AUG182027

ICDAR 2027 — 21st International Conference on Document Analysis and Recognition

by ICDAR 2027 Organizing Committee

conferenceIn-PersonUpcomingStarts in 13 months

The flagship document AI conference, now built around multimodal LLMs

August 18–22, 2027·Kuala Lumpur, Malaysia·450 attendees·Paid
Share:
THE DAILY BRIEF
AUG182027
ICDAR 2027 — 21st International Conference on Document Analysis and Recognition

by ICDAR 2027 Organizing Committee

conferenceIn-Personupcoming

The flagship document AI conference, now built around multimodal LLMs

About This Event

ICDAR 2027, the 21st International Conference on Document Analysis and Recognition, runs 18-22 August 2027 in Kuala Lumpur, Malaysia. ICDAR is the field's flagship venue: the research presented here — layout analysis, handwriting and scene-text recognition, table and formula extraction, signature verification, document forensics — is the science underneath every intelligent document processing product an enterprise buys. The 2027 call for papers has shifted decisively toward generative AI, explicitly soliciting work on foundation models for document understanding, multimodal LLMs for documents, agents for document AI and document reasoning models, alongside the classical tracks on historical documents, medical documents and private and secure document understanding. The organising committee expects roughly 450 participants across keynote talks, main-track sessions, panel discussions, poster exhibitions and side workshops — small enough that the authors of the papers your vendors will cite are actually reachable in the hallway.

At a Glance

Date
August 18–22, 2027
Location
Kuala Lumpur, Malaysia
Format
In-Person
Event Type
Conference
Status
Upcoming
Expected Attendance
450
Pricing
Paid
Organizer
ICDAR 2027 Organizing Committee
Added
Aug 16, 2026
Updated
Aug 16, 2026

Topics & Focus Areas

Document AIFoundation models for document understandingMultimodal LLMs for documentsHandwriting and scene text recognitionTable and formula recognitionInformation extraction and document retrievalDocument forensics and provenancePrivate and secure document understanding

Who Should Attend

  • Chief Data Officers
  • Heads of AI/ML
  • Intelligent document processing product leaders
  • Computer vision and NLP researchers
  • ML engineers
  • Enterprise architects in insurance, banking and healthcare

Organizer

I

ICDAR 2027 Organizing Committee

icdar2027.org/

THE DAILY BRIEF

Enterprise AI insights for technology and business leaders, twice weekly.

beri.net

Subscribe at beri.net/subscribe for twice-weekly AI insights delivered to your inbox.

LinkedIn: linkedin.com/in/rberi  |  X: x.com/rajeshberi

© 2026 Rajesh Beri. All rights reserved.

About This Event

ICDAR 2027, the 21st International Conference on Document Analysis and Recognition, runs 18-22 August 2027 in Kuala Lumpur, Malaysia. ICDAR is the field's flagship venue: the research presented here — layout analysis, handwriting and scene-text recognition, table and formula extraction, signature verification, document forensics — is the science underneath every intelligent document processing product an enterprise buys. The 2027 call for papers has shifted decisively toward generative AI, explicitly soliciting work on foundation models for document understanding, multimodal LLMs for documents, agents for document AI and document reasoning models, alongside the classical tracks on historical documents, medical documents and private and secure document understanding. The organising committee expects roughly 450 participants across keynote talks, main-track sessions, panel discussions, poster exhibitions and side workshops — small enough that the authors of the papers your vendors will cite are actually reachable in the hallway.

Key Dates & Deadlines

  • Abstract deadline
  • Full paper deadline
  • Reviews released to authors
  • Rebuttal deadline
  • Acceptance notification
  • Camera-ready deadline

At a Glance

Date
August 18–22, 2027
Location
Kuala Lumpur, Malaysia
Format
In-Person
Event Type
Conference
Status
Upcoming
Expected Attendance
450
Pricing
Paid
Organizer
ICDAR 2027 Organizing Committee
Added
Aug 16, 2026
Updated
Aug 16, 2026

Topics & Focus Areas

Document AIFoundation models for document understandingMultimodal LLMs for documentsHandwriting and scene text recognitionTable and formula recognitionInformation extraction and document retrievalDocument forensics and provenancePrivate and secure document understanding

Who Should Attend

  • Chief Data Officers
  • Heads of AI/ML
  • Intelligent document processing product leaders
  • Computer vision and NLP researchers
  • ML engineers
  • Enterprise architects in insurance, banking and healthcare

Why Attend

If your roadmap includes automating claims, invoices, contracts, onboarding packets or KYC files, this is where you learn what document AI can genuinely do roughly a year before a vendor demo asserts it. You leave with a calibrated read on where multimodal LLMs still fail — dense tables, handwriting, low-resource scripts, degraded scans, multi-page reasoning — the benchmarks and datasets your evaluation harness should be using, and direct contact with the researchers whose work your document-processing vendor will productise next.

Event Features

  • Main-track topic area: foundation models for document understanding
  • Main-track topic area: multimodal LLMs for documents and agents for document AI
  • Main-track topic area: recognition of tables and formulas
  • Main-track topic area: private and secure document understanding
  • Main-track topic area: document forensics and provenance, and signature verification
  • Programme format: keynote talks, main-track sessions, panel discussions, poster exhibitions, side workshops and social gatherings

What Makes This Unique

  • The only top-tier venue devoted entirely to document analysis — the research layer beneath every IDP and document-extraction product on the market
  • The 2027 call for papers names foundation models for document understanding, multimodal LLMs for documents, agents for document AI and document reasoning models as first-class topics, not workshop side-tracks
  • At roughly 450 expected participants it is a working conference, not a trade show — the authors of the benchmarks your vendors quote are in the room
  • Twenty-first edition of a series that has defined the field's benchmarks and competitions since 1991

Industries Represented

Financial ServicesInsuranceHealthcareLegalGovernment

Pricing Details

Registration rates for ICDAR 2027 are not yet published — the conference site currently carries only the programme outline and the call for papers. Historically ICDAR uses tiered IAPR-member, non-member and student rates announced after acceptance notifications.

Organizer

I

ICDAR 2027 Organizing Committee

icdar2027.org/

Frequently Asked

When is ICDAR 2027?
ICDAR 2027 takes place August 18–22, 2027 in Kuala Lumpur, Malaysia.
Where is ICDAR 2027 held?
ICDAR 2027 is held in Kuala Lumpur, Malaysia. The event format is in-person.
What is the ICDAR 2027 submission deadline?
The next ICDAR 2027 deadline is abstract deadline on Jan 31, 2027. Remaining dates: Full paper deadline — Feb 20, 2027; Reviews released to authors — Apr 20, 2027; Rebuttal deadline — Apr 27, 2027; Acceptance notification — May 15, 2027; Camera-ready deadline — Jun 5, 2027.
How much does ICDAR 2027 cost?
ICDAR 2027 is paid. Registration rates for ICDAR 2027 are not yet published — the conference site currently carries only the programme outline and the call for papers. Historically ICDAR uses tiered IAPR-member, non-member and student rates announced after acceptance notifications.
Who should attend ICDAR 2027?
ICDAR 2027 is aimed at Chief Data Officers, Heads of AI/ML, Intelligent document processing product leaders, Computer vision and NLP researchers, ML engineers, Enterprise architects in insurance, banking and healthcare.
What topics does ICDAR 2027 cover?
ICDAR 2027 covers Document AI, Foundation models for document understanding, Multimodal LLMs for documents, Handwriting and scene text recognition, Table and formula recognition, Information extraction and document retrieval, Document forensics and provenance, Private and secure document understanding.

Sources

This page was written from 3 sources, 1 on domains other than icdar2027.org.

  1. 1.icdar2027.orgicdar2027.orgvendor
  2. 2.icdar2027.orgcall for papersvendor
  3. 3.cse.lehigh.educonferences
Newsletter

Stay Ahead of the Curve

Weekly enterprise AI insights for technology leaders. No spam, no vendor pitches—unsubscribe anytime.

Subscribe