Short papers
Up to 4 pages (excluding references), following ACL two-column format; archival submissions reviewed by the workshop.
NLP/CORE Unranked
Insights 2026
Workshop on Insights from Negative Results in NLP
Aug 01 2026Official workshop site Site reachable
The Insights Workshop 2026 is a venue dedicated to promoting the publication and discussion of negative results in NLP, addressing the field's bias toward positive benchmark improvements. It encourages submissions that reveal methodological flaws, failed experiments, or overlooked issues in models, datasets, and evaluation practices to foster more rigorous and reproducible research.
Key deadlines
Verified September 9, 2026Full paper
August 15, 2026
AoE
Workshop timeline
Submission and decisions
Full paperKey deadline
August 15, 2026 · AoE
Paper fit
Contribution paths
A strong submission should clearly identify its contribution and evaluate it appropriately.
Non-archival abstract submissions
1-2 page abstracts for papers already published elsewhere (e.g., main conferences or non-NLP venues); non-archival and intended for discussion.
Research areas in scope
01
Negative Results in NLP
broadly applicable recommendations for training/fine-tuning/promptingablation studies showing unexpected contributionsdatasets or probing tasks revealing lack of generalizationtrivial baselines that work suspiciously wellcross-lingual failures of techniquesinstability of results due to hardware, initialization, or preprocessingtheoretical arguments against proposed methodsissues with data processing/collection/annotation pipelinesproblems with evaluation metrics (e.g., F1, BLEU, LLM-as-judge)failures to follow policies or unintended consequences in agentic systemsunder-reporting of training details, test data contamination, invalid comparisonspreliminary results of non-mainstream approaches
Policies worth checking twice
- Submissions must follow the official ACL two-column format; non-conforming papers will be rejected without review.
- Workshop cannot accept work that has been or will be published elsewhere and is concurrently submitted to overlapping venues.
- Authors may submit papers rejected from other venues and provide original reviews and responses; reviewers will not see this information, but organizers may consider it.
- All accepted papers must be presented at the workshop to appear in the proceedings.
- At least one author of each accepted paper must register for the workshop.
- Authors may use one extra page in the camera-ready version to address reviewer comments.
- Previous presentations (e.g., arXiv preprints) must be noted in the camera-ready version but not in the anonymized submission.
- The workshop follows the ACL Anonymity Policy.
Official sources
Compiled from the official call for papers. The organizers’ pages remain authoritative.
Last verified September 9, 2026