paper-with-me

홈 › Papers

REKEY: Metadata-Grounded Visual-Key Regeneration for Contamination-Resilient VQA Evaluation

2026-06-17 · Tengjie Lin, Yutao Sun, Jingwei Ni, Shuhan Ge, Hao-Xuan Ma, Yanting Miao, Wangyue Lu, Mingshuai Chen, Tiancheng Zhao, Jianwei Yin arxiv

Static visual question answering (VQA) benchmarks age quickly: Once the items leak into training corpora, scores can reflect memorization rather than genuine visual ability, thus obscuring real progress. Rebuilding high-quality benchmarks such as V*Bench requires substantial human annotation, yet each static release can quickly become another leaked artifact. We propose ReKey, a live benchmark protocol that randomly regenerates the answer-bearing local detail, or visual key, in real images at evaluation time. Using human-validated edit slots, ReKey samples fresh instances with new answers, construction-grounded labels, and controlled visual-search difficulty. On V*Bench, the ReKey regenerated benchmark reveals a sharp score jump across eight frontier vision-language models (VLMs): The original items score 9.5--18.8 percentage points higher than the regenerated variants. By making the visual key renewable, ReKey keeps evaluation fresh as models and training data evolve.

📄 PDF Abstract BibTeX arXiv:2606.20736

Code (0)

등록된 구현이 없습니다.

Tasks

Visual Question Answering

Similar Papers 제목 키워드 기반

From Guidelines to Guarantees: A Graph-Based Evaluation Harness for Domain-Specific Evaluation of LLMs

2025-08-28 · Jessica M. Lundin, Usman Nasir Nakakana, Guillaume Chabot-Couture arxiv

Rigorous evaluation of domain-specific language models requires benchmarks that are comprehensive, contamination-resistant, and maintainable. Static, manually curated datasets do not satisfy these properties. We present …

Predicting Microbial Ontology and Pathogen Risk from Environmental Metadata with Large Language Models

2025-07-29 · Hyunwoo Yoo, Gail L. Rosen arxiv

Traditional machine learning models struggle to generalize in microbiome studies where only metadata is available, especially in small-sample settings or across studies with heterogeneous label formats. In this work, we …

Fine-Grained Benchmark Generation for Comprehensive Evaluation of Foundation Models

2026-05-12 · Mohammed Saidul Islam, Negin Baghbanzadeh, Farnaz Kohankhaki, Afshin Cheraghi 외 arxiv

Evaluation of foundation models often rely on aggregate scores from benchmarks that lack comprehensive coverage and metadata for a fine-grained evaluation. We introduce a framework for automated benchmark generation. Our…

Search-Time Contamination in Deep Research Agents: Measuring Performance Inflation in Public Benchmark Evaluation

2026-06-03 · Yongjie Wang, Xinyue Zhang, Kunhong Yao, Zhiwei Zeng 외 arxiv

Public benchmarks enable fair and reproducible evaluation of LLM reasoning, but they become fragile for deep research agents that actively search the web during inference. Such agents may retrieve public benchmark metada…

Appear2Meaning: A Cross-Cultural Benchmark for Structured Cultural Metadata Inference from Images

2026-04-08 · Yuechen Jiang, Enze Zhang, Md Mohsinul Kabir, Qianqian Xie 외 arxiv

Recent advances in vision-language models (VLMs) have improved image captioning for cultural heritage. However, inferring structured cultural metadata (e.g., creator, origin, period) from visual input remains underexplor…

Image Captioning