Main Track - Archival
New research submissions that may be published in the ACL Anthology as part of the workshop proceedings; must present substantially new and original work.
NLP/CORE Unranked
SURGeLLM 2026
SURGeLLM: Structured Understanding, Retrieval, and Generation in LLMs era
SURGeLLM 2026 is a workshop at ACL 2026 that brings together researchers from NLP, information retrieval, data management, and visualization to advance structure-aware understanding, retrieval, generation, and evaluation of structured artifacts such as tables, charts, maps, flowcharts, and diagrams. The workshop emphasizes making these structured elements first-class citizens in LLM systems through novel methods, benchmarks, and real-world applications.
Key deadlines
Verified September 9, 2026Full paper
March 26, 2026
AoE
Conference timeline
Submission and decisions
Full paperKey deadline
March 26, 2026 · AoE
Paper fit
Contribution paths
A strong submission should clearly identify its contribution and evaluate it appropriately.
Main Track - Non-Archival
New research submissions intended for presentation only, without archival publication in the ACL Anthology.
Presentation-only Track
Recently accepted or published work (e.g., from ACL Findings or main conferences) presented at the workshop without archival publication.
Research areas in scope
01
Structured Understanding
Methods for tabular understanding and question answering (multi-table reasoning, multi-hop inference, and multimodal integration)Natural language interfaces to structured data (Text-to-SQL, semantic parsing, schema linking, and data discovery)Structured retrieval (tables, cells/rows, chart elements, maps, workflows, and code fragments)Data-centric AI for LLMs on structured data (representation learning, augmentation, robustness, and domain adaptation)Agentic systems for structured data understanding and analysis (time series, graphs, tabular, etc)Benchmarking and evaluation (scalability, throughput, contamination, reproducibility, and human-centered assessment)Applications and governance for LLMs over structured data (DataOps, privacy, fairness, and reliability)
Policies worth checking twice
- Submissions may be under review elsewhere at the time of submission.
- Archival submissions must comply with ACL policy on overlap/self-plagiarism and commit to publishing in exactly one venue.
- Authors must indicate at submission time whether their paper is intended for archival or non-archival presentation.
- Submissions must follow ACL formatting guidelines (up to 8 pages for long papers, up to 4 pages for short papers, excluding references and appendices).
- Non-public resources (e.g., proprietary data) must be responsibly summarized and comply with ACL policies.
- The workshop follows the ACL Anti-Harassment Policy and ACL Code of Ethics.
Official sources
Compiled from the official call for papers. The organizers’ pages remain authoritative.
Last verified September 9, 2026