Long Papers
Complete research contributions with novel findings, experimental results, and thorough analysis; up to 8 pages + references.
AI/CORE Unranked
Eval4SD 2026
First Workshop on Evaluating LLMs for Specialized Domains
Eval4SD 2026 is the first workshop focused on evaluating large language models (LLMs) for specialized domains such as law, medicine, digital humanities, and social sciences. It aims to address the lack of reliable evaluation resources in niche areas by promoting benchmarking, replication of domain research, and novel evaluation methodologies, with an emphasis on open-weight models and low-resource languages.
Paper fit
Contribution paths
A strong submission should clearly identify its contribution and evaluate it appropriately.
Short & Position Papers
Preliminary results, position papers, system descriptions, and focused contributions; up to 4 pages + references.
Research areas in scope
01
LLM Benchmarking
Evaluating multiple models, datasets, inference methods, or prompting techniquesIntroducing novel specialized benchmarking datasetsPerformance in low- and medium-resource languages
02
Domain Research Replication
Replicating existing domain research using altered LLM setupsTesting robustness of domain conclusions to changes in prompting, model weights, or training dataEvaluation of open-weight models for replicability
03
Metrics and Evaluation Methodology
Methodology for assessing LLM outputs in complex tasksLLM judge setupsNovel rule-based metrics for specialized tasks
Policies worth checking twice
- Submissions undergo double-blind peer review
- Submissions must follow the ACL template
- Submitted papers (except non-archival) must not be under review elsewhere
- Non-archival submissions are allowed for recently published work or work-in-progress without violating dual-submission policies
- Non-archival papers will be presented but not included in the official proceedings
Official sources
Compiled from the official call for papers. The organizers’ pages remain authoritative.
Last verified September 9, 2026