Mix-Ecom: Towards Mixed-Type E-Commerce Dialogues with Complex Domain Rules
Paper • 2509.23836 • Published
Dataset Viewer
The dataset viewer is not available for this split.
Cannot load the dataset split (in streaming mode) to extract the first rows.
Error code: StreamingRowsError
Exception: CastError
Message: Couldn't cast
task_id: string
item_id: string
asr_result: string
item_title: string
item_leaf_category_name: string
photo_id: string
start_time: string
end_time: string
item_price: string
is_support_7d_back: string
item_promise_deliver_time: string
category_prop: struct<类型: string, 制冷方式: string, 容量: string, 能效: string, 屏幕尺寸: string, � (... 594 chars omitted)
child 0, 类型: string
child 1, 制冷方式: string
child 2, 容量: string
child 3, 能效: string
child 4, 屏幕尺寸: string
child 5, 处理器: string
child 6, 存储容量: string
child 7, 材质: string
child 8, 保温时长: string
child 9, 风格: string
child 10, 功效: string
child 11, 质地: string
child 12, 适用肤质: string
child 13, 降噪功能: string
child 14, 续航时间: string
child 15, 车架材质: string
child 16, 轮径尺寸: string
child 17, 避震类型: string
child 18, 适用场景: string
child 19, 香型: string
child 20, 适用人群: string
child 21, 酒精度: string
child 22, 原料: string
child 23, 年份: string
child 24, 尺寸: string
child 25, 轮组: string
child 26, 开合方式: string
child 27, 功率: string
child 28, 功能: string
child 29, 风嘴类型: string
child 30, 长度: string
child 31, 调性: string
child 32, 锅底类型: string
child 33, 涂层: string
child 34, 口味: string
child 35, 甜度: string
child 36, 包装: string
is_fresh_item: string
item_weight: string
qa_pair: struct<question: string, answer: string>
child 0, question: string
child 1, answer: string
snapshot_id: string
database: struct<order_id: string, item_id: string, logistics_id: string, user_id: string, shop_id: string>
chil
...
hild 38, 酒精度: list<item: string>
child 0, item: string
child 39, 原料: list<item: string>
child 0, item: string
child 40, 年份: list<item: string>
child 0, item: string
child 41, 轮组: list<item: string>
child 0, item: string
child 42, 开合方式: list<item: string>
child 0, item: string
child 43, 长度: list<item: string>
child 0, item: string
child 44, 调性: list<item: string>
child 0, item: string
child 45, 图案样式: list<item: string>
child 0, item: string
child 46, 特殊功能: list<item: string>
child 0, item: string
child 47, 适用机型: list<item: string>
child 0, item: string
child 48, 款式: list<item: string>
child 0, item: string
child 49, 适用季节: list<item: string>
child 0, item: string
question_type: list<item: string>
child 0, item: string
trajectory: string
conversation: string
key_answer: list<item: string>
child 0, item: string
score: struct<key_answer_score: double, database_score: int64, flag_remark: int64, flag_l_o: int64>
child 0, key_answer_score: double
child 1, database_score: int64
child 2, flag_remark: int64
child 3, flag_l_o: int64
database_gt: struct<remark_info: string>
child 0, remark_info: string
label: int64
user_profile_st1: struct<reason: string, image_verification: string, solution: string, mood: string>
child 0, reason: string
child 1, image_verification: string
child 2, solution: string
child 3, mood: string
image_paths: list<item: string>
child 0, item: string
to
{'task_id': Value('string'), 'snapshot_id': Value('string'), 'database': {'order_id': Value('string'), 'item_id': Value('string'), 'logistics_id': Value('string'), 'shop_owner_id': Value('string'), 'user_id': Value('string')}, 'user_profile_st1': {'reason': Value('string'), 'image_verification': Value('string'), 'solution': Value('string'), 'mood': Value('string')}, 'user_profile': Value('string'), 'first_query': Value('string'), 'question_type': List(Value('string')), 'user_address': Value('string'), 'trajectory': Value('string'), 'conversation': Value('string'), 'key_answer': List(Value('string')), 'label': Value('int64'), 'database_gt': {'order_info': {'order_id': Value('string'), 'item_id': Value('string'), 'item_num': Value('string'), 'item_title': Value('string'), 'send_address': Value('string'), 'receive_address': Value('string'), 'pay_time': Value('string'), 'payment_fee': Value('string'), 'order_state': Value('string')}, 'logistics_info': {'order_id': Value('string'), 'item_id': Value('string'), 'logistics_id': Value('string'), 'logistics_state': Value('string'), 'receive_address': Value('string'), 'logistics_time': {'collect_time': Value('string'), 'sign_time': Value('string'), 'sign_off_time': Value('string'), 'delivery_time': Value('string'), 'refund_time': Value('string')}, 'send_address': Value('string'), 'courier_brand': Value('string')}, 'remark_info': Value('string')}, 'image_paths': List(Value('string'))}
because column names don't match
Traceback: Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 149, in get_rows_or_raise
return get_rows(
dataset=dataset,
...<4 lines>...
column_names=column_names,
)
File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
return func(*args, **kwargs)
File "/src/services/worker/src/worker/utils.py", line 129, in get_rows
rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
File "/src/services/worker/src/worker/utils.py", line 489, in safe_iter
yield from ds.decode(False) if ds.features else ds
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2818, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2355, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2380, in _iter_arrow
for key, pa_table in self.ex_iterable._iter_arrow():
~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 536, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 419, in _iter_arrow
for key, pa_table in self.generate_tables_fn(**gen_kwags):
~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 343, in _generate_tables
self._cast_table(pa_table, json_field_paths=json_field_paths),
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
pa_table = table_cast(pa_table, self.info.features.arrow_schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2369, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2297, in cast_table_to_schema
raise CastError(
...<3 lines>...
)
datasets.table.CastError: Couldn't cast
task_id: string
item_id: string
asr_result: string
item_title: string
item_leaf_category_name: string
photo_id: string
start_time: string
end_time: string
item_price: string
is_support_7d_back: string
item_promise_deliver_time: string
category_prop: struct<类型: string, 制冷方式: string, 容量: string, 能效: string, 屏幕尺寸: string, � (... 594 chars omitted)
child 0, 类型: string
child 1, 制冷方式: string
child 2, 容量: string
child 3, 能效: string
child 4, 屏幕尺寸: string
child 5, 处理器: string
child 6, 存储容量: string
child 7, 材质: string
child 8, 保温时长: string
child 9, 风格: string
child 10, 功效: string
child 11, 质地: string
child 12, 适用肤质: string
child 13, 降噪功能: string
child 14, 续航时间: string
child 15, 车架材质: string
child 16, 轮径尺寸: string
child 17, 避震类型: string
child 18, 适用场景: string
child 19, 香型: string
child 20, 适用人群: string
child 21, 酒精度: string
child 22, 原料: string
child 23, 年份: string
child 24, 尺寸: string
child 25, 轮组: string
child 26, 开合方式: string
child 27, 功率: string
child 28, 功能: string
child 29, 风嘴类型: string
child 30, 长度: string
child 31, 调性: string
child 32, 锅底类型: string
child 33, 涂层: string
child 34, 口味: string
child 35, 甜度: string
child 36, 包装: string
is_fresh_item: string
item_weight: string
qa_pair: struct<question: string, answer: string>
child 0, question: string
child 1, answer: string
snapshot_id: string
database: struct<order_id: string, item_id: string, logistics_id: string, user_id: string, shop_id: string>
chil
...
hild 38, 酒精度: list<item: string>
child 0, item: string
child 39, 原料: list<item: string>
child 0, item: string
child 40, 年份: list<item: string>
child 0, item: string
child 41, 轮组: list<item: string>
child 0, item: string
child 42, 开合方式: list<item: string>
child 0, item: string
child 43, 长度: list<item: string>
child 0, item: string
child 44, 调性: list<item: string>
child 0, item: string
child 45, 图案样式: list<item: string>
child 0, item: string
child 46, 特殊功能: list<item: string>
child 0, item: string
child 47, 适用机型: list<item: string>
child 0, item: string
child 48, 款式: list<item: string>
child 0, item: string
child 49, 适用季节: list<item: string>
child 0, item: string
question_type: list<item: string>
child 0, item: string
trajectory: string
conversation: string
key_answer: list<item: string>
child 0, item: string
score: struct<key_answer_score: double, database_score: int64, flag_remark: int64, flag_l_o: int64>
child 0, key_answer_score: double
child 1, database_score: int64
child 2, flag_remark: int64
child 3, flag_l_o: int64
database_gt: struct<remark_info: string>
child 0, remark_info: string
label: int64
user_profile_st1: struct<reason: string, image_verification: string, solution: string, mood: string>
child 0, reason: string
child 1, image_verification: string
child 2, solution: string
child 3, mood: string
image_paths: list<item: string>
child 0, item: string
to
{'task_id': Value('string'), 'snapshot_id': Value('string'), 'database': {'order_id': Value('string'), 'item_id': Value('string'), 'logistics_id': Value('string'), 'shop_owner_id': Value('string'), 'user_id': Value('string')}, 'user_profile_st1': {'reason': Value('string'), 'image_verification': Value('string'), 'solution': Value('string'), 'mood': Value('string')}, 'user_profile': Value('string'), 'first_query': Value('string'), 'question_type': List(Value('string')), 'user_address': Value('string'), 'trajectory': Value('string'), 'conversation': Value('string'), 'key_answer': List(Value('string')), 'label': Value('int64'), 'database_gt': {'order_info': {'order_id': Value('string'), 'item_id': Value('string'), 'item_num': Value('string'), 'item_title': Value('string'), 'send_address': Value('string'), 'receive_address': Value('string'), 'pay_time': Value('string'), 'payment_fee': Value('string'), 'order_state': Value('string')}, 'logistics_info': {'order_id': Value('string'), 'item_id': Value('string'), 'logistics_id': Value('string'), 'logistics_state': Value('string'), 'receive_address': Value('string'), 'logistics_time': {'collect_time': Value('string'), 'sign_time': Value('string'), 'sign_off_time': Value('string'), 'delivery_time': Value('string'), 'refund_time': Value('string')}, 'send_address': Value('string'), 'courier_brand': Value('string')}, 'remark_info': Value('string')}, 'image_paths': List(Value('string'))}
because column names don't matchNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
Mix-ECom Dataset
Data files for Mix-ECom, a benchmark for evaluating LLM-based agents on e-commerce customer service tasks. See the code repository and paper for the agent, evaluation harness, and usage instructions.
Contents
.
├── eval/
│ ├── eval_gt_a.json # After-sales split (91 samples, text + image)
│ ├── eval_gt_d.json # During-sales split (100 samples, text + video)
│ └── eval_gt_l.json # Logistics split (108 samples, text)
├── database/ # Simulated e-commerce database (shop / order / item / logistics / user / product)
├── images/ # Product & complaint images referenced by the after-sales split
└── videos/ # Live-stream clips referenced by the during-sales split
All identifiers (user IDs, order IDs, addresses, phone numbers, etc.) in this dataset are synthetically generated and do not correspond to real individuals or transactions.
Usage
Download and place under data/ in the Mix-ECom code repository:
hf download zhourax977/Mix-ECom --repo-type dataset --local-dir data
License
CC BY 4.0 — see DATA_LICENSE in the code repository.
Citation
@misc{zhou2025mixecommixedtypeecommercedialogues,
title={Mix-Ecom: Towards Mixed-Type E-Commerce Dialogues with Complex Domain Rules},
author={Chenyu Zhou and Xiaoming Shi and Hui Qiu and Xiawu Zheng and Haitao Leng and Yankai Jiang and Shaoguo Liu and Tingting Gao and Rongrong Ji},
year={2025},
eprint={2509.23836},
archivePrefix={arXiv},
primaryClass={cs.AI},
url={https://arxiv.org/abs/2509.23836},
}
- Downloads last month
- 18