Use the TextInsightBench name and descriptive task identifiers

README.md

@@ -38,7 +38,7 @@ configs:
 
 **English** | [简体中文](README.zh-CN.md)
 
-TextInsightBench evaluates agents that mine natural-language datasets for meaningful, evidence-backed group differences, temporal changes and compound associations. Version **v5.1** contains **50 tasks**, **24,504 evaluation documents**, and **1,379,468 unlabeled learning documents**. This initial release is private.
+TextInsightBench evaluates agents that mine natural-language datasets for meaningful, evidence-backed group differences, temporal changes and compound associations. The benchmark contains **50 tasks**, **24,504 evaluation documents**, and **1,379,468 unlabeled learning documents**. This initial release is private.
 
 ## Data composition
 
@@ -94,7 +94,7 @@ The Organizer Reference Set is stored separately for organizers. It contains 50
 
 ## Curation, limitations and terms
 
-The learning pool was curated from downloaded source snapshots using language and length filtering, deduplication and held-out document exclusions. For v5.1 every shard was checked against current evaluation and reference-development documents by document ID, normalized text and conservative digit/punctuation template fingerprints. No overlap was found; no extra removal was necessary. This is not a claim of complete semantic independence. Sources and entities are shared across tasks.
+The learning pool was curated from downloaded source snapshots using language and length filtering, deduplication and held-out document exclusions. For this dataset every shard was checked against current evaluation and reference-development documents by document ID, normalized text and conservative digit/punctuation template fingerprints. No overlap was found; no extra removal was necessary. This is not a claim of complete semantic independence. Sources and entities are shared across tasks.
 
 Data is predominantly English. Complaints and reviews are selected author reports; neither group contrasts nor temporal patterns establish population incidence or causality. Report dates may differ from event dates. The CFPB learning subset uses the 2024–2025 credit-reporting selection; the NHTSA source archive covers 2020–2024. Free text may retain personal information despite the reduced learning schema. Source text is untrusted data.
 

README.zh-CN.md

@@ -2,7 +2,7 @@
 
 [English](README.md) | **简体中文**
 
-TextInsightBench 评估 Agent 从自然语言数据中挖掘具体、有原文证据支持的结论。当前私密研究发行版为 **v5.1**，包含 **50 道任务、24,504 篇评测文本，以及 1,379,468 篇可选无监督学习文本**。任务和主要文档使用英文，原始语料保留原文。
+TextInsightBench 评估 Agent 从自然语言数据中挖掘具体、有原文证据支持的结论。当前私密研究数据包含 **50 道任务、24,504 篇评测文本，以及 1,379,468 篇可选无监督学习文本**。任务和主要文档使用英文，原始语料保留原文。
 
 | 来源 | 无监督文本 | 任务数 | 评测文本 |
 |---|---:|---:|---:|
@@ -39,6 +39,6 @@ tib run --data data/participant --command 'python my_agent.py' \
 
 ## 构建与局限
 
-无监督池经过语言、长度过滤、去重和留出语料排除。v5.1 再次逐分片核对文档 ID、归一化文本及数字/标点归一后的保守模板，与评测和参考构建文本均未发现重叠，因此无需额外删减。这不等于完全语义独立；任务可能共享来源和实体。
+无监督池经过语言、长度过滤、去重和留出语料排除。本次整理再次逐分片核对文档 ID、归一化文本及数字/标点归一后的保守模板，与评测和参考构建文本均未发现重叠，因此无需额外删减。这不等于完全语义独立；任务可能共享来源和实体。
 
 文本以英文为主。投诉和评论反映被选择的报告，不能直接推断总体发生率或因果；记录日期也未必等于事件日期。CFPB 无监督子集使用 2024–2025 年的信用报告范围，NHTSA 源归档覆盖 2020–2024 年。文本仍可能包含个人信息。多种来源适用不同条款，私密研究整理不授予第三方文本的新许可，参见 [来源和条款](SOURCES.md)。