Long papers
Up to 9 pages with unlimited references and appendices
AI/CORE Unranked
EvalMG 2026
The Second Workshop on Evaluation of Multimodal Generation
EvalMG 2026 is the second workshop focused on evaluation methods for multimodal generation and retrieval systems, held in conjunction with SIGIR 2026. It brings together researchers from information retrieval, NLP, computer vision, and multimodal AI to address challenges in assessing the quality, fairness, and robustness of multimodal systems, with a special emphasis on RAG, agentic AI, and cross-modal evaluation.
Key deadlines
Verified September 9, 2026Full paper
May 28, 2026
AoE
Workshop timeline
Submission and decisions
Full paperKey deadline
May 28, 2026 · AoE
Paper fit
Contribution paths
A strong submission should clearly identify its contribution and evaluate it appropriately.
Short papers
Up to 4 pages with unlimited references and appendices
Non-archival submissions
Papers accepted to other conferences (e.g., ACL, EMNLP, WWW, KDD, SIGIR) may be submitted for presentation without inclusion in workshop proceedings
SIGIR'26 Fast Track submissions
SIGIR'26 submissions that were not accepted may be resubmitted with the original paper and review comments attached
Research areas in scope
01
Evaluation Topics
Multimodal retrieval for RAG, Agentic AI, recommendation systemsEvaluation of retrieved cross-modal samples, without relying on augmented generationMulti-aspect evaluation methods capturing inter- and intra-modal coherence, relevance, grounding, and contextual consistencyBenchmark retrieval datasets, evaluation protocols and annotations for text–image–audio–video–3D generationAutomatic and human-centric metrics for informativeness, factuality, fluency, faithfulness, calibration, and usability for multimodal generationMethodology for detecting, analysing, and mitigating multimodal bias, stereotypes, toxicity, and hallucinationsEvaluation in multimodal low-resource and multilingual settings, including culturally aware and cross-lingual metricsAgent-based evaluation of multimodal generation in multi-turn, tool-use, or iterative editing scenariosGame-theoretic or optimization-based formulations of evaluation objectives and protocolsEvaluation of the generation quality of synthetic multimodal data, provenance/attribution, and downstream impact on training and deploymentEthical considerations in the evaluation of multimodal text generation, including bias detection and mitigation strategiesEvaluation of Security and Privacy Dimensions in Multimodal Applications
Policies worth checking twice
- Papers must be anonymous for double-blind review
- Submissions must strictly follow SIGIR submission guidelines
- Direct submissions are accepted via OpenReview
- ARR paper commitment is accepted via a dedicated OpenReview page
- SIGIR'26 Fast Track submissions require both the paper and review comments as attachments; submissions without reviews will be desk-rejected
- Non-archival submissions are allowed for papers already accepted at other conferences and will not be included in proceedings
Official sources
Compiled from the official call for papers. The organizers’ pages remain authoritative.
Last verified September 9, 2026