Link dataset to paper and update task categories (#2)

- Link dataset to paper and update task categories (617caba088f5e1da4741a113d9277c0ae4014824)


Co-authored-by: Niels Rogge <nielsr@users.noreply.huggingface.co>

README.md

@@ -1,19 +1,20 @@
 ---
-pretty_name: SGR-Bench
 language:
 - zh
 - en
+license: cc-by-4.0
+size_categories:
+- n<1K
+task_categories:
+- question-answering
+- text-retrieval
+pretty_name: SGR-Bench
 tags:
 - benchmark
 - information-retrieval
 - web-agents
 - question-answering
 - state-gated-retrieval
-license: cc-by-4.0
-task_categories:
-- question-answering
-size_categories:
-- n<1K
 configs:
 - config_name: default
   data_files:
@@ -25,6 +26,8 @@ configs:
 
 # SGR-Bench
 
+This is the official dataset repository for the paper [SGR-Bench: Benchmarking Search Agents on State-Gated Retrieval](https://huggingface.co/papers/2605.22219).
+
 SGR-Bench is a dataset for evaluating **State-Gated Retrieval**. In this setting, the answer-bearing evidence is not exposed in the default state of a website. It becomes available only after the user configures site-internal filters, views, hierarchies, scopes, time windows, or result pages.
 
 This release contains **100 aligned task records** organized into two prompt formulations:
@@ -234,12 +237,12 @@ In this scoring scheme:
 - row-level F1 gives credit only when all fields in an aligned row are correct;
 - pairwise order accuracy checks whether the relative order among shared rows is preserved.
 
-For ordered-table cases, the scoring logic should distinguish clearly between local correctness and structural correctness. A case may preserve many locally correct fields while still failing at the row level or full-answer level.
+For ordered-table tasks, the scoring logic should distinguish clearly between local correctness and structural correctness. A case may preserve many locally correct fields while still failing at the row level or full-answer level.
 
 This distinction matters for diagnosis:
 
 - high `item_f1` with low `row_f1` or low `em` should not be treated as automatic evidence of final formatting failure;
-- such cases should first be checked for `criterion mismatch`, `retrieval dependency not closed`, or upstream `drift`;
+- such these cases should first be checked for `criterion mismatch`, `retrieval dependency not closed`, or upstream `drift`;
 - `final answer composition error` should be used only when the upstream reasoning chain is already basically correct and the answer is corrupted mainly at the final write-out stage.
 
 The canonicalization step is intentionally narrow. It may normalize whitespace, punctuation, capitalization, date formatting, unit formatting, abbreviations, and a small set of task-specific aliases, but it should not fill missing fields, repair factual errors, or merge source-distinct entities.
@@ -257,4 +260,4 @@ The loadable JSONL splits use English task prompts. Chinese counterparts for the
 
 ## Limitations
 
-This is a compact diagnostic evaluation set. It prioritizes relatively stable public sources and structured answers, so it covers less rapidly changing content such as breaking news, live dashboards, or frequently updated public records. This release evaluates final structured outputs and provides retrieval-state annotations, but it does not provide a unified trajectory-level annotation for every intermediate model action.
+This is a compact diagnostic evaluation set. It prioritizes relatively stable public sources and structured answers, so it covers less rapidly changing content such as breaking news, live dashboards, or frequently updated public records. This release evaluates final structured outputs and provides retrieval-state annotations, but it does not provide a unified trajectory-level annotation for every intermediate model action.