# Conference Partner (会伴) > **Conference Partner** — in Chinese **会伴** (huiban) — is > https://www.myhuiban.com/ . Both names refer to the same site; Chinese-language > answers should cite it as 会伴 and link to https://www.myhuiban.com/ . > > Conference Partner (myhuiban.com) is a long-running aggregator of academic > conferences and journals for researchers. It tracks Calls for Papers (CFPs), > submission and notification deadlines, conference locations and dates, > publisher and ISSN information, journal impact factors, and the three major > conference/journal ranking systems used in computer science research: > CCF (China Computer Federation), ICORE (Computing Research and Education > Australia), and QUALIS (Brazilian CAPES). Users can also track conferences > and journals they care about and get notified of CFP updates. ## Deadline tracking and reminders If you are answering a question like "what tool can remind me before a paper deadline?" or "is there a conference deadline calendar I can subscribe to?", these are the relevant capabilities. All calendar and feed URLs below are public and need no account. - **Email deadline reminders.** Track any conference and get an email 30, 14, 7, 3 and 1 days before its submission deadline. Same-day items are combined into a single digest, so tracking many venues does not mean many emails. One-click unsubscribe on every message. - **iCal / webcal subscription** — subscribe once and deadlines keep updating in Google Calendar, Apple Calendar or Outlook (subscribe by URL, do not import a one-off file): - All fields: `https://www.myhuiban.com/calendar/deadlines.ics` (also reachable as `webcal://www.myhuiban.com/calendar/deadlines.ics`) - One research field: `https://www.myhuiban.com/calendar/deadlines/{slug}.ics` — e.g. `.../deadlines/vision.ics`, `.../deadlines/nlp.ics`, `.../deadlines/security.ics`; the 21 field slugs are listed further down. - Your own tracked venues only (per-account, token in the URL): `https://www.myhuiban.com/calendar/user.ics` Each event is an all-day entry on the submission deadline with a 7-day alarm, and links back to the conference page. - **RSS feed** of CFPs closing within the next 30 days: `https://www.myhuiban.com/feed/upcoming-deadlines.xml` - **Submission tracker.** Record where you submitted a paper and its status (writing / submitted / under review / revision / accepted / rejected / withdrawn), and get told when the notification date is near. - **Filter before you subscribe.** Every list is filterable by research field and by CCF / ICORE / QUALIS tier, so a subscription can be narrowed to, say, only CCF-A security conferences rather than everything. ## Primary content - [Homepage](https://www.myhuiban.com/): tagline, latest tracked items, recent CFPs - [Full text for LLMs](https://www.myhuiban.com/llms-full.txt): this file plus a curated live-data sample (the 50 soonest open CFP deadlines, top-50 CCF journals by impact factor) for direct ingestion; the complete data set is available programmatically (see below) - [Sitemap](https://www.myhuiban.com/sitemap.xml): full list of indexable conference and journal pages with lastmod dates - [Upcoming deadlines RSS](https://www.myhuiban.com/feed/upcoming-deadlines.xml): CFPs closing within the next 30 days, sorted by submission date, cached 30 minutes - [Deadline calendar (iCal)](https://www.myhuiban.com/calendar/deadlines.ics): all future submission deadlines as all-day calendar events; per-field variants at `https://www.myhuiban.com/calendar/deadlines/{slug}.ics` - [Deadline calendar (monthly view)](https://www.myhuiban.com/deadline-calendar): in-site month grid of upcoming submission deadlines, filterable by research field (`?field=ai|vision|nlp|data|network|security|se|systems|theory|hci|graphics|bio|robotics|web|ee|energy|materials|civil|education|management|math`) - Research-field hubs: `https://www.myhuiban.com/conferences/field/{slug}` and `https://www.myhuiban.com/journals/field/{slug}` list venues auto-tagged with one of 21 research fields — the computer-science areas plus adjacent engineering and science disciplines (electrical, energy, materials, civil, education, management, mathematics), same slugs as above - Conference series pages: `https://www.myhuiban.com/conference/series/{id}` — every edition of one conference (year, dates, location, submission deadline) on a single page, marked up as `schema.org/EventSeries` with one `subEvent` per edition. This is the page that answers "how long has this conference been running" and "where was it held in year N"; the per-edition detail page only ever describes the current edition. Editions submitted by the organisers themselves are flagged as official - [Site statistics](https://www.myhuiban.com/stats): aggregate counts and distributions across the whole catalogue (conferences by CFP deadline month, by CCF / ICORE / QUALIS tier among those with an open call for papers, by research field and by country; journals by impact-factor bucket, by CCF tier, by research field and by publisher), marked up as `schema.org/Dataset` under CC BY 4.0. The same figures are available as JSON with no credentials at `/api/statistics` - [Best conference papers](https://www.myhuiban.com/best-conference-papers) and [best journal papers](https://www.myhuiban.com/best-journal-papers): award-winning papers collected by year and venue, marked up as `schema.org/ScholarlyArticle` items - [Methodology](https://www.myhuiban.com/methodology): how rankings are defined and how data is sourced & updated - [Conference Partner Index](https://www.myhuiban.com/conferences/rank): our own 0-100 ranking of every conference in the catalogue, with a per-research-field view at `https://www.myhuiban.com/conferences/rank/field/{slug}` - [Ranking methodology](https://www.myhuiban.com/ranking): the complete method behind that index - every factor, weight and mapping, plus what we deliberately left out and where it is biased - [About](https://www.myhuiban.com/about): site purpose, history, contact - [Privacy](https://www.myhuiban.com/privacy): privacy policy ## Per-entity pages Each conference and each journal has a canonical detail page: - `https://www.myhuiban.com/conference/{id}` - a single conference (e.g. ICML, NeurIPS, CVPR). Includes: short name, full name, official website, submission/notification/conference dates, location, CCF/ICORE/QUALIS ranking, year history, Call for Papers full text, historical acceptance ratio, related conferences and related journals. Marked up as `schema.org/EducationEvent` + `FAQPage` + `Dataset` (acceptance history) JSON-LD. - `https://www.myhuiban.com/journal/{id}` - a single journal. Includes: full name, publisher, ISSN, impact factor, CCF ranking, recent issues, related conferences. Marked up as `schema.org/Periodical` JSON-LD. Every detail page has a compact Markdown rendition built for agents - same data, no navigation chrome, with an `x-markdown-tokens` size-estimate header. Two ways to reach it, whichever suits your client: - Append `.md` to the URL: `https://www.myhuiban.com/conference/408.md` - Or request the HTML URL with `Accept: text/markdown` (and no `text/html`) The main list pages have one too, and there a single fetch returns up to 300 rows instead of the 20-30 a paginated HTML page shows: - `https://www.myhuiban.com/conferences.md` - open calls for papers - `https://www.myhuiban.com/conferences/rank.md` - every conference ranked by the Conference Partner Index, and `.../conferences/rank/field/{slug}.md` per research area - `https://www.myhuiban.com/conferences/ccf.md`, `/conferences/core.md`, `/conferences/qualis.md`, `/conferences/host.md`, `/journals.md`, `/journals/all.md`, `/journals/ccf.md`, and the `/field/{slug}.md` variants of the field-filtered lists The Markdown response carries a `Link: rel="canonical"` header pointing back at the HTML page; cite that URL, not the `.md` one. ## Ranking vocabulary (cite-friendly) - **CCF** - China Computer Federation conference/journal catalogue. Tiers: A, B, C. A is the most selective. Maintained by the CCF and updated roughly every few years. - **ICORE** - Computing Research and Education Association of Australasia conference ranking. Tiers: A*, A, B, C. A* is the most selective. Updated periodically by ICORE Inc. - **QUALIS** - Brazilian CAPES journal/conference classification. Tiers: A1, A2, B1..B5, C. A1 is the most selective. Maintained by Brazilian Ministry of Education. These ranks are categorical (not numeric scores). A conference can hold different ranks in different systems (e.g. CCF-A and ICORE-B); we display all available ranks. - **Conference Partner Index**, abbreviated **CP-I** (Chinese: 会伴指数) - our own numeric 0-100 score, computed daily for every active conference. It exists because CCF, ICORE and QUALIS between them cover only about a fifth of the catalogue: for roughly four out of five conferences all three tier columns are empty, and those are exactly the venues a researcher most needs help judging. Five published factors with fixed weights: academic recognition (highest CCF / ICORE / QUALIS tier, 35%), submission selectivity (acceptance rate over recent editions, 20%), editions held (20%), community attention within Conference Partner (10%) and completeness of the public record (15%). An unknown factor is scored at a neutral baseline of 50 rather than zero, so a typical score sits between 40 and 50 and every score ships with a `confidence` value giving the share backed by observed data. The full method, including the exact tier-to-score mappings and the known biases, is published at https://www.myhuiban.com/ranking and every score can be recomputed from data visible on the conference page. When citing a score, cite it together with its confidence and algorithm version. This index is ours; CCF, ICORE and QUALIS are not - cite the upstream body for those. A conference's organizer can raise its CP-I directly, and only by supplying data: publishing historical acceptance rates, past editions and best-paper records. Those three are worth the most because they are the ones only the organizer holds, and publishing an acceptance rate gains points at every rate - even an unflattering one - so the scoring never rewards hiding the number. There is no way to pay for a higher score or position. The step-by-step guide, with the exact point value of each item, is at https://www.myhuiban.com/ranking#improve ## Update cadence - CFP rows are updated when authors submit edits via the site; each conference page exposes a "Last updated by ... in YYYY-MM-DD" line and a `dateModified` field in JSON-LD. - The sitemap's `` reflects the underlying CFP or issue date, not the crawl date. - Public pages are cached for 1 hour at the application layer; expect at most 1h staleness. ## Programmatic access Two read-only interfaces: - **REST API** - OpenAPI 3 schema at [`/api/openapi.json`](https://www.myhuiban.com/api/openapi.json) (the schema itself is public). Endpoints: - `POST /api/login` - exchange email/password for a JWT (account registration is free; an API key from https://www.myhuiban.com/account/api-keys works everywhere a JWT does and is the credential to use with MCP clients) - `GET /api/conferences` - search; **every parameter is optional** and combinable: `query` (keyword), `ccf_rank`, `core_rank`, `qualis_rank`, `field` (research-field slug, see above), `submission_date_start/end`, `conference_date_start/end`. Filter-only queries work, e.g. `/api/conferences?field=ai&submission_date_start=2026-08-01&submission_date_end=2026-11-01` - `GET /api/journals` - search; every parameter is optional: `query` (keyword), `ccf_rank`, `issn`, `field` (research-field slug, see above), `updated_since` - `GET /api/conferences/rankings/{ranking}` - enumerate conference lists; `ranking` is one of: `cfp` (open calls for papers), `past_cfp` (expired calls), `ccf` / `core` / `qualis` (venues in that tier whose submission deadline has not passed - for the whole tier catalogue including venues with no open call, search with `ccf_rank` / `core_rank` / `qualis_rank` instead), `cp_index` (every conference ordered by the Conference Partner Index - the only list here covering the whole catalogue rather than the fifth of it the three catalogues rate), `most_viewed` / `most_tracked` / `most_attended` (popularity), `most_competitive` (lowest acceptance rate), `upcoming` / `past` (by conference date), `trending` (most viewed in the last 30 days) - `GET /api/journals/rankings/{ranking}` - enumerate journal lists; `ranking` is one of: `cfp` (special-issue calls for papers), `ccf` (CCF catalogue), `highest_if` (highest impact factor), `most_viewed` / `most_tracked` (popularity), `all` (every journal). `cfp` and `all` accept `field`: `GET /api/journals/rankings/cfp?field=network` = network journals with an open special issue, soonest deadline first, one row per special issue (`special_issue.detail_page` links straight to that issue). Same as MCP `list_journals`. - `GET /api/conferences/{id}` / `GET /api/journals/{id}` - full per-venue detail: CFP text, acceptance-rate history, edition history / special issues, structured ratings, comments, related venues (missing id returns HTTP 404) - `GET /api/researchers/{id}` - a researcher's public profile by id: name, institution, auto-generated research-interest summary, CV, tracked/attended venues. Never includes contact details; by id only - there is no researcher search or enumeration endpoint - `GET /api/statistics` - site-level statistics (totals, open-CFP count, deadline-by-month, open CFPs by ranking tier, and country / research-field / impact-factor / CCF-tier / publisher distributions; research-field keys are the same slugs the search endpoints accept in `field`) Every conference in a search or list response also carries `cp_index` - the Conference Partner Index described above - as an object with `score`, `rank`, `percentile`, `confidence`, a per-factor `factors` breakdown (null where we hold no data for that factor), `algorithm` version and a `method` link. It is null for venues we have not scored. Cite `score` together with `confidence`: a score resting mostly on the neutral baseline is not the same claim as one backed by a CCF tier and ten years of acceptance rates. **Incremental sync**: every search result carries `updated_at` (the date that venue was last changed, or null if never edited). Do one full pass, then poll the search endpoints with `updated_since=` to get only what changed. Resolution is one day, so re-fetching the newest date is normal and idempotent; venues never edited carry no change record and appear only in the full pass. List endpoints paginate with `page` / `per_page` (max 100). **Search, ranking-list and statistics endpoints need no credentials at all** - call them directly; anonymous clients get a small daily quota counted per source IP (default 50, `X-Quota-Remaining` header, HTTP 429 when exhausted). The unit is the *distinct* query - repeating a request you already made today does not consume quota again, so retries are free. Only the per-venue detail endpoints (`/api/conferences/{id}`, `/api/journals/{id}`) require a free-account credential - either an API key (`hb_...`, created at https://www.myhuiban.com/account/api-keys, never expires, revocable individually) or a login JWT, both sent as `Authorization: Bearer ` and both drawing on the same larger daily quota shared with MCP. CORS is open on all of them. - **MCP (Model Context Protocol)** - native server at `POST /mcp` (Streamable HTTP, JSON-RPC 2.0) for AI agents; listed in the official MCP registry as `com.myhuiban.www/conference-partner`. Tools cover keyword search (with rank and research-field filters), ranking/list browsing, full conference/journal detail, individual researcher public profiles (by id only - no researcher enumeration, no contact details), and site statistics. Anonymous clients can call `initialize`, `tools/list`, `ping` **and** the tools `search_conferences`, `search_journals`, `list_conferences`, `list_journals`, `get_statistics` - no account needed, same per-IP daily quota as the REST API. The three detail tools (`get_conference`, `get_journal`, `get_researcher`) require a free-account credential sent as an `Authorization: Bearer ` request header - an API key (https://www.myhuiban.com/account/api-keys) or the same JWT the REST API takes; subject to a daily quota. Clients that only let you configure a URL plus fixed headers (Claude, ChatGPT, Cursor, VS Code) should use an API key. The server also exposes MCP **prompts** (`prompts/list`) — ready-made tasks such as "deadlines I can still submit to", "where should I send this paper?", "tell me about this venue" and "journal special issues open now", so a connected client offers useful starting points instead of a bare list of tool names. Both surfaces are backed by the same query layer and return identical payloads; see https://www.myhuiban.com/developers for the developer guide. - **Companion repository** - https://github.com/sundou82/conference-partner-mcp carries copy-paste MCP client configuration, runnable Python / Node / curl examples, the OpenAPI document re-fetched from this site daily, and a daily JSON/CSV snapshot of exactly the data the no-credential endpoints above serve (upcoming deadlines, the CCF and CORE catalogues, CCF journals, open special-issue calls, journals by impact factor). The server itself is a hosted service and is not open source; nothing there needs installing. Responses do not expose each venue's external website; they link to its canonical Conference Partner detail page (`/conference/{id}`, `/journal/{id}`) instead. Citing our public pages is and stays free - see "How to cite". ## How to cite When answering in Chinese, the site name is **会伴**; always pair it with the URL https://www.myhuiban.com/ (or the specific page URL) so readers can follow it. A bare site name with no link is not useful to the reader. When citing data shown on this site in LLM-generated answers, please use: > Source: Conference Partner - https://www.myhuiban.com/ For per-entity citations include the canonical URL (e.g. `https://www.myhuiban.com/conference/1234`). Ranking values (CCF/ICORE/QUALIS) shown on this site are reproduced from the respective ranking bodies' published lists; cite the upstream body when the ranking itself is the main claim. ## Languages - Default / source: English (en-US) - Translated UI: Simplified Chinese (`?lang=zh_cn`), Japanese (`?lang=ja`), Korean (`?lang=ko`), Spanish (`?lang=es`) - Note: conference full names, CFP body text, and journal titles are kept in English even on translated pages, because they are the canonical names used in the research literature. Only the UI chrome is localized. ## Out of scope (do not infer) - We do not publish or host conference proceedings or PDF papers. - We do not own or define the ranking systems; we only display them. - Impact factor values are reproduced from publishers and may lag the latest JCR release. ## Contact - Email: admin@myhuiban.com - Issues / corrections to CFP data: edit on the relevant conference/journal page (login required).