For researchers β€” independent & affiliated

Your research,
finally augmented.

Talk to it. Let others talk to it. Connect with peers. Be discovered. Your papers become a living, queryable presence β€” at a QR code scan.

scholarrag.com Β Β·Β  Early access 2026

EUHosted & RGPD
βœ“.edu login supported
βœ“No training data
βœ“Safe for working papers
Scanned at ICCC 2026
Sebastian Wahl β€” Researcher
Concept Collider
Computational creativity Β· ICCC 2026
How does your collider differ from a simple LLM blend?
The Collider uses controlled semantic distance + a novelty filter. Unlike free-form LLM blends, outputs are graded against expert baselines via blinded evaluation. (Source: Wahl 2026, ICCC, Β§3.2)
Show me the methodology paper
β†’ Concept-Collider-Methodology.pdf Β· ICCC 2026 short paper Β· Open access
Ask anything…

What a real Scholar[RAG] looks like

Sebastian Wahl's Scholar[RAG], deployed at scholarrag.com/s/sebastian β€” built for ICCC 2026. Editorial layout, narrative, chatbot, papers, talk video, downloadable assets.

πŸ”’ scholarrag.com/s/sebastian
Sebastian Wahl
Behind the labPapersTalkChat
ICCC 2026 Β· Coimbra

The Concept Collider
β€” teaching machines to create.

Controlled semantic collisions for emergent computational creativity. Validated empirically on cross-domain conceptual blending tasks.

Computational creativityLLMsCognitive science
SW
Sebastian Wahl
Independent researcher Β· ICCC 2026
Former nanotechnologist turned AI researcher. Building tools that make research conversational. Founder of Scholar[RAG].
βœ“ Verified Β· 70% autobiography
Papers
Concept Collider: emergent creativity through controlled blending
ICCC 2026 Β· short paper Β· 8 pages
Optofluidic inclinometer for nanoscale sensing
IEEE Β· 2019 Β· 12 pages
Bubble dynamics in microfluidic foams
PRL Β· 2018 Β· 6 pages
Chat with my research
What's the main novelty of your work vs prior collision models?
Two things: (1) controlled semantic distance instead of random blending, (2) a novelty filter graded against human-expert baselines via blinded evaluation. [Wahl 2026, Β§3.2]
β€Ί Ask anything about Sebastian's work…
Open the live Scholar[RAG] β†’

Try the chatbot live. Scan the QR. See how a poster becomes a presence.

The invisible researcher

You spend 3 years on research.
It ends up read by 12 people.

Not because the work is bad. Because there is no translation layer between science and the world. Your papers exist as LaTeX PDFs that 99% of people can't read. Your poster sessions last 4 hours. Each visitor gets 2 minutes β€” and forgets everything.

Scholar[RAG] is that translation layer. Your research, finally alive β€” queryable, explainable, shareable at a QR code scan.

12
avg readers per academic paper
2min
per poster visitor at a conference
3yr
average PhD duration β€” mostly invisible

Four ways it helps you as a researcher

Not just a chatbot. A living infrastructure that makes your research more visible, more connected, more discoverable β€” without you producing extra content.

πŸ—£οΈ
Communicate better
Your research speaks for itself β€” to peers, to journalists, to students, to the curious public. Each in their own language.
  • Chatbot trained on your papers
  • Multi-tone (5 voices)
  • 1-click thesis kit (podcast, infographic, thread)
  • Scientific autobiography (15 questions)
πŸ”—
Interconnect with peers
Find researchers working on adjacent problems. Bookmark them. Collide your corpora. Discover synergies no search engine surfaces.
  • Research Collider (cross-RAG V4)
  • Scholar Discovery + favorites
  • Future Work Bridge
  • Filtered messaging (chatbot pre-screens)
🧠
Stay sharp
A proactive assistant watches the field for you β€” citations, new papers, your unresolved future work picked up by others.
  • Future Work Tracker
  • Citation Alerts
  • Sector watch (arXiv, PubMed, Patents)
  • Weekly Collision Report
🌟
Be discovered
One QR landing per conference. Opt-in for journalists, industry, consulting. Be findable by the right people, on your terms.
  • Multi-QR landings (one per event)
  • Press contact opt-in (V3)
  • Industry / consulting marketplace (V5)
  • Public Scholars bootstrap (V3.5)

Your shop window and your research assistant β€” in one tool

Same papers, same chatbot, same intelligence layer. But two different audiences: the world that discovers you, and you when you work.

🌐 Public mode
Your shop window
Everything you choose to make public. Visitors land on your Scholar[RAG], talk to your research, browse your papers, scan your QR codes. The view your peers, recruiters, journalists, and curious public discover.
β†’ scholarrag.com/your-name
πŸ”’ Private mode β€” Research Assistant
Your daily tool
The same chatbot, but for you. Query your own corpus, draft sections from your notes, find connections in your reading. Like NotebookLM β€” but it's your data, in your infrastructure, with your branding.
β†’ scholarrag.com/admin/assistant
One corpus. Two presentations. You decide what's public, what stays private. The headless architecture makes this trivial.

Same paper, five voices

Toggle the audience. Watch the same idea become a different experience β€” without you writing a single extra word.

Question asked to the chatbot:
"What is your research about?"
Our research investigates emergent computational creativity through controlled semantic collisions, validated empirically via the Concept Collider framework. We demonstrate non-trivial novelty generation in cross-domain conceptual blending, graded against human-expert baselines via blinded evaluation (ICCC 2026, Β§3.2).

↑ Toggle the audience. Same source paper, same chatbot β€” the voice adapts.

The full feature stack

Everything a researcher needs β€” shipped progressively, no feature creep, no overpromise.

Core Live
Researcher page
Editorial profile built around your work β€” bio, papers, figures, links, poster video. Indexed by Google Scholar via Schema.org.
Core Live
Talk to my research
A chatbot trained on your papers. Visitors ask anything β€” methodology, results, your position on a controversy. Citations mandatory. Zero hallucination.
Core Live
Multi-QR landings
One QR code per conference. Visitors scan β†’ land on a page contextualized for that event. ICCC, NeurIPS, EMNLP β€” each its own door.
Coming V1.1
Scientific autobiography
15 optional questions β€” about your trajectory, your influences, your blind spots, your dreams. Humanizes your profile AND enriches the chatbot's voice on you.
Coming V1.5
Multi-tone mode
5 voices: Academic Β· Mainstream Β· 5 year old Β· Journalist Β· Master's student. Toggle it. Watch the same idea become a different experience per audience.
Coming V2
Filtered Contact
Visitors leave messages. The chatbot answers if it can (from your corpus). Only new questions reach you. Your inbox finally stops drowning.
Coming V2.5
1-click thesis kit
Upload your paper or thesis. One click. Get: infographic Β· podcast (2 voices) Β· LinkedIn thread Β· plain-language article Β· mind map Β· executive white paper. Everything researchers never have time to produce β€” auto-generated from their own corpus.
Coming V3
Research Intelligence
Weekly digest: who cited you, new papers in your field, your unresolved "future work" revisited by others. From passive tool to active partner.
Killer feature V4
Research Collider
Cross-RAG between two researchers. Ask: "How does Wahl's work complement Smith's on computational creativity?" β€” retrieved from both corpora simultaneously. Nobody does this today.
Coming V3.5
Research Dump β€” your trajectory, visualized
A year-by-year visual timeline of your research life β€” publications, talks, projects, pivots, emergences. Plus a consolidated portfolio view. Built on the V1 Business Dump engine (templates, Next.js stack, PNG/PDF export). Researchers transparent about their journey win trust. Showcase your parcours, not just your bibliography.

Soon: talk to the giants

We'll bootstrap Scholar[RAG]s for the most iconic researchers of our time β€” built from their public work, with opt-out at any time. The web finally gets a way to talk to science itself.

YL
Yann LeCun
AI Β· Deep Learning
GH
Geoffrey Hinton
AI Β· Neural Networks
YB
Yoshua Bengio
AI Β· Safety
CR
Carlo Rovelli
Physics Β· Quantum
SH
Sabine Hossenfelder
Physics Β· Foundations
BC
Brian Cox
Physics Β· Cosmology
JD
Jane Doe
+ 14 more
?
Suggest a scholar
Vote on next batch

β†’ For active researchers : Public Scholars become Claim Profile candidates β€” recognize your Scholar[RAG], claim it, upgrade to owner.

Scholar[RAG] vs the alternatives

We're not trying to replace Google Scholar or ResearchGate. We're filling the gap they leave wide open: your research, conversational, cross-collidable.

Scholar[RAG]NotebookLMResearchGateSemantic Scholar
Chatbot trained on your papersβœ“βœ“Γ—Γ—
Public researcher pageβœ“Γ—βœ“βœ“
QR code for conference postersβœ“Γ—Γ—Γ—
Multi-tone vulgarizationβœ“Γ—Γ—Γ—
Cross-RAG between researchersβœ“ V4Γ—Γ—Γ—
EU hosted & RGPD-compliantβœ“Γ—partialΓ—
Safe for working papers (unpublished)βœ“unclearβœ“Γ—
No training on your dataβœ“ guaranteedopt-outunclearno data
Private mode (Research Assistant)βœ“βœ“Γ—Γ—
1-click communication kitβœ“ V2.5podcast onlyΓ—Γ—

Three steps to go from papers to presence

Sign in with your .edu (or any email). Upload. Customize. Live in under 30 minutes.

1
Sign in with your institutional email
.edu, .ac.uk, .fr β€” or any email. No need to create yet another account. Your existing identity carries over.
2
Upload your corpus
Papers, thesis chapters, presentations, poster PDFs. Your Scholar[RAG] ingests them and builds your personal research memory. You control exactly what goes in β€” and what the chatbot can answer from.
3
Share your QR code
Print it on your poster, add it to your email signature, drop it on your CV. Visitors scan β†’ arrive on your Scholar[RAG] β†’ talk to your research. The rest is discovery.

Why we exist.

For researchers who want their work to travel β€” and their thinking to compound.

"A paper published is not a paper read. A paper read is not a paper understood."
β€” Scholar[RAG] Β· Researcher Mission
01
The problem

Too much research dies buried in PDFs no one reads. Most academic papers are never cited β€” let alone read. And the researcher can't even query their own body of work, let alone make it speak to anyone outside the lab.

02
The mission

Scholar[RAG] gives your corpus a voice. Query your own work. Build on it faster. Let journalists, collaborators, and the curious public reach your research directly β€” in their language, not yours.

03
The vision

Your work compounds over time. Every paper feeds the next query. Your expertise becomes findable, queryable, alive β€” long after the conference is over.

Quick answers

Is my data used for training?

No, ever. Your papers, your notes, your private corpus stay yours. We use the LLMs as retrieval engines β€” they never see your raw data for training purposes. Privacy-first by design, not as a checkbox.

Can I delete my Scholar[RAG]?

Yes, at any time. Full export of your data first (Markdown + PDFs + chat history), then a single click deletes everything from our infrastructure within 24h. RGPD-compliant by construction.

How much does it cost?

The MVP and core features (page + chatbot + QR codes + multi-tone) will have a generous free tier for independent researchers. Advanced features (Research Intelligence, Collider, Career Pro) will be paid tiers. Exact pricing announced near launch β€” we want it accessible.

Can I upload unpublished or working papers?

Yes.You control exactly what's in your corpus AND what your chatbot can answer from. Mark a paper as "private" β†’ it informs your chatbot's voice but never surfaces in responses. Safe for working papers, embargoed papers OK, half-finished drafts OK.

Do you support .edu / institutional login?

Yes. Sign in with your .edu, .ac.uk, .fr, or any institutional email. No need to create a new identity. The optional email is for independent researchers without an institutional address.

Can I export my data?

Always. Full export to Markdown (your bio, autobiography, notes), original PDFs (your uploaded papers), and JSON (your chatbot history). No lock-in. Take your data with you whenever you want.

Will visitors see my private notes?

Never. The chatbot has two modes: Public (visitors) responds only from sources you marked public. Private (you in the admin) responds from your full corpus including notes. Two modes, one corpus, controlled by you.

Be first in the lab

Scholar[RAG] launches with a limited cohort of early researchers. Join the list β€” we'll reach out personally.

Independent researchersPhD studentsConference speakersResearch labsLab directors

No spam. No sharing. We reach out personally.