Update croissant.json

croissant.json

@@ -54,7 +54,7 @@
       "@id": "repo",
       "name": "repo",
       "description": "The Hugging Face git repository.",
-      "contentUrl": "https://huggingface.co/datasets/ProcessBench-2026/ProcessBench-Anom/tree/refs%2Fconvert%2Fparquet",
+      "contentUrl": "https://huggingface.co/datasets/ProcessBench-2026/RoboProcessBench/tree/refs%2Fconvert%2Fparquet",
       "encodingFormat": "git+https",
       "sha256": "https://github.com/mlcommons/croissant/issues/80"
     },
@@ -97,7 +97,7 @@
     {
       "@type": "cr:RecordSet",
       "@id": "default",
-      "description": "ProcessBench-2026/ProcessBench-Anom - 'default' subset\n\nAdditional information:\n- 2 splits: train, eval",
+      "description": "ProcessBench-2026/RoboProcessBench - 'default' subset\n\nAdditional information:\n- 2 splits: train, eval",
       "field": [
         {
           "@type": "cr:Field",
@@ -592,11 +592,11 @@
     }
   ],
   "conformsTo": "http://mlcommons.org/croissant/1.1",
-  "name": "ProcessBench-Anom",
-  "description": "\n\t\n\t\t\n\t\tProcessBench\n\t\n\n\n\t\n\t\t\n\t\tDataset Summary\n\t\n\nProcessBench is a process-aware benchmark for vision-language robotic manipulation understanding. It evaluates whether VLMs can infer how a manipulation execution unfolds, including phase, contact, motion, bimanual coordination, primitive-local progress, temporal order, outcome, and primitive-level transitions.\nThis release contains 57,892 QA rows: 48,841 SFT rows and 9,051 evaluation rows across 12 task families and 260 manipulation tasks.… See the full description on the dataset page: https://huggingface.co/datasets/ProcessBench-2026/ProcessBench-Anom.",
+  "name": "RoboProcessBench",
+  "description": "\n\t\n\t\t\n\t\tRoboProcessBench\n\t\n\n\n\t\n\t\t\n\t\tDataset Summary\n\t\n\nRoboProcessBench is a process-aware benchmark for vision-language robotic manipulation understanding. It evaluates whether VLMs can infer how a manipulation execution unfolds, including phase, contact, motion, bimanual coordination, primitive-local progress, temporal order, outcome, and primitive-level transitions.\nThis release contains 57,892 QA rows: 48,841 SFT rows and 9,051 evaluation rows across 12 task families and 260 manipulation tasks.… See the full description on the dataset page: https://huggingface.co/datasets/ProcessBench-2026/ProcessBench-Anom.",
   "alternateName": [
-    "ProcessBench-2026/ProcessBench-Anom",
-    "ProcessBench"
+    "ProcessBench-2026/RoboProcessBench",
+    "RoboProcessBench"
   ],
   "creator": {
     "@type": "Organization",
@@ -625,12 +625,12 @@
     "manipulation"
   ],
   "license": "https://choosealicense.com/licenses/other/",
-  "url": "https://huggingface.co/datasets/ProcessBench-2026/ProcessBench-Anom",
-  "rai:dataLimitations": "ProcessBench evaluates VLM-side process understanding in robotic manipulation, not closed-loop robot control or robot safety. It is derived from GM-100, RH20T, REASSEMBLE, and AIST-Bimanual, so source coverage, viewpoints, embodiments, sensing modalities, and task-family support are heterogeneous. Not every source supports every task family, and primitive-aware tasks depend on available primitive annotations. The benchmark uses multiple-choice QA and should be interpreted with task-level scores, random/majority baselines, human audit, and bootstrap CIs. It is not recommended for robot safety certification, deployment readiness claims, generic VLM ranking, or causal claims about downstream VLA policy improvement without additional validation.",
-  "rai:dataBiases": "ProcessBench inherits biases from its four upstream robotic datasets, including robot embodiment, camera viewpoint, tabletop scene layout, object category, task-family coverage, sensing availability, annotation granularity, and success/failure distribution. Some process cues are available only for sources with suitable native signals or annotations, so task-family coverage is uneven. Labels are generated by deterministic builders from source-native signals, and label quality depends on upstream signal resolution, annotation quality, and calibration choices. Models may learn source-specific visual or embodiment priors if results are not interpreted by task and source.",
+  "url": "https://huggingface.co/datasets/ProcessBench-2026/RoboProcessBench",
+  "rai:dataLimitations": "RoboProcessBench evaluates VLM-side process understanding in robotic manipulation, not closed-loop robot control or robot safety. It is derived from GM-100, RH20T, REASSEMBLE, and AIST-Bimanual, so source coverage, viewpoints, embodiments, sensing modalities, and task-family support are heterogeneous. Not every source supports every task family, and primitive-aware tasks depend on available primitive annotations. The benchmark uses multiple-choice QA and should be interpreted with task-level scores, random/majority baselines, human audit, and bootstrap CIs. It is not recommended for robot safety certification, deployment readiness claims, generic VLM ranking, or causal claims about downstream VLA policy improvement without additional validation.",
+  "rai:dataBiases": "RoboProcessBench inherits biases from its four upstream robotic datasets, including robot embodiment, camera viewpoint, tabletop scene layout, object category, task-family coverage, sensing availability, annotation granularity, and success/failure distribution. Some process cues are available only for sources with suitable native signals or annotations, so task-family coverage is uneven. Labels are generated by deterministic builders from source-native signals, and label quality depends on upstream signal resolution, annotation quality, and calibration choices. Models may learn source-specific visual or embodiment priors if results are not interpreted by task and source.",
   "rai:personalSensitiveInformation": "None of the listed sensitive categories are intentionally included.",
-  "rai:dataUseCases": "ProcessBench measures process-aware VLM understanding in robotic manipulation: whether a model can infer how an execution is unfolding from visual observations. It covers phase, contact, motion, coordination, primitive-local progress, temporal order, outcome, and primitive-level transitions. Validated use cases include diagnostic VLM evaluation, task-family-level failure analysis, and process-aware VLM post-training, supported by deterministic GT builders, strict split isolation, human audit, random/majority baselines, and bootstrap CIs. Not validated: closed-loop robot policy evaluation, robot safety certification, deployment readiness, generic VLM ranking, fairness auditing of human populations, or causal claims about downstream VLA improvement.",
-  "rai:dataSocialImpact": "ProcessBench can improve transparency around VLM failure modes in robotic manipulation by evaluating process-level cues beyond final success. It may support research on visual critics, progress monitors, failure detectors, and process-aware VLM evaluators. The main risk is over-interpreting benchmark accuracy as evidence of safe or reliable robot behavior. Mitigations include intended-use and out-of-scope statements, task-level reporting, random/majority baselines, human audit summaries, bootstrap CIs, and reconstruction notes. Full upstream raw videos are not redistributed, and downstream robot deployment requires additional validation.",
+  "rai:dataUseCases": "RoboProcessBench measures process-aware VLM understanding in robotic manipulation: whether a model can infer how an execution is unfolding from visual observations. It covers phase, contact, motion, coordination, primitive-local progress, temporal order, outcome, and primitive-level transitions. Validated use cases include diagnostic VLM evaluation, task-family-level failure analysis, and process-aware VLM post-training, supported by deterministic GT builders, strict split isolation, human audit, random/majority baselines, and bootstrap CIs. Not validated: closed-loop robot policy evaluation, robot safety certification, deployment readiness, generic VLM ranking, fairness auditing of human populations, or causal claims about downstream VLA improvement.",
+  "rai:dataSocialImpact": "RoboProcessBench can improve transparency around VLM failure modes in robotic manipulation by evaluating process-level cues beyond final success. It may support research on visual critics, progress monitors, failure detectors, and process-aware VLM evaluators. The main risk is over-interpreting benchmark accuracy as evidence of safe or reliable robot behavior. Mitigations include intended-use and out-of-scope statements, task-level reporting, random/majority baselines, human audit summaries, bootstrap CIs, and reconstruction notes. Full upstream raw videos are not redistributed, and downstream robot deployment requires additional validation.",
   "rai:hasSyntheticData": false,
   "prov:wasDerivedFrom": [
     {
@@ -661,7 +661,7 @@
         "@id": "https://www.wikidata.org/wiki/Q4929239"
       },
       "prov:label": "Source data collection",
-      "sc:description": "ProcessBench is derived from existing robotic manipulation datasets: GM-100, RH20T, REASSEMBLE, and AIST-Bimanual. The original robotic trajectories, videos, sensor signals, task metadata, primitive annotations, and success/failure records were collected by the upstream dataset creators. ProcessBench does not recollect raw robot trajectories and does not redistribute full upstream raw videos or full frame dumps."
+      "sc:description": "RoboProcessBench is derived from existing robotic manipulation datasets: GM-100, RH20T, REASSEMBLE, and AIST-Bimanual. The original robotic trajectories, videos, sensor signals, task metadata, primitive annotations, and success/failure records were collected by the upstream dataset creators. ProcessBench does not recollect raw robot trajectories and does not redistribute full upstream raw videos or full frame dumps."
     },
     {
       "@type": "prov:Activity",
@@ -669,7 +669,7 @@
         "@id": "https://www.wikidata.org/wiki/Q5227332"
       },
       "prov:label": "Source-aware preprocessing",
-      "sc:description": "We standardize source metadata into a unified ProcessBench schema with item IDs, source identifiers, task-family IDs, questions, answer choices, GT answers, visual references, reconstruction keys, split identifiers, builder versions, and prompt versions. Source-specific fields are preserved as reconstruction metadata when needed. Samples with unstable source references, missing visual inputs, or ambiguous local decision units are filtered."
+      "sc:description": "We standardize source metadata into a unified RoboProcessBench schema with item IDs, source identifiers, task-family IDs, questions, answer choices, GT answers, visual references, reconstruction keys, split identifiers, builder versions, and prompt versions. Source-specific fields are preserved as reconstruction metadata when needed. Samples with unstable source references, missing visual inputs, or ambiguous local decision units are filtered."
     },
     {
       "@type": "prov:Activity",