11th Workshop on
Automated Knowledge Base Construction

Co-located with EMNLP 2026 in Budapest, Hungary — October 28, 2026 Room F8.

While Large Language Models (LLMs) have revolutionized NLP, they remain prone to hallucinations, reasoning “mode-collapse” in open-ended generation, and a lack of factual provenance. The Automated Knowledge Base Construction (AKBC) workshop addresses a key missing piece of the generative era: structured knowledge. Knowledge Bases (KBs) serve as ground truth for fact verification, the semantic backbone for constrained decoding in generation, and as a resource behind Retrieval-Augmented Generation (RAG).

The workshop contributes to the growing momentum around integrating structured knowledge into generative models, both at training time and at inference time.

It follows a successful series of previous editions: as an independent conference in 2022, 2021, 2020, and 2019; as a workshop in 2017 at NIPS, in 2016 at NAACL, in 2014 at NIPS, in 2013 at CIKM, in 2012 at NAACL; and in 2010 as a stand-alone event in Grenoble, France.

Stay in the loop. Follow AKBC on LinkedIn for deadlines, speaker announcements, and shared task updates. Follow →

Sponsors

We are grateful to our sponsors:

News

  • We are grateful to our sponsors Bloomberg and SLB!
  • September 8th, 2026: Out of 58 submissions, 27 papers were accepted — see the list of accepted papers
  • September 8th, 2026: Parisa Kordjamshidi (Michigan State University) is confirmed as a keynote speaker
  • September 1st, 2026: Shared task results are out — see the winners on the shared task page and the full final leaderboard
  • August 12th, 2026: Denny Vrandečić (Wikimedia) is confirmed as a keynote speaker
  • July 30th, 2026: Amir Globerson (Tel Aviv University & Google) is confirmed as a keynote speaker
  • July 23rd, 2026: Dan Roth (University of Pennsylvania & Oracle) is confirmed as a keynote speaker
  • May 28th, 2026: Alon Halevy (Google) is confirmed as a keynote speaker
  • May 27th, 2026: The workshop day is fixed to October 28, 2026
  • May 21st, 2026: We are excited to announce Heng Ji (UIUC) as a keynote speaker
  • May 21st, 2026: We are excited to announce Mausam (IIT Delhi) as a keynote speaker
  • April 27th, 2026: We receive news that the workshop will be held on October 28, 2026
  • April 13th, 2026: The Web page of AKBC goes live

Accepted Papers

Accepted papers are presented as posters, with a selection invited for lightning talks. Remote viewing is possible; remote authors upload their poster to the workshop website instead of presenting physically.

Research Papers

  • A Rhetorical Citation Event Graph Model for Characterizing Scientific Findings: Resource Construction and Application to Plant HealthAnne-Sophie Foussat, Robert Bossy, Vincent Guigue, Sauvion, Claire Nédellec
  • A Temporal Knowledge Graph for Music Festival Lineup ForecastingJulia Gastinger, Thilo Ignaz Dieing, Christian Meilicke, Heiner Stuckenschmidt
  • Auditing a KB Elicitation of Frontier LLM Knowledge: A Multi-dimensional Analysis of GPTKB v1.5Shrestha Ghosh, Luca Giordano, Yujia Hu, Tuan-Phong Nguyen, Simon Razniewski
  • Can We Do Interpretable NLI with Graphs Based on Atomic Propositions?Younes Boufouss, Luc Pommeret, Thomas Gerald, Patrick Paroubek, Sophie Rosset
  • Distill-RE: Few-Shot Relation Extraction via Teacher LLM Reasoning DistillationDanielle Sullivan-Pao, Sheila Alemany, Olga Simek
  • Edge Knowledge Bases for Sustainable and Explainable AIAparna S. Varde, Emil Njor, Francesco Daghero, Mahyar Tourchi Moghaddam
  • Epistemic Twins: Enabling a Symbolic Science of Language Model KnowledgeSimon Razniewski, Shrestha Ghosh, Luca Giordano, Yujia Hu, Tuan-Phong Nguyen
  • EventSocraticKG: Event-Centric Knowledge Graph Construction for Narrative Retrieval-Augmented GenerationWilliam Rosengren
  • Evidence-Carrying Claims: Fail-Closed Verification Gates for Language-Model-Assisted Research as Knowledge Base ConstructionRob Sneiderman
  • Evidence-Guided Schema Normalization for Temporal Tabular ReasoningAshish Thanga, Vibhu Dixit, Abhilash Shankarampeta, Vivek Gupta
  • Explaining Textual Entailment with Lexical Entailments: Using LLMs to Supply Lexical Relations for Formal ProofsJorryt de Jong, Stefan Moraca, Ettore Cesari, Lasha Abzianidze
  • FALCON: A Model and Dataset Agnostic Framework for Synthetic Data Generation for NL2SQL PairsDarian Lee, Shannon Rumsey, Jack St. Clair, Xinyi Tang, Aditya Bansal, Yuanming Shi
  • Fine-Grained Table Retrieval Through the Lens of Complex QueriesWojciech Kosiuk, Xingyu Ji, Yeounoh Chung, Fatma Ozcan, Madelon Hulsebos
  • From Imitation to Reasoning: Sample-Efficient LLM Agents for Knowledge Graph CompletionChenjun Zheng, Zhe Wang, Zhigang Wang, Wenjia Zhang, Xiaolin Yang
  • GRACE: Graph-Grounded Reflective Agent Copilot Engine for Expert-in-the-Loop Knowledge ExpansionJohn Seon Keun Yi, Joshua R Minot, Dokyun Lee
  • GRACE-Mem: Graph Retrieval with Agentic Curation of Evidence for Long-Term Conversational MemoryHeng Wang, Xuan Ni Chen, Pei-Yun Chang, Chih-Hsi Chen, Hsin-Yu Huang, Chienyi Lin, I-Ling Chung
  • HTTP for Meaning: Towards Linguistically Complete Knowledge Representations with Knowledge State ProtocolIan Yuen
  • LMKG: A Flexible Method for Ontology-Aware Knowledge Graph Extraction from TextGianmarco Pappacoda, Paolo Torroni
  • MUSART: A Scalable Multi-Domain Benchmark for Measuring Factual Recall in Open-Weight Language ModelsRajaa El Hamdani, Thomas Bonald, Fragkiskos D. Malliaros
  • Operationalization Is Part of the Claim: Operationalization-Aware Knowledge in the Era of Generated AnswersJan-Christoph Kalo
  • Revisiting RAG Retrievers: An Information Theoretic BenchmarkWenqing Zheng, Dmitri Kalaev, Noah Fatsi, Daniel Barcklow, Owen Reinert, Igor Melnyk, Senthil Kumar, C. Bayan Bruss
  • Scaling E-Commerce Attribute Extraction with Parallel DecodingNikhita Vedula, Dushyanta Dhyani, Bryan Wang, Shervin Malmasi
  • Schema-Guided Extraction of Interaction Knowledge Bases from Narrative Text: A Folktale Case StudySergey Kovalchuk, Anastasia Ugleva, Oleg Kubryak
  • The Canonical Order Problem: When Large Language Models Are Unreliable Knowledge Bases for Multi-Valued RelationsTimo Pierre Schrader, Annemarie Friedrich, Simon Razniewski, Lukas Lange
  • When Relevance Is Not Evidence: A Controlled Diagnostic of Claim-Conditioned Support RankingKe Lu
  • When String Matching Stands In for a Knowledge Base: A Role-Aware Audit of WikiBigEditEdison Yang, Neel Marripalapu, Akshaj Agadi
  • Zero-Shot Relation Classification with Large Language Models: Binary Prompting and Relevance ElicitationSheila Alemany, Danielle Sullivan-Pao, Olga Simek

Shared Task Papers

  • 🥇 Winner — Coverage Is Not Knowledge: Why Closed-Book Knowledge Base Construction Looks Selection BoundMaksim Silchenko
  • 🥈 2nd place — MAVE: Multi-Angle Voting for Emission in Closed-Book Knowledge Base ConstructionJonas Kunze, Andreas Both
  • 🥉 3rd place — The Instruction Tax on Factual Recall: Base Models Can Beat Their Instruction-Tuned CounterpartsChristophe Khalil, Aya Mourad
  • Closed-Book Knowledge Base Construction with Calibrated Abstention and Numeric Self-ConsistencyTanvir Ahmed
  • A Forced Answer Channel for Closed-Book Knowledge Base Construction at LM-KBC 2026Yiqun Sun, Pengfei Wei, Lawrence B. Hsieh
  • REAP: Relation-Aware Elicitation and Parsing for Closed-Book Knowledge Base Construction from LLMsThanh-Dan Bui, Do Thanh Trung, Tuan-Phong Nguyen
  • SetProbe: Description-First Relation-Wise Self-Consistency for Closed-Book Knowledge Base ConstructionJannic Alexander Cutura
  • Knowledge Base Construction from Language Models: Mechanisms and Calibrations for Multi-Object Relational Extraction under Parameter BudgetsChengcheng Wei, Ze Chen, Jiewen Zheng, Jiarong He
  • Pre-training Aware Directed Reasoning for Closed-Book Knowledge Base ConstructionMikołaj Tym
  • Exemplars Primarily Shift Retrieval, Verification Instructions Can Shift Rejection: An Anatomy of Inference-Time Interventions for Closed-Book Knowledge Base ConstructionTakashi Sasaki
  • COVER-KBC: Evidence-Centric Adaptive Inference for Closed-Book Knowledge Base ConstructionQuoc-Linh Vo
  • MuseList: Confidence-Gated Multi-Surface Recall for Closed-Book Knowledge Base ConstructionAdrian Zimmermann
  • One Workflow Does Not Fit All Relations: A System Description for the AKBC 2026 LM-KBC Shared TaskAndreea-Elena Cochintele, Anca Mădălina Dobrovăț, Sergiu Nisioi
  • CoRe: Construct-versus-Recall Routing for Knowledge Base Construction Under a Parameter BudgetFatih Naufal Habibillah, Filbert Aurelian Tjiaranata
  • Ensembling Elicitation Strategies for Knowledge Base ConstructionVlad-Madalin Ungureanu, Neacșu Matei, Ianis Belu
  • DataMind at AKBC Shared Task 2026: Relation-Specific Multiple Prompting for Closed-Book Knowledge Base CompletionTanaree Patcharatrakul, Peerapon Vateekul
  • Relation-Specific Elicitation of a Base Model under a 32B Closed-Book Budget for LM-KBC 2026Ruggero Marino Lazzaroni, Tim Booker, Pau Soler Valadés, Mathias Angermaier, Kirill Solovev

Shared Task

Large language models contain a substantial amount of factual knowledge. Turning that knowledge into reliable knowledge base entries, however, is much harder than answering a single factual question.

Given a subject s and a relation r, predict the complete set of correct object strings {o₁, o₂, …, oₖ}. Unlike standard factual QA, a subject may have zero, one, or many correct objects. The goal is to construct a complete and precise KB entry.

Further details are on the shared task page.

Invited Speakers

Denny Vrandečić
Denny Vrandečić
Wikimedia
Mausam
Mausam
IIT Delhi
Heng Ji
Heng Ji
University of Illinois Urbana-Champaign
Alon Halevy
Alon Halevy
Google
Dan Roth
Dan Roth
University of Pennsylvania/Oracle
Amir Globerson
Amir Globerson
Tel Aviv University/Google
Parisa Kordjamshidi
Parisa Kordjamshidi
Michigan State University

Provisional Schedule

The program is provisional and subject to change.

TimeItem
08:30–08:45Opening
08:45–09:15Invited talk 1: Denny Vrandecic
09:15–09:45Invited talk 2: Mausam
09:45–10:15Invited talk 3: Heng Ji: Constructing a 16-Million-Edge Knowledge Network for Agentic Scientific Discovery
10:15–10:30Featured paper 1: Explaining Textual Entailment with Lexical Entailments: Using LLMs to Supply Lexical Relations for Formal Proofs, de Jong et al.
10:30–11:00Coffee break
11:00–11:30Invited talk 4: Amir Globerson
11:30–12:00Invited talk 5: Parisa Kordjamshidi
12:00–12:15Featured paper 2: Zero-Shot Relation Classification with Large Language Models: Binary Prompting and Relevance Elicitation. Alemany et al.
12:15–12:30Featured paper 3: GRACE-Mem: Graph Retrieval with Agentic Curation of Evidence for Long-Term Conversational Memory, Wang et al.
12:30–13:30Lunch break
13:45–14:30Shared task session
14:30–16:00Poster session
15:30–16:00Coffee break (overlapping with posters)
16:00-16:30Invited talk 6: Alon Halevy: Designing Knowledge Agents
16:30–17:00Invited talk 7: Dan Roth
17:00–17:15Community discussion
17:15–17:30Closing
18:00-open endAKBC reception at Sky bar of Expo Tower Hotel (conference hotel) (TBC)

Organization

Organizing Committee

Contact us at akbc2026@gmail.com for workshop inquiries, or at akbc2026-shared-task@googlegroups.com for shared task inquiries.

Program Committee