Spaces:
Build error
Build error
File size: 2,213 Bytes
c32ee7d |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 |
dataset: asset
subset: ratings
templates:
09b2a13b-cba6-4473-8a46-3fa24be71ce2: !Template
answer_choices: null
id: 09b2a13b-cba6-4473-8a46-3fa24be71ce2
jinja: "{% set questions= [ \"Does the second sentence better convey the information?\"\
, \"Is the second sentence more fluent?\", \"Is the second sentence easier\
\ to understand?\"] %}\n\nFirst sentence: {{original}}\n\nSecond sentence: {{simplification}}\n\
\n{{questions[aspect]}} \n\n|||\n\n{% if rating > 50 %}\n Yes\n{% else %}\n\
\ No\n{% endif %}"
metadata: !TemplateMetadata
choices_in_prompt: null
metrics: []
original_task: false
name: asset_ratings1
reference: Taking questions from the original paper, we use rating to establish
a binary classification problem.
47142040-4121-4144-98b9-61cb5cbb1313: !Template
answer_choices: null
id: 47142040-4121-4144-98b9-61cb5cbb1313
jinja: 'First sentence: {{original}}
Second sentence: {{simplification}}
I am scoring these simplification exercises. How easier to read is the second
sentence on a scale from 0 (harder to read) to 100 (easier to read)?
|||
{{rating}}'
metadata: !TemplateMetadata
choices_in_prompt: null
metrics: []
original_task: true
name: asset_ratings3
reference: Prompt model to rate how simplified the sentence is in the general
sense, instead of an particular aspect.
d2bed959-29ab-4962-a106-dc91c00f3f03: !Template
answer_choices: null
id: d2bed959-29ab-4962-a106-dc91c00f3f03
jinja: "{% set statements= [ \"the second sentence expresses the underlying meaning\
\ the best.\", \"the second sentence is more fluent.\", \"the second sentence\
\ is easier to read and understand.\"] %}\n\nFirst sentence: {{original}}\n\n\
Second sentence: {{simplification}}\n\nRate the following statement from 0 (strongly\
\ disagree) to 100 (strongly agree): {{statements[aspect]}} \n\n|||\n\n{{rating}}"
metadata: !TemplateMetadata
choices_in_prompt: null
metrics: []
original_task: true
name: asset_ratings2
reference: Require the model to output the rating
|