Update README.md

README.md

@@ -18,9 +18,9 @@ configs:
 - config_name: default
   data_files:
   - split: eval
-    path: data/processbench_eval.parquet
+    path: splits/processbench_eval.parquet
   - split: train
-    path: data/train-*
+    path: splits/processbench_sft.parquet
 dataset_info:
   features:
   - name: image
@@ -37,27 +37,7 @@ dataset_info:
 
 ## Dataset Summary
 
-ProcessBench is a process-aware benchmark for robotic manipulation understanding. This release build contains `57,892` public QA rows: `9,051` eval rows and `48,841` SFT rows across `12` task families.
-
-## Included in this local release build
-
-- `data/processbench_eval.parquet` and `data/processbench_eval.jsonl`
-- `data/processdata_sft.parquet` and `data/processdata_sft.jsonl`
-- dataset-specific split files for all four sources
-- `metadata/split_summary.json`, `metadata/eval_manifest.json`, `metadata/sft_manifest.json`
-- `metadata/task_distribution.csv`
-- `metadata/schema.md`, `metadata/reconstruction.md`, `metadata/prompt_templates.md`
-- post-trained `ProcessEval-7B` results under `ProcessEval_results/`
-- post-trained `ProcessEval-7B` model LoRA adapter weights and configuration under `ProcessEval-7B/`
-- benchmark suite `ProcessData-58k` under `ProcessData-58k/`
-
-
-## Sources
-
-- `GM-100`
-- `RH20T`
-- `REASSEMBLE`
-- `AIST-Bimanual`
+ProcessBench is a process-aware benchmark for robotic manipulation understanding. This release build contains `57,892` QA rows: `9,051` Eval rows and `48,841` SFT rows across `12` task families. The split follows strict episode / recording / scene isolation. Upstream dataset sources include `GM-100`, `RH20T`, `REASSEMBLE`, and `AIST-Bimanual`.
 
 ## Task Families
 
@@ -74,11 +54,131 @@ ProcessBench is a process-aware benchmark for robotic manipulation understanding
 - `T11`: Next Primitive Prediction
 - `T12`: Primitive Chain Restoration
 
-## Splits
+## SFT-Eval Split Reconstruction
+
+### Relevant Assets
+
+- Public QA rows for `eval` and `sft`.
+- Split manifests and split counts.
+- Public release-relative references for each item.
+
+### How to use the public references
+
+Each row exposes three reconstruction-oriented fields:
+
+- `visual_ref`
+- `source_episode_ref`
+- `reconstruction_key_json`
+
+These are designed to be stable, path-free references. They are sufficient to map a released item back to the corresponding source episode or recording once you have obtained the upstream dataset under its original terms.
+
+### Source-specific units
+
+- `GM-100`: grouped by `(task_id, episode_id)`.
+- `RH20T`: grouped by `recording_id`.
+- `REASSEMBLE`: grouped by `recording_id`.
+- `AIST-Bimanual`: grouped by `recording_id`.
+
+### Expected workflow
+
+1. Obtain upstream datasets under their original terms.
+2. Use `source`, `source_task_id`, `source_unit_id`, `camera`, and `frame_indices_json` from the public release.
+3. Reconstruct the required visual inputs with the source-specific evaluation scripts in the main codebase.
+
+
+## Public QA fields Schema
+
+- `item_id`: Stable public item identifier.
+- `split`: `eval` or `sft`.
+- `source`: `GM-100`, `RH20T`, `REASSEMBLE`, or `AIST-Bimanual`.
+- `source_slug`: Short source slug used in release-relative references.
+- `source_task_id`: Source-side task identifier without local paths.
+- `source_unit_type`: Split-isolation unit type, e.g. `episode` or `recording`.
+- `source_unit_id`: Public unit identifier without local paths.
+- `task_id`: Canonical paper task ID `T1` ... `T12`.
+- `task_name`: Canonical paper task name.
+- `task_type_legacy`: Legacy engineering task name when it differs from paper naming.
+- `input_type`: Public input format description.
+- `question`: Public question text.
+- `choice_A` ... `choice_F`: Flattened answer choices.
+- `answer`: Public answer label in the release schema.
+- `answer_text`: Public answer text mapped from `answer`.
+- `num_choices`: Number of non-null answer choices.
+- `num_frames`: Number of referenced frames.
+- `frame_indices_json`: JSON-encoded frame indices.
+- `display_labels_json`: JSON-encoded panel labels such as `["X", "Y"]` or `["X", "Z", "Y"]`.
+- `camera`: Source camera identifier when available.
+- `arm_type`: Source-side arm configuration tag when available.
+- `visual_ref`: Release-relative visual reference string. This is not a local filesystem path.
+- `source_episode_ref`: Release-relative source unit reference.
+- `reconstruction_key_json`: Minimal JSON payload needed to reconstruct the sample with upstream data and source-specific scripts.
+- `task_meta_in_source`: Whether source-side task metadata/context exists upstream.
+- `task_meta_public`: Always `false` in this release build. Separate structured task-meta fields are intentionally withheld.
+- `split_version`: Current split version.
+- `split_group_id`: Source-side split group identifier.
+- `builder_version`: Public builder version tag.
+- `prompt_version`: Public prompt-serialization version tag.
+- `sft_target`: SFT-only supervised target string. Present only in `processdata_sft.*`.
+
+### Notes
+
+- `T5` is equivalent to `T_progress`.
+- `T8` is equivalent to `T_temporal`.
+- `T9` is equivalent to `T_binary`.
+- `T8` is normalized into six explicit permutation choices so the public table remains flat and Parquet-friendly.
+- `T9` is normalized to `A/B` choices even when the source data originally used `X/Y`.
+
+
+
+## Prompt Templates
+
+This file documents the public prompt serialization used by the current release build. Separate source-side task-meta prompting is intentionally disabled in this public release.
+
+### Standard multiple-choice template
+
+```text
+{question}
+
+A: {choice_A}
+B: {choice_B}
+...
+
+Choose exactly one option label.
+
+Output protocol:
+- You may include brief reasoning before the final answer.
+- The final line must be exactly: <ANSWER>A</ANSWER>
+- Do not output anything after </ANSWER>.
+```
+
+### `T8` Temporal Ordering template
+
+```text
+You are shown 3 frames from a robot manipulation task.
+The frames are labeled X, Y, Z (these labels are arbitrary identifiers, not positional, and not time-ordered).
+Determine the correct chronological order of these frames (from earliest to latest).
+Choose exactly one 3-letter permutation, for example: YXZ means Y happened first, then X, then Z.
+
+Output protocol:
+- You may include brief reasoning before the final answer.
+- The final line must be exactly: <ANSWER>XYZ</ANSWER>
+- Do not output anything after </ANSWER>.
+```
+
+### `T9` Temporal Priority template
+
+```text
+A single comparison image shows two labeled robot-manipulation panels from the same episode.
+The left-right placement of the panels and the labels are arbitrary identifiers and do not indicate temporal order.
+Which labeled panel happened earlier in the real manipulation sequence, X or Y?
+Choose exactly one label: X or Y.
+
+Output protocol:
+- You may include brief reasoning before the final answer.
+- The final line must be exactly: <ANSWER>X</ANSWER>
+- Do not output anything after </ANSWER>.
+```
 
-- `eval`: `9,051`
-- `sft`: `48,841`
-- split rule: strict episode / recording / scene isolation
 
 ## License and Terms