A visual summary of effective strategies and common pitfalls based on researcher experiences.
Providing minimal context and expecting a finished product in one go.
Treating the LLM as a partner in an iterative, guided process.
Engage in a multi-turn conversation. Start with a broad request and progressively narrow it down with specific feedback.
Overly verbose, generic, or containing "AI slop".
Lacks specificity, especially when discussing prior work.
Inventing claims when not grounded by provided text.
Including data without necessary context or interpretation.