Upload CulturalMenuBench v3 candidate

MenuBench_media_v3_candidate_20261021_01-of-02.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8cc3def270788f288f0bf10960a2f83bbc3e31b0554a018021dca37f0914979f
+size 502735343

MenuBench_media_v3_candidate_20261021_02-of-02.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28fb81b85d8ef46304219cf68622573381d8b6d3afb6a2500121af706a403b6d
+size 249100731

MenuBench_metadata_v3_candidate_20261021.zip

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:779165eac90c48514e8c4f987c0f5692e6259509b3c8f3f1a7e15e938e748156
+size 7219480

README.md

@@ -0,0 +1,111 @@
+---
+pretty_name: CulturalMenuBench
+language:
+  - zh
+  - en
+tags:
+  - multimodal
+  - culinary-reasoning
+  - benchmark
+  - multiple-choice
+size_categories:
+  - 10K<n<100K
+---
+
+# CulturalMenuBench
+
+**CulturalMenuBench: Probing the Knowledge-Application Gap in Multimodal Culinary Reasoning**
+
+[Paper](https://arxiv.org/abs/2609.03526) · [Code](https://github.com/BobTsang-NLP/CulturalMenuBench)
+
+## Status and permissions
+
+This is the v3 **local release candidate**, prepared for a possible Hugging Face
+dataset repository. It has not been uploaded by this packaging step. Company
+authorization to export these files and permission to redistribute third-party
+recipe images and text must be confirmed before any external upload, including
+to a private repository. See `RIGHTS.md`. No dataset license is granted here;
+the code repository's Apache-2.0 license does not license third-party content.
+
+## Dataset scope
+
+| Suite | Questions |
+| --- | ---: |
+| Binary | 4,870 |
+| Chinese multiple-choice | 2,607 |
+| Foreign multiple-choice | 3,217 |
+| Subregion | 354 |
+| **Total** | **11,048** |
+
+The candidate contains 587 main dishes and 12,057 physical JPEG images addressed
+by 12,060 image IDs; three IDs share existing files. All questions and images
+are included. This is not a sample-only package.
+
+## Download and extraction
+
+Download these **three independent archives** and `SHA256SUMS`:
+
+- `MenuBench_metadata_v3_candidate_20261021.zip`
+- `MenuBench_media_v3_candidate_20261021_01-of-02.zip`
+- `MenuBench_media_v3_candidate_20261021_02-of-02.zip`
+
+Do not concatenate the archives. Verify the downloaded files, then extract all
+three archives into the **same empty directory**:
+
+```bash
+sha256sum -c SHA256SUMS
+mkdir data
+python3 -m zipfile -e MenuBench_metadata_v3_candidate_20261021.zip data
+python3 -m zipfile -e MenuBench_media_v3_candidate_20261021_01-of-02.zip data
+python3 -m zipfile -e MenuBench_media_v3_candidate_20261021_02-of-02.zip data
+cd data
+sha256sum -c SHA256SUMS
+python3 scripts/load_dataset.py --questions questions/binary.jsonl
+python3 scripts/load_dataset.py --questions questions/mc-chinese.jsonl
+python3 scripts/load_dataset.py --questions questions/mc-foreign.jsonl
+python3 scripts/load_dataset.py --questions questions/subregion.jsonl
+```
+
+For the first checksum command, keep this repository's `README.md` and
+`RIGHTS.md` alongside the three archives. The second checksum command checks
+the extracted dataset rather than the download files.
+
+Python 3.9+ with the standard library is sufficient for the included loader and
+scorer. This repository layout is a downloadable archive distribution; it does
+not promise automatic Hugging Face Dataset Viewer or `datasets.load_dataset`
+support. Use the included loader or the code repository's scripts.
+
+## Schema and evaluation
+
+`questions/*.jsonl` contains the four task suites. `items.jsonl` contains dish
+metadata, and `media_manifest.jsonl` resolves `asset://image:...` references to
+image files. Preserve ordered question input blocks, including repeated images.
+Never supply answer labels, source recipe metadata, or review records as model
+inputs. `scripts/score.py` provides local scoring; it makes no model API calls.
+
+## Version boundary and limitations
+
+Dish display names and their task-specific question spans were cleaned, and
+non-JPEG parent images were encoded as native-resolution JPEG Q95/4:4:4.
+Existing JPEG files retain their original bytes. Question IDs, answers, option
+order, image IDs and image order remain unchanged from the v2 candidate.
+There is no super-resolution or new image repair in this packaging step.
+
+Earlier image preprocessing includes LaMa/Qwen repair. Neither that processing
+nor JPEG encoding guarantees watermark-free images; residual or uncertain
+overlays may remain. Source attribution and preprocessing records are retained
+inside the metadata archive, not discarded to make this bundle smaller.
+
+This revised candidate has **not been reevaluated**. Historical paper scores
+must not be presented as results measured on these revised inputs.
+`KNOWN_ISSUES.jsonl` retains eight answer-change proposals and 21 scope proposals;
+they have not been applied. See the extracted `README.md`, `RELEASE.json`, and
+preprocessing/name-review records for detailed version information.
+
+## Publication checklist
+
+After obtaining export and redistribution approval, create the intended Hugging
+Face dataset repository and upload the **six extracted files** at its root.
+The outer transport ZIP is for local transfer only, not a fourth data archive.
+Before making the dataset public, finalize licensing, attribution, a public
+contact/takedown channel, and the candidate/public release status consistently.

RIGHTS.md

@@ -0,0 +1,16 @@
+# Rights and distribution
+
+This is a local release candidate, not authorization to export or redistribute.
+Obtain the appropriate company authorization before transferring these files to
+an external service, even if the destination repository is private.
+
+Recipe images, source text and translations have not been granted an open
+license here. Source URLs and image hashes support attribution review, not proof
+of permission. Image processing does not create redistribution permission.
+
+The existing Apache-2.0 license of the companion code repository does not extend
+to third-party recipe images or text. No blanket dataset license is granted.
+The authors must finalize redistribution permissions, annotation licensing,
+attribution, a public contact and a takedown channel before public release.
+
+The dataset's existing rights status remains `unreviewed_do_not_publish`.

SHA256SUMS

@@ -0,0 +1,5 @@
+8cc3def270788f288f0bf10960a2f83bbc3e31b0554a018021dca37f0914979f  MenuBench_media_v3_candidate_20261021_01-of-02.zip
+28fb81b85d8ef46304219cf68622573381d8b6d3afb6a2500121af706a403b6d  MenuBench_media_v3_candidate_20261021_02-of-02.zip
+779165eac90c48514e8c4f987c0f5692e6259509b3c8f3f1a7e15e938e748156  MenuBench_metadata_v3_candidate_20261021.zip
+87d6f7a7e13c5e2ce4ac686b78df1f68f068972f58c10309df37ba406c73d7e3  README.md
+545e4f439044f54bf63954e87dc745c8b7d350003988f381b4ffe650c97f59aa  RIGHTS.md