How do ML researchers actually use AI tools to improve their writing? [D]

2026-06-04 · Source: Machine Learning · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering · Depth: Intermediate, short

Summary

ML researchers are actively using AI tools, particularly Large Language Models (LLMs), to enhance their writing and research workflows. Common applications include generating code for figures, which can save over 2 hours by avoiding documentation lookups, and structuring technical text. One workflow involves drafting outlines with tools like Claude, expanding thoughts, integrating data, forming paragraphs, and then using Claude to rewrite sentences for density and neatness. However, a significant debate exists regarding LLMs' impact on cognitive skills and originality. While some argue LLMs enhance abilities by offloading "brain dead" tasks, others cite research like "AI Assistance Reduces Persistence and Hurts Independent performance" and personal experiences of worsened writing and coding skills, emphasizing the importance of independent thought and manual work for original ideas. Practical uses also extend to line-by-line text analysis and literature review.

Key takeaway

If you are an ML researcher optimizing your writing workflow, strategically integrate LLMs for tasks like code generation for figures or refining sentence structure, but remain vigilant about potential cognitive skill degradation. Balance AI assistance with dedicated manual effort, especially for critical thinking, original idea generation, and developing your core writing and coding competencies. Consider using LLMs as an interactive analytical tool for specific sentence feedback rather than a full content generator to maintain intellectual ownership and skill development.

Key insights

LLMs offer efficiency in technical writing and coding but risk diminishing cognitive skills and originality if over-relied upon.

Principles

• LLMs can offload "brain dead" tasks like figure coding.
• Overuse of LLMs may worsen writing and coding skills.
• Effective LLM use requires strong prompt engineering skills.

Method

A workflow involves drafting outlines (e.g., with Claude), writing initial thoughts, enhancing with data, forming paragraphs, and using LLMs to refine sentence density and neatness.

In practice

• Generate code for figures (e.g., sophisticated TikZ).
• Perform line-by-line text analysis and correction.
• Assist with literature review to find specific papers.

Topics

• Large Language Models
• Technical Writing
• Research Workflow
• Code Generation
• Cognitive Load
• Prompt Engineering
• Literature Review

Best for: AI Scientist, Research Scientist

Related on AIssential

• Build a Powerful AI Research Pipeline with LM Studio and NotebookLM — Combining local LLMs with source-grounded knowledge tools enhances research and content creation workflows.
• The Crutch or the Ceiling? How Different Generations of LLMs Shape EFL Student Writings — Advanced LLMs boost surface fluency in EFL writing but may hinder deep learning and mask true student ability.
• Helping people write code again — LLM-assisted coding empowers experienced individuals to resume programming by reducing time barriers.
• Can AI Write Your Code? — Trusting AI-generated code for complex quantitative research requires rigorous validation against benchmarks and expert human supervision.
• Pokémon Built a Robot Brain — LLMs are trained in stages, from broad knowledge acquisition to fine-tuned conversational and reasoning capabilities.
• How To Use AI for the Ancient Art of Close Reading — LLMs can transform traditional close reading into a dynamic, personalized, and deeply engaging learning experience.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Machine Learning.