Accelerate Research Topic Selection with LLMs
Bollegala · x · 2026-07-14
The author shares how they recently used an LLM to help students determine research topics. What used to take two to three weeks can now be done in about a day, and the process is much more relaxed.
Their method roughly consists of four steps:
- Define the problem and direction yourself: True research taste, novelty, and the overarching direction shouldn't be directly handed over to the LLM.
- Share the problem, initial solutions, and concerns with the LLM: The input might be long and messy, but that's completely fine.
- Iterate and refine: The LLM suggests various approaches, while the author continuously points out things that don't fit their research style, are too complex, or missing. They frequently summarize and organize the discussion to keep it on track.
- Output an actionable plan for the student: Finally, have the LLM generate a PDF research plan detailing datasets, models, evaluation metrics, baselines, and what constitutes a successful result.
Ultimately, they share both the final plan and the entire chat history with the student, followed by a meeting to discuss it.
More from Research
- SUFLECA shows NOC-based correspondence can improve CAD-to-image alignment — ducha_aiki · 2026-07-21
- OpenAI-style autonomous researchers could become real scientific collaborators — Promptmethus · 2026-07-21
- Soft Clamp cuts tool-call overuse in multi-teacher distillation, from 13.7% to 9.0% — antgroup · 2026-07-21
- ShotPlan adds learnable planning tokens for cinematic multi-shot video generation — Tele-AI · 2026-07-21
- A silicon photonic reservoir chip compensates fiber distortion in real time at 28 Gbps — bravo_abad · 2026-07-21
- A developer maps out six design rules for CLIs that humans and AI agents can both use — yujiezha · 2026-07-21