Rename the three media archives without changing their bytes; update only media_json.archive in multimodal Parquet rows. Add a concise media README and consistently named checksum manifest. Requests, image members, scores, embeddings, prediction arrays, and report metrics are unchanged.
Remove development-only records from the public benchmark package
20,167 exact-match successful score restorations; known failures stay zero. Unverified legacy scores unchanged. Shared CLIP text/all-image late fusion, 12 canonical public baselines and ID-aligned prediction replay for all three scenarios.
Yesterday
Align all displayed benchmark naming with RobustRouteBench
Use RobustRouteBench for the displayed results-source name and archive description. Preserve the valid source URL, result values, dataset configuration, and all data assets.
Align RobustRouteBench documentation and figures with the technical report
Adopt the supplied card layout and banner; update result tables and quality-cost figure from the report frozen results. Clarify pointwise run aggregation, pooled Robust metrics, evaluation scope, and shared Qwen3-VL-Embedding-8B text-and-image representations. Dataset files, media, schema, model metadata, and the original release manifest are unchanged.
