback_data = {
"output": llm_output,
"rating": user_rating, # e.g., "good", "bad", "neutral"
"timestamp": "2023-10-27T10:30:00Z" # For simplicity
}
print(f"Feedback recorded: {feedback_data}")
# In a real app, you'd save this to a database or log file.
if __name__ == "__main__":
generated_text = "The capital of France is Paris."
print(f"LLM output: '{generated_text}'")
# Simulate user giving feedback
user_choice = input("Was this output good or bad? (good/bad): ")
collect_feedback(generated_text, user_choice)
generated_text_2 = "The sun revolves around the Earth."
print(f"\nLLM output: '{generated_text_2}'")
user_choice_2 = input("Was this output good or bad? (good/bad): ")
collect_feedback(generated_text_2, user_choice_2)Turning Feedback into Growth
Once feedback is collected, it's not just stored; it's used to make the LLM better:
- Prompt Refinement: If many users rate an output as "bad" for a specific prompt, you can adjust the prompt to be clearer or more specific.
- Data for Fine-tuning: High-quality human-corrected data can be used to fine-tune the LLM, teaching it directly from good examples.
- Reinforcement Learning from Human Feedback (RLHF): Feedback can be used to train a reward model, which then guides the LLM to generate preferred outputs.
This closes the loop and improves performance.
Hurdles in Human-in-the-Loop
While HITL is powerful, it comes with challenges:
- Scalability: Manually reviewing every LLM output for a large application is impossible.
- Cost: Human annotators or user feedback mechanisms can be expensive to build and maintain.
- Bias in Feedback: Human reviewers can also introduce their own biases, which might then be reinforced in the model.
- Subjectivity: What one person considers a "good" response, another might not.
Careful design is needed to overcome these.
Understanding HITL
You've learned about Human-in-the-Loop systems. Let's test your understanding!
Recap: HITL for Better LLMs
In this lesson, we explored Human-in-the-Loop (HITL) systems. We learned that HITL integrates human judgment to continuously improve LLM performance by addressing limitations like hallucinations and biases.
You now understand the components of a HITL system, the difference between explicit and implicit feedback, and how this feedback drives model improvement. You also saw a simple code example for collecting feedback and are aware of the challenges involved.
Keep exploring how to build robust LLM applications!
Perguntas Frequentes
A aula “Sistemas de feedback com participação humana” é grátis?
Sim — o texto completo de “Sistemas de feedback com participação humana” é grátis para ler aqui na web. Para praticá-la interativamente (um editor de código integrado e um tutor de IA 24/7) e desbloquear o restante do curso de Prompt Engineering & LLM Optimization for Developers, atualize para CoddyKit PRO. O curso de Prompt Engineering & LLM Optimization for Developers inclui 4 aulas no total.
O que vou aprender em “Sistemas de feedback com participação humana”?
Projete e implemente sistemas que integrem feedback humano para melhorar continuamente o desempenho dos LLMs e corrigir erros. Você pratica Prompt Engineering & LLM Optimization for Developers com código prático que executa diretamente no navegador, e um tutor de IA 24/7 responde suas dúvidas enquanto trabalha na aula.
Preciso ter experiência prévia para começar Prompt Engineering & LLM Optimization for Developers?
Nenhuma experiência prévia é necessária. Prompt Engineering & LLM Optimization for Developers no CoddyKit é estruturado para alunos iniciantes até avançados, então você pode começar aqui ou desde o início e aprender no seu ritmo. Esta é a aula 2 de 4.
Quanto tempo leva a aula “Sistemas de feedback com participação humana”?
A maioria das aulas CoddyKit leva cerca de 5–10 minutos. Cada uma é compacta e interativa, então você faz progresso constante e retoma exatamente de onde parou entre web e app.
Posso escrever e executar código nesta aula de Prompt Engineering & LLM Optimization for Developers?
Sim. Cada aula de Prompt Engineering & LLM Optimization for Developers inclui um editor de código integrado, então você escreve e executa código real direto no navegador e recebe feedback de IA instantaneamente — nenhuma configuração local necessária.
Todas as aulas deste curso
- Métricas e referências de avaliação de LLMs
- Sistemas de feedback com participação humana
- Injeção de prompts e boas práticas de segurança
- Detectando e Mitigando Alucinações