Research Stage

I am done with learning…I am not done. I still think learning new topics is important but I have to focus on something else than learning and ideas. and that is being rigorous on a topic. Read, write and think about a topic rigorously and ask questions. I remember that paper on latent disentanglement by rishabh parihar. It was so rigorous. It feels like he just did and did experiments.

Instead of being perfect in work, I have to be doing more and more work instead of getting ideas.

I think this is the most important answers from chatgpt for a newbie as a researcher.

Question :

I am in 2nd year of phd I have learnt many things like how to read a paper how to find problems pattern recognition and learning new concepts….but i am still not getting confidence there are two thing i want to ask……first how the journey of other phd student like what are the steps in a phd journey where iam in those steps….and how can i improve myself in research like what shouldI do next becasue i have learnt that i cant rely on ideas all the time in research

Answer :

This is such a thoughtful question 👏 — it shows you’re reflecting critically, which itself is a sign of growth as a researcher. Let me break this down into two parts:

---

1. PhD Journey – Common Stages

Every PhD is unique, but many students go through broadly similar phases:

1. Orientation / Survival (Year 1)

   • Learning how to read papers efficiently.

   • Understanding your field and the “big problems.”

   • Feeling lost is normal — everything feels too broad.

2. Consolidation (Year 2) ← you are here

   • You’ve learned methods, tools, how to scan papers, identify gaps.

   • Now comes the hardest transition: from being a learner to a creator.

   • Confidence dips here because you realize: “I know enough to see how much I don’t know.”

   • Most students underestimate themselves at this point.

3. Direction / First Contributions (Years 2–3)

   • You carve out a niche problem.

   • You write your first serious paper(s).

   • This is where persistence beats inspiration — not every idea works, but persistence yields results.

4. Expansion (Years 3–4)

   • You have a research identity.

   • You branch out — combine methods, extend theories, collaborate.

5. Wrapping Up (Final Year)

   • Tie together your thesis story.

   • Publish your strongest contributions.

   • Think about your “next chapter” (postdoc, industry, etc.).

---

2. How to Improve Yourself in Research (Next Steps for You)

Since you’re in Year 2, here’s what matters most now:

🔎 A. Shift from “learning” → “doing”

• You don’t need to know everything to start producing.

• Pick one well-scoped problem (even if small!) and aim to solve it with rigor.

• Example: Instead of “robust audio-visual learning,” narrow to “how audio helps recover corrupted video features in sign language recognition.”

📚 B. Use papers differently

• Early on: you read to understand.

• Now: read to position your work.

  • Where exactly is the gap?

  • Who would care about the answer?

🛠️ C. Build habits of iteration

• Research isn’t about having brilliant ideas all the time.

• It’s about iterating small ideas, running experiments, and refining.

  • 1% improvements compound.

  • Even “failed” experiments can become a section of your paper (“we tried X, it failed, so we did Y”).

💡 D. Work on confidence actively

• Confidence doesn’t come before results — it comes from small results.

• Set mini-milestones:

  • Reproduce a paper.

  • Run a baseline on your dataset.

  • Write a short workshop paper (lower pressure than a top-tier conference).

🤝 E. Don’t do it alone

• Talk with your advisor, peers, even on forums.

• Explaining your work to others is the fastest way to clarify your own thinking.

---

3. Ideas vs Process

You’re right — you cannot rely on “grand ideas” all the time.

• Good research is 90% persistence, 10% inspiration.

• The process:

  1. Start with a small, concrete question.

  2. Test it rigorously.

  3. Share it, get feedback, refine.

That process itself generates new ideas — one solved problem reveals another.

---

✅ So where are you?

• You’ve already mastered reading, recognizing patterns, and learning concepts.

• You’re at the transitional step: from consuming knowledge to producing small, testable contributions.

• The next big leap is: pick one narrow problem and start building + writing.

Tags: newbie, researcher's life, learning, doing