Chapter 7: Where to Find the Literature¶
“You can’t build on prior work if you never find the right work. Research starts with search.”
Why This Chapter Matters¶
One of the most common questions new researchers ask is:
“Where do I even find the papers I’m supposed to read?”
Sure, your advisor might throw a few PDFs your way. But building a strong thesis—or publishing your own work—requires a wider net.
You need to know where to look, what to trust, and how to dig deeper.
This chapter will give you a toolbox of platforms, search engines, and research discovery tools so you can find credible, relevant, and up-to-date work—without getting lost in a swamp of random PDFs.
Conceptual Breakdown¶
🔹 Academic Search Engines vs. Libraries
Let’s distinguish two types of tools:
| Type | Examples | Purpose |
|---|---|---|
| Search Engines | Google Scholar, Semantic Scholar | Broad discovery, citation chaining |
| Digital Libraries | ACM DL, IEEE Xplore, SpringerLink | Authoritative source, full-text access |
Search engines cast a wide net. Libraries give you official, citable copies.
🔹 Platforms You Should Know
| Platform | Use Case |
|---|---|
| Google Scholar | Fastest way to find papers + citation counts + related works |
| Semantic Scholar | AI-powered highlights, TLDRs, and citation context |
| DBLP | Great for structured CS publication lists and author profiles |
| IEEE Xplore | Best for CS/EE fields (AI, vision, robotics, signal processing) |
| ACM Digital Library | Excellent for HCI, software engineering, systems research |
| arXiv | Open preprints (especially for ML/AI/NLP/vision), often before peer review |
🔎 Pro Tip: Always check if your school gives institutional access—this unlocks many paywalled papers via your university library.
🔹 Specialized Discovery Tools
Here are tools that make literature exploration smarter, not harder:
-
Connected Papers: Build a visual graph of related work from a single seed paper
→ https://connectedpapers.com -
Litmaps: Track related papers as they’re published, auto-organize by topic
→ https://www.litmaps.com -
ResearchRabbit: Create “rabbit holes” of paper relationships across time and topics
→ https://www.researchrabbit.ai
These are gold for building your literature review, seeing evolution of ideas, and finding missing gaps.
🔹 Which Platform Should I Use When?
| Goal | Best Tools |
|---|---|
| Get quick overview of a topic | Google Scholar, Semantic Scholar |
| Find the original source of a paper | DBLP, IEEE Xplore, ACM DL |
| Explore related works visually | Connected Papers, Litmaps, ResearchRabbit |
| Access cutting-edge work (pre-peer-review) | arXiv, Twitter threads from researchers |
| Track an author or lab | DBLP, Google Scholar profiles |
Self-Check Questions¶
- What platform do you default to when looking for papers? Why?
- Have you tried using any visual literature mapping tools?
- Do you know which platforms are most relevant to your field (e.g., NLP vs HCI)?
Try This Exercise¶
Build a Paper Discovery Toolkit
Choose one topic you’re interested in (e.g., “object detection for drones” or “LLMs for legal document analysis”).
Search that topic using the following tools and record what you find:
- Google Scholar
- Semantic Scholar
- Connected Papers
- IEEE Xplore or ACM DL
- DBLP
For each, ask:
- Which gave the most useful papers?
- Which showed citation counts?
- Which offered recent vs. older papers?
You’ll start to see patterns—and preferences.
Researcher’s Compass¶
Good research is built on good sources.
And good sources are only useful if you can actually find them.
Don’t rely on a single platform.
Instead, build your own search stack—a reliable set of tools that helps you:
- Discover new ideas
- Trace citation trails
- Collect trustworthy, peer-reviewed work
Research isn’t about knowing everything. It’s about knowing where to look—and how to follow the thread.