Dataset Viewer
Auto-converted to Parquet Duplicate
id
stringclasses
5 values
query
stringclasses
5 values
query_zhtw
stringclasses
5 values
tools
stringclasses
3 values
think
stringclasses
5 values
answer
stringclasses
5 values
messages
listlengths
3
3
24443d31d9a445adac133c40acf491ee
Search for bills proposed in the 11th Legislative Yuan that contain 'Artificial Intelligence' in their name, particularly looking for proposals where legislator Lai Jui-lung (賴瑞隆) was involved.
查詢第 11 屆立法院中,議案名稱含有『人工智慧』的法律提案,並篩選出立法委員賴瑞隆參與提案的議案狀態。
[{"name": "query_rows", "description": "讀取已 normalise 資料集的實際列;支援聚合查詢(v1.11.2+)。\n\n參數:\n dataset_id: 資料集 ID\n where: 原生 SQL WHERE 條件(不含 WHERE 關鍵字)。欄位名稱含中文/特殊字\n 元時請用雙引號包起來。例: \"\"年度\" = '113'\", \"agency ILIKE '%板橋%'\"\n columns: 投影欄位清單,**接受 list 或逗號分隔字串兩種格式**(v1.13.7+)。\n 可帶聚合表達式如 \"COUNT(*) AS ...
根據立法院 fast-path 提示,查詢立法院議案(ly-bills)時不需要先搜尋資料集,可以直接使用 `query_rows` 工具。使用者想查詢第 11 屆立法院中,議案名稱含有『人工智慧』的法律提案,並篩選出立法委員賴瑞隆參與的議案。為了找出這些議案,我將直接查詢 `ly-bills` 資料集,設定過濾條件為 `"屆"='11'` 且 `"議案名稱" ILIKE '%人工智慧%'`。由於人工智慧相關的提案數量預期不會太多,先將所有符合此名稱條件的議案取回,再從中篩選出含有「賴瑞隆」作為提案人或連署人的議案及其狀態。
[{"name": "query_rows", "arguments": {"dataset_id": "ly-bills", "where": "\"屆\"='11' AND \"議案名稱\" ILIKE '%人工智慧%'", "limit": 100}}]
[ { "role": "system", "content": "You are a function calling AI model. You are provided with function signatures within <tools></tools> XML tags. You may call one or more functions to assist with the user query. Don't make assumptions about what values to plug into functions.\n<tools>\n{\"name\": \"query_rows...
e11b01c01c3d40c6870a47675cb07bb9
Please search for invention patent publications related to "folding screen mechanism" or "flexible panel folding", and for the most relevant hit, retrieve its full claims and detailed description.
請幫我搜尋與「折疊螢幕機構」或「柔性面板折疊」相關的發明專利,並針對最符合的一筆,獲取其完整的專利申請範圍與實施方式。
[{"name": "search_patents", "description": "以自然語言查詢 TIPO 發明專利**公開案** corpus (data.gov.tw dataset 15992,\nOGDL 商用 OK)。背後用 Qwen3-Embedding-4B 做 semantic search,適合白話 query。\n\n參數:\n query: 自然語言描述,例如「防止電池爆炸的技術」「AI 偵測癌症的方法」\n 「環保可分解包裝設計」。不必用專利術語,用日常語言即可。\n year_from / year_to: 公開年 (西元) 範圍,例 year_from=2024\n ...
使用者想要尋找與「折疊螢幕機構」或「柔性面板折疊」相關的發明專利,並獲取最符合的一筆之完整專利內容。我需要先使用 `search_patents` 工具進行語意搜尋,設定 query 為「折疊螢幕機構 柔性面板折疊」,並限制回傳數量以利後續挑選最符合的專利。取得搜尋結果後,下一步才能使用 `get_patent_body` 取得該專利的詳細實施方式與申請範圍。因此,第一步先執行專利搜尋。
[{"name": "search_patents", "arguments": {"query": "折疊螢幕機構 柔性面板折疊", "limit": 5}}]
[ { "role": "system", "content": "You are a function calling AI model. You are provided with function signatures within <tools></tools> XML tags. You may call one or more functions to assist with the user query. Don't make assumptions about what values to plug into functions.\n<tools>\n{\"name\": \"search_pat...
66df6528baba4d19815f9c2ef2ab0489
I would like to search for all legal bills proposed by the 11th term legislator Hsu Chiao-hsin that contain "fraud" or "scam" in the bill name.
我想查詢第 11 屆立委徐巧芯所提出,且提案名稱包含「詐欺」或「詐騙」的所有法律提案。
[{"name": "query_rows", "description": "讀取已 normalise 資料集的實際列;支援聚合查詢(v1.11.2+)。\n\n參數:\n dataset_id: 資料集 ID\n where: 原生 SQL WHERE 條件(不含 WHERE 關鍵字)。欄位名稱含中文/特殊字\n 元時請用雙引號包起來。例: \"\"年度\" = '113'\", \"agency ILIKE '%板橋%'\"\n columns: 投影欄位清單,**接受 list 或逗號分隔字串兩種格式**(v1.13.7+)。\n 可帶聚合表達式如 \"COUNT(*) AS ...
根據立法院 fast-path 說明,查詢立法院議案(法律提案)應直接使用 `query_rows` 工具查詢 `ly-bills` 資料集,而不需要先進行 `search_datasets`。使用者要查詢第 11 屆立委徐巧芯提出且名稱包含「詐欺」或「詐騙」的提案。因此,我們設定 `dataset_id` 為 "ly-bills",並在 `where` 條件中過濾 `"屆" = '11'`、`"提案人" ILIKE '%徐巧芯%'`,以及 `("議案名稱" ILIKE '%詐欺%' OR "議案名稱" ILIKE '%詐騙%')`。欄位名稱包含中文,因此依說明使用雙引號包覆。
[{"name": "query_rows", "arguments": {"dataset_id": "ly-bills", "where": "\"屆\" = '11' AND \"提案人\" ILIKE '%徐巧芯%' AND (\"議案名稱\" ILIKE '%詐欺%' OR \"議案名稱\" ILIKE '%詐騙%')", "limit": 100}}]
[ { "role": "system", "content": "You are a function calling AI model. You are provided with function signatures within <tools></tools> XML tags. You may call one or more functions to assist with the user query. Don't make assumptions about what values to plug into functions.\n<tools>\n{\"name\": \"query_rows...
38327be24a8c48819a5ac9a82c56939b
Search for all awarded government procurement projects won by Chunghwa Telecom Co., Ltd. in 2024, sorted by the award price in descending order, and list the top 10 results.
查詢中華電信股份有限公司在 2024 年所有已決標的政府採購案,並依決標金額從高到低排序,列出前 10 筆。
[{"name": "query_rows", "description": "讀取已 normalise 資料集的實際列;支援聚合查詢(v1.11.2+)。\n\n參數:\n dataset_id: 資料集 ID\n where: 原生 SQL WHERE 條件(不含 WHERE 關鍵字)。欄位名稱含中文/特殊字\n 元時請用雙引號包起來。例: \"\"年度\" = '113'\", \"agency ILIKE '%板橋%'\"\n columns: 投影欄位清單,**接受 list 或逗號分隔字串兩種格式**(v1.13.7+)。\n 可帶聚合表達式如 \"COUNT(*) AS ...
根據系統提示中的「政府採購 / 廠商查詢 fast-path」,當查詢包含「政府採購、廠商、決標、得標」等關鍵字時,應直接使用 `query_rows` 查詢 `pcc-tender` 資料集,跳過 `search_datasets`。 使用者希望查詢「中華電信股份有限公司」在「2024 年」所有「已決標」的採購案,並依「決標金額從高到低排序」,列出「前 10 筆」。 因此,我們設定參數如下: - `dataset_id`: "pcc-tender" - `where`: "announcement_type='決標公告' AND companies ILIKE '%中華電信股份有限公司%' AND date BETWEEN '...
[{"name": "query_rows", "arguments": {"dataset_id": "pcc-tender", "where": "announcement_type='決標公告' AND companies ILIKE '%中華電信股份有限公司%' AND date BETWEEN '2024-01-01' AND '2024-12-31'", "order_by": "CAST(award_price AS BIGINT) DESC", "limit": 10, "columns": ["agency", "tender_name", "award_price", "date", "companies"]}}...
[ { "role": "system", "content": "You are a function calling AI model. You are provided with function signatures within <tools></tools> XML tags. You may call one or more functions to assist with the user query. Don't make assumptions about what values to plug into functions.\n<tools>\n{\"name\": \"query_rows...
fa6b66298015437f94b2dd38bdf318b6
"I want to search for the drug license of \"Panadol\" and query its drug label for contraindications(...TRUNCATED)
"我想搜尋「普拿疼」的藥品許可證,並查詢其仿單中有關「肝衰竭」或「妊(...TRUNCATED)
"[{\"name\": \"search_drug\", \"description\": \"衛福部食藥署 全部藥品許可證 search (da(...TRUNCATED)
"為了完成使用者的請求,我需要採取以下步驟:\n1. 使用 `search_drug` 工具搜(...TRUNCATED)
"[{\"name\": \"search_drug\", \"arguments\": {\"name\": \"普拿疼\", \"active_only\": true}}, {\"n(...TRUNCATED)
[{"role":"system","content":"You are a function calling AI model. You are provided with function sig(...TRUNCATED)
README.md exists but content is empty.
Downloads last month
65