docs(card): simplify dataset and project descriptions

NOTICE.md

@@ -1,6 +1,10 @@
 # Attribution and licensing notice
 
-This repository is a community distribution prepared by zhangdw. Original dataset ownership and authorship remain with **Meta AI Research / Meta Platforms, Inc.** It redistributes the official GAIA2 mini validation Parquet and published GAIA2 filesystem resources without changing their bytes. Directory placement is changed for convenient local use; the row index, manifests, validation metadata and this repository's documentation are newly generated.
+GAIA2 is a dataset by **Meta AI Research / Meta Platforms, Inc.** This distribution,
+prepared by zhangdw, packages the original mini validation Parquet and published
+filesystem files for local use. It adds a row index, manifests, validation metadata
+and documentation. Source revisions, file hashes and directory mappings are recorded
+in `provenance.json` and `data-manifest.json`.
 
 Original sources:
 
@@ -20,7 +24,10 @@ The filesystem card additionally states:
 
 > Some files come from a subset of the Wikipedia and are licensed under the [Wikipedia own license](https://en.wikipedia.org/wiki/Wikipedia:Copyrights).
 
-The source cards request that evaluation data not be used for training. Their notices and benchmark canary markers are retained. Individual documents, images and other third-party resources retain their respective rights and notices; this repository does not relicense them. The Creative Commons Attribution 4.0 International terms are available at https://creativecommons.org/licenses/by/4.0/legalcode.
+The source cards request use of the data for evaluation. Individual documents,
+images and other third-party resources are distributed under their respective
+licenses and notices. The [Creative Commons Attribution 4.0 International
+terms](https://creativecommons.org/licenses/by/4.0/legalcode) apply as specified above.
 
 ## Citation
 

README.md

@@ -46,10 +46,9 @@ dataset_info:
 
 </div>
 
-The official GAIA2 **mini / validation** subset, packaged with its published
-filesystem resources for native ARE evaluation. This community distribution keeps
-all scenario and resource bytes unchanged. **32 scenarios per capability**:
-Execution · Search · Adaptability · Time · Ambiguity.
+GAIA2-mini packages the official GAIA2 **mini / validation** subset with its
+published filesystem resources for native ARE evaluation. **160 scenarios,
+32 per capability**: Execution · Search · Adaptability · Time · Ambiguity.
 
 ## 🗂️ What's inside
 
@@ -92,7 +91,7 @@ tasks = load_dataset("zhangdw/GAIA2-mini", "mini", split="validation")
 Each row's `data` field is a native ARE scenario JSON string; `tasks.jsonl` indexes
 the original row order. To use the downloaded resources, set `DEMO_FS_PATH` to the
 absolute path of `GAIA2-mini/filesystem/demo_filesystem` before starting ARE.
-Install ARE and configure your agent and judge models separately.
+Install ARE and configure your agent and judge models.
 
 </details>
 
@@ -103,18 +102,17 @@ and [filesystem](https://huggingface.co/datasets/meta-agents-research-environmen
 [Revisions & validation](provenance.json) · [Checksums](data-manifest.json) · [Licenses & citation](NOTICE.md).
 
 All 160 scenarios passed ARE 1.1.0 loading and oracle preprocessing.
-Four upstream file references are unavailable; see
-[known resource gaps](known-missing-files.json). The upstream authors request that
-this evaluation data not be used for training.
+The [resource gap report](known-missing-files.json) lists four missing upstream
+file references. The upstream authors request that this data be used for evaluation.
 
 ## Related project · Bara
 
 **[Bara — Better ARE ReAct Agent](https://github.com/zhangdw156/Bara)** collects ARE ReAct source code for
 studying **recursive self-improvement (RSI) on GAIA2**.
-It includes five research reimplementations inspired by **AutoSaddler,
-ActiveSaddler, ParetoSaddler (two variants) and HarnessMM**, drawn from public code
-and paper appendices. Each source can be studied, modified and evaluated with the
-`bara` CLI; its implementation notes describe coverage and differences from the paper.
+Its five research reimplementations draw on public code and paper appendices from
+**AutoSaddler, ActiveSaddler, ParetoSaddler (two variants) and HarnessMM**.
+Each agent has a source directory you can install, modify and evaluate with the
+`bara` CLI.
 
 [Browse sources & papers →](https://github.com/zhangdw156/Bara#collection) ·
 [Evaluation guide →](https://github.com/zhangdw156/Bara/blob/main/docs/EVALUATION.md)

data-manifest.json

@@ -2,7 +2,7 @@
   "schema_version": 1,
   "source_snapshot_complete": true,
   "files_count": 303,
-  "total_bytes": 421727536,
+  "total_bytes": 421727216,
   "excluded": [
     "data-manifest.json",
     "Hub-managed .gitattributes",
@@ -11,13 +11,13 @@
   "files": [
     {
       "path": "NOTICE.md",
-      "size": 3288,
-      "sha256": "392df90abc6b20da4bda06dee1cefb2feb70c8c774270da6da0caf59e4981932"
+      "size": 3131,
+      "sha256": "621aa095acbb858962cec1282d22cd440eb237bd29f9ef0b11622eadf8194b23"
     },
     {
       "path": "README.md",
-      "size": 4415,
-      "sha256": "0cae65d3dd271e9e0188c26016ecdbfea9206730e833c03758584fbe4e59f03d"
+      "size": 4252,
+      "sha256": "2c956f5af7d658b6bbebafab23ee6177f1ad2c44f1d93c7ee6b9e9652fa5a6f0"
     },
     {
       "path": "assets/overview.svg",