PromiseEval - Corporate ESG Promise Verification
Verification of corporate ESG (Environmental, Social, Governance) promises, requiring annotators to identify whether a statement contains a promise, whether supporting evidence exists, the clarity of the promise-evidence pair, and when the promise can be verified. Based on SemEval-2025 Task 6 (PromiseEval).
Configuration Fileconfig.yaml
This Potato config reproduces the annotation task. Save it as config.yaml and run potato start config.yaml to try it.
# PromiseEval - Corporate ESG Promise Verification
# Based on Chen et al., SemEval-2025 Task 6
# Paper: https://aclanthology.org/2025.semeval-1.321/
# Task site / dataset: https://sites.google.com/view/promiseeval/promiseeval
#
# This task verifies promises found in corporate ESG (Environmental,
# Social, Governance) communications. Annotators read a paragraph from
# a corporate report and assess it along four subtasks:
# 1. Promise Identification - does the text contain a promise?
# 2. Supporting Evidence - is there evidence supporting the promise?
# 3. Clarity - how clearly does the evidence relate to the promise?
# 4. Verification Timing - when can the promise be verified?
annotation_task_name: "PromiseEval - Corporate ESG Promise Verification"
task_dir: "."
data_files:
- sample-data.json
item_properties:
id_key: "id"
text_key: "text"
output_annotation_dir: "annotation_output/"
output_annotation_format: "json"
port: 8000
server_name: localhost
annotation_schemes:
- annotation_type: radio
name: promise_identification
description: "Does this paragraph contain a promise (a commitment about future action)?"
labels:
- "Yes"
- "No"
keyboard_shortcuts:
"Yes": "1"
"No": "2"
tooltips:
"Yes": "The paragraph contains a promise or commitment about future action"
"No": "The paragraph does not contain any promise or commitment"
- annotation_type: radio
name: supporting_evidence
description: "Is there supporting evidence for the promise? (Answer only if a promise is present.)"
labels:
- "Yes"
- "No"
keyboard_shortcuts:
"Yes": "3"
"No": "4"
tooltips:
"Yes": "Evidence is provided that supports the identified promise"
"No": "No supporting evidence is provided for the promise"
- annotation_type: radio
name: evidence_clarity
description: "How clearly does the evidence relate to and support the promise?"
labels:
- "Clear"
- "Not Clear"
- "Misleading"
keyboard_shortcuts:
"Clear": "5"
"Not Clear": "6"
"Misleading": "7"
tooltips:
"Clear": "The evidence directly and clearly supports the promise"
"Not Clear": "The relationship between evidence and promise is vague or weak"
"Misleading": "The evidence is potentially misleading with respect to the promise"
- annotation_type: radio
name: verification_timing
description: "When should a reader be able to verify whether the promise is fulfilled?"
labels:
- "Within 2 years"
- "2-5 years"
- "Longer than 5 years"
- "Other"
keyboard_shortcuts:
"Within 2 years": "q"
"2-5 years": "w"
"Longer than 5 years": "e"
"Other": "r"
tooltips:
"Within 2 years": "The promise can be verified within two years"
"2-5 years": "The promise can be verified in two to five years"
"Longer than 5 years": "The promise can be verified more than five years out"
"Other": "Already verifiable, or no specific timeline is given"
- annotation_type: text
name: evidence_reasoning
description: "Optionally note the evidence or reasoning for your assessment."
annotation_instructions: |
You will be shown a paragraph drawn from a corporate ESG (Environmental,
Social, Governance) report along with company context.
Your tasks are:
1. Decide whether the paragraph contains a promise (a commitment about future action).
2. If a promise is present, decide whether supporting evidence is provided.
3. Rate how clearly that evidence relates to and supports the promise
(Clear, Not Clear, or Misleading).
4. Judge when a reader could verify whether the promise was fulfilled.
Optionally record the evidence or reasoning behind your assessment.
html_layout: |
<div style="padding: 15px; max-width: 800px; margin: auto;">
<div style="background: #f0f9ff; border: 1px solid #bae6fd; border-radius: 8px; padding: 16px; margin-bottom: 16px;">
<strong style="color: #0369a1;">Report Paragraph:</strong>
<p style="font-size: 16px; line-height: 1.7; margin: 8px 0 0 0;">{{text}}</p>
</div>
<div style="display: grid; grid-template-columns: 1fr 1fr 1fr; gap: 10px; margin-bottom: 16px;">
<div style="background: #f8fafc; border: 1px solid #e2e8f0; border-radius: 8px; padding: 12px;">
<strong style="color: #475569;">Company:</strong> {{company}}
</div>
<div style="background: #f8fafc; border: 1px solid #e2e8f0; border-radius: 8px; padding: 12px;">
<strong style="color: #475569;">Year:</strong> {{year}}
</div>
<div style="background: #f8fafc; border: 1px solid #e2e8f0; border-radius: 8px; padding: 12px;">
<strong style="color: #475569;">Source:</strong> {{source}}
</div>
</div>
</div>
allow_all_users: true
instances_per_annotator: 50
annotation_per_instance: 2
allow_skip: true
skip_reason_required: false
Sample Datasample-data.json
[
{
"id": "promise_001",
"text": "By 2030, we commit to reducing our absolute Scope 1 and Scope 2 greenhouse gas emissions by 50% against our 2020 baseline. This target has been validated by the Science Based Targets initiative.",
"company": "Northwind Energy",
"year": "2023",
"source": "Annual Sustainability Report"
},
{
"id": "promise_002",
"text": "Our total energy consumption decreased by 4.1% year over year, reflecting efficiency upgrades completed across three manufacturing sites during the reporting period.",
"company": "Meridian Manufacturing",
"year": "2024",
"source": "ESG Data Supplement"
}
]
// ... and 8 more itemsGet This Design
Clone or download from the repository
Quick start:
git clone https://github.com/davidjurgens/potato-showcase.git cd potato-showcase/semeval/2025/task06-promiseeval potato start config.yaml
Dataset & paper
Chen et al., SemEval-2025 Task 6
Citation (BibTeX)
@inproceedings{chen-etal-2025-semeval,
title = "{S}em{E}val-2025 Task 6: Multinational, Multilingual, Multi-Industry Promise Verification",
author = {Chen, Chung-Chi and Seki, Yohei and Shu, Hakusen and Lhuissier, Ana{\"i}s and Kang, Juyeon and Lee, Hanwool and Day, Min-Yuh and Takamura, Hiroya},
booktitle = "Proceedings of the 19th International Workshop on Semantic Evaluation (SemEval-2025)",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
pages = "2461--2471"
}Details
Annotation Types
Domain
Use Cases
Tags
Found an issue or want to improve this design?
Open an IssueRelated Designs
Multilingual and Crosslingual Fact-Checked Claim Retrieval
Fact-checked claim retrieval task requiring annotators to judge whether a previously fact-checked claim matches a claim from a social media post, supporting multilingual and crosslingual evaluation. Based on SemEval-2025 Task 7.
BRAINTEASER - Commonsense-Defying QA
Lateral thinking and commonsense-defying question answering task requiring annotators to select answers to brain teasers that defy default commonsense assumptions and provide explanations. Based on SemEval-2024 Task 9 (BRAINTEASER).
Check-COVID: Fact-Checking COVID-19 News Claims
Fact-checking COVID-19 news claims. Annotators verify claims against evidence, identify supporting/refuting spans, and provide verdicts with explanations. Based on the Check-COVID dataset targeting misinformation during the pandemic.