where research convenes
A fast, searchable home for accepted papers from top conferences and journals — browse the field, follow authors and institutions, and turn papers into insight.
English · 中文
Every conference and journal publishes its accepted papers on a different site, in a different format. confer gathers them into one place with a single, consistent interface — so you can scan a whole field in minutes instead of clicking through a dozen program pages.
It's a static website backed by a small scraping pipeline: each venue is normalized
to one shared Paper shape, then enriched from bibliographic/open metadata sources
for DOI, abstracts, publication details, and open-access links where available. No
accounts, no backend — your favorites and saved searches live in your browser.
- 🔎 Field-aware search. Search everything by default, then narrow with prefixes
like
author:,title:, orinst:and exclude terms with-. - 🏷 Authors & affiliations. Hover an author to see their institution; click it to pull up every paper from that institution. Same-name authors are disambiguated in analytics and networks.
- 📊 Insights panel. Live charts of the top institutions, authors, and tracks for whatever is currently in view — click any bar to drill in.
- 🕸 Relationship networks. Explore co-author and institution networks for the current result set.
- ⭐ Favorites & saved searches. Star papers across venues; save a filter set to return to later.
- 📤 Selection & export. Select papers from the list, then copy BibTeX or download CSV with DOI and publication metadata.
- ⚡ Fast & private. A single pre-rendered page; all filtering happens client-side.
Light/dark themes, keyboard shortcuts (
⌘K,⌘/), and a responsive mobile layout.
confer currently brings together conferences and journals across EDA, computer architecture, software engineering, testing, and programming languages, with more added purely through configuration. Browse them all from the category sidebar. The current data set includes multiple yearly editions for several venue series, grouped by area, series, and year.
config/venues.yaml ─▶ scraper + enrichers ─▶ unified JSON per venue ─▶ Astro site ─▶ static host
- Config lists venues, their primary scraper, and optional metadata enrichers.
- Adapters each understand one source platform and emit the same
Papershape. - Enrichers merge Crossref/OpenAlex metadata such as DOI, abstract, publication date, volume/issue/pages, keywords, author metadata, and open-access/PDF links.
- Site consumes only the unified data — adding a venue never touches the UI. It uses the enriched metadata for search, export, disambiguation, and network views.
See AGENTS.md for the architecture, the Paper schema, and the
adapter contract.
Build the data (Python, via uv):
cd scraper
uv run confer list # show configured venues
uv run confer build # build all enabled venues → web/public/data/
uv run confer build --venue <venue_id> # build a single venue
uv run confer build --refresh # ignore cache, refetch over the networkEach venue is cached under data/cache/<venue_id>/, so re-runs are offline unless you
pass --refresh.
Run the site (Astro, Node ≥ 18):
cd web
npm install
npm run dev # local dev server
npm run build # static build → web/dist/The site reads web/public/data/ at build time and outputs a static dist/ you can
host anywhere. The committed JSON is the build input, so a deploy only runs the Astro
build — no Python at deploy time.
- Add an entry to
config/venues.yaml(fields are documented inline). - Point its
scraper:at a registered adapter (dateconf,dblp,linklings,researchr,sigarch). - Add
source.enrichers(crossref,openalex) when the primary source lacks publication metadata. uv run confer build --venue <id>and checkweb/public/data/<id>.json.
To support a new platform, add an adapter under scraper/src/confer/scrapers/ and
register it — see AGENTS.md, "How to add a scraper adapter".