Hugging Face's logo

Datasets:
Lexmount
/
LexBench-Browser

Tasks:
Question Answering
Text Generation
Reinforcement Learning
Languages:
Chinese
English
Size:
n<1K
Tags:
browser-agent
web-agent
benchmark
evaluation
automation
License:
Dataset card Files
xet
 Community
1
waple0820 commited on
Commit
e7da01c
·
verified ·
1 Parent(s): 11d8e0d

Update LexBench-Browser public dataset card after task removal

Browse files

Align public dataset card with the 208-task public release after removing task ids 2021 and 2022.

Files changed (1) hide show
  1. README.md +16 -12
README.md CHANGED
@@ -21,7 +21,7 @@ tags:
21
  # LexBench-Browser
22
 
23
  LexBench-Browser is a public browser-agent dataset for evaluating agents on real-web workflows.
24
- The v1.0 snapshot contains 210 no-login tasks across 107 distinct websites, with Chinese and
25
  English instructions, task-level reference steps, key points, common mistakes, scoring rubrics,
26
  and robustness tags.
27
 
@@ -53,6 +53,7 @@ LexBench-Browser/
53
  |-- task.jsonl
54
  |-- task_global.jsonl
55
  |-- task_lexmount.jsonl
 
56
  `-- VERSION_HISTORY.md
57
  ```
58
 
@@ -60,9 +61,10 @@ Splits:
60
 
61
  | Split | File | Tasks | Notes |
62
  | --- | --- | ---: | --- |
63
- | `All` | `task.jsonl` | 210 | Default public v1.0 split |
64
- | `global` | `task_global.jsonl` | 92 | Global-region task subset |
65
  | `lexmount` | `task_lexmount.jsonl` | 118 | Lexmount-region task subset |
 
66
 
67
  ## Fields
68
 
@@ -89,14 +91,14 @@ Language:
89
 
90
  | Language | Tasks |
91
  | --- | ---: |
92
- | `zh` | 137 |
93
  | `en` | 73 |
94
 
95
  Reasoning type:
96
 
97
  | Reasoning type | Tasks |
98
  | --- | ---: |
99
- | `single_step` | 117 |
100
  | `multi_step` | 70 |
101
  | `deep_analysis` | 23 |
102
 
@@ -104,13 +106,15 @@ Domain:
104
 
105
  | Domain | Tasks |
106
  | --- | ---: |
107
- | `finance_gaming` | 44 |
108
- | `video_platform` | 42 |
109
- | `tools_education` | 40 |
110
- | `general` | 34 |
111
- | `social_lifestyle` | 26 |
112
- | `ecommerce` | 23 |
113
- | `gaming` | 1 |
 
 
114
 
115
  ## Robustness Tags
116
 
 
21
  # LexBench-Browser
22
 
23
  LexBench-Browser is a public browser-agent dataset for evaluating agents on real-web workflows.
24
+ The v1.0 snapshot contains 208 no-login tasks across 107 distinct websites, with Chinese and
25
  English instructions, task-level reference steps, key points, common mistakes, scoring rubrics,
26
  and robustness tags.
27
 
 
53
  |-- task.jsonl
54
  |-- task_global.jsonl
55
  |-- task_lexmount.jsonl
56
+ |-- task_sample50.jsonl
57
  `-- VERSION_HISTORY.md
58
  ```
59
 
 
61
 
62
  | Split | File | Tasks | Notes |
63
  | --- | --- | ---: | --- |
64
+ | `All` | `task.jsonl` | 208 | Default public v1.0 split |
65
+ | `global` | `task_global.jsonl` | 90 | Global-region task subset |
66
  | `lexmount` | `task_lexmount.jsonl` | 118 | Lexmount-region task subset |
67
+ | `sample50` | `task_sample50.jsonl` | 50 | Small sample subset |
68
 
69
  ## Fields
70
 
 
91
 
92
  | Language | Tasks |
93
  | --- | ---: |
94
+ | `zh` | 135 |
95
  | `en` | 73 |
96
 
97
  Reasoning type:
98
 
99
  | Reasoning type | Tasks |
100
  | --- | ---: |
101
+ | `single_step` | 115 |
102
  | `multi_step` | 70 |
103
  | `deep_analysis` | 23 |
104
 
 
106
 
107
  | Domain | Tasks |
108
  | --- | ---: |
109
+ | `media_entertainment` | 48 |
110
+ | `education_research` | 42 |
111
+ | `gaming` | 25 |
112
+ | `commerce` | 23 |
113
+ | `safety` | 23 |
114
+ | `local_lifestyle` | 19 |
115
+ | `finance` | 17 |
116
+ | `productivity_tools` | 6 |
117
+ | `social_community` | 5 |
118
 
119
  ## Robustness Tags
120