id stringlengths 17 21 | artifact_type stringclasses 5
values | category stringclasses 4
values | prompt stringlengths 302 4.4k | platform_instructions stringclasses 6
values | rubric_json stringlengths 30.6k 249k | evaluation_prompt stringclasses 4
values | acceptance_groups_json stringlengths 127 803 | inputs_manifest_json stringlengths 352 6.33k | task_directory stringlengths 28 32 | inputs_directory stringlengths 35 39 |
|---|---|---|---|---|---|---|---|---|---|---|
animations-videos-021 | dynamic_media | animations-videos | # State changes when a series-circuit switch closes
Using the materials in `input/`, create **a 15-second animated SVG** for readers new to the subject. Write the deliverable in English.
See `input/reading-scope.md` for the reading scope. Use the provided facts and assets; do not add unsupported product, statistical,... | Deliver only the artifacts and companion files requested in the task statement. The task statement specifies formats, counts, dimensions, duration, editability, previews, source attribution and companion documentation.
Delivered files must actually open, render or run with the appropriate tools, with valid references t... | [
{
"id": "IF01",
"dimension": "Task Fulfillment / Compliance",
"criterion": "Requirement: the circuit must consist of an ideal DC source, a switch and two series resistors.\nVerification method: check whether the circuit diagram contains an ideal DC source, a switch, R1 and R2, and whether R1 and R2 are ... | Evaluate artifact/. Read prompt.md, platform_instructions.txt and every rubric.json criterion with its answers. rubric.json is the complete authority for Agent item requirements, applicability boundaries and rating anchors; use the supplied materials to verify facts. Apply task-wide scope limits to all related items wi... | {"G01": ["IF01"], "G02": ["IF02", "IF03"], "G03": ["IF04"], "G04": ["IF05"], "G05": ["IF06", "IF07", "IF08"], "G06": ["IF09", "IF11"], "G07": ["IF10", "IF12"], "G08": ["IF13"], "G09": ["IF14"], "G10": ["IF15"], "G11": ["IF16", "IF17"], "G12": ["IF18", "IF19"], "G13": ["IF20", "IF21"], "G14": ["IF22"], "G15": ["IF23"], ... | {
"files": [
{
"path": "input/image_sources.json",
"bytes": 2074,
"sha256": "7ce063b79969321cc80c6faa210f258b7e4f84aa9d58064e8d1c40f6933117c0"
},
{
"path": "input/images/series_wiki_01.svg",
"bytes": 6293,
"sha256": "088b37f61353a749c747d8470ee9905dd374b20ec7bb30fdf2701... | data/tasks/animations-videos-021 | data/tasks/animations-videos-021/inputs |
animations-videos-022 | dynamic_media | animations-videos | "# Gradient descent: iteration and convergence\n\nUsing the materials in `input/`, create **a 20-sec(...TRUNCATED) | "Deliver only the artifacts and companion files requested in the task statement. The task statement (...TRUNCATED) | "[\n {\n \"id\": \"IF01\",\n \"dimension\": \"Task Fulfillment / Compliance\",\n \"criteri(...TRUNCATED) | "Evaluate artifact/. Read prompt.md, platform_instructions.txt and every rubric.json criterion with (...TRUNCATED) | "{\"G01\": [\"IF01\"], \"G02\": [\"IF02\"], \"G03\": [\"IF03\"], \"G04\": [\"IF04\"], \"G05\": [\"IF(...TRUNCATED) | "{\n \"files\": [\n {\n \"path\": \"input/d2l_gd.html\",\n \"bytes\": 188856,\n \(...TRUNCATED) | data/tasks/animations-videos-022 | data/tasks/animations-videos-022/inputs |
animations-videos-023 | dynamic_media | animations-videos | "# Renewable electricity shares in four countries\n\nThe supplied input materials are in `input/work(...TRUNCATED) | "Deliver only the artifacts and companion files requested in the task statement. The task statement (...TRUNCATED) | "[\n {\n \"id\": \"IF01\",\n \"dimension\": \"Task Fulfillment / Compliance\",\n \"criteri(...TRUNCATED) | "Evaluate artifact/. Read prompt.md, platform_instructions.txt and every rubric.json criterion with (...TRUNCATED) | "{\"G01\": [\"IF01\", \"IF02\", \"IF04\"], \"G02\": [\"IF03\", \"IF06\", \"IF07\", \"IF08\", \"IF09\(...TRUNCATED) | "{\n \"files\": [\n {\n \"path\": \"input/workspace/README.md\",\n \"bytes\": 810,\n (...TRUNCATED) | data/tasks/animations-videos-023 | data/tasks/animations-videos-023/inputs |
animations-videos-024 | dynamic_media | animations-videos | "# Three selling points of Raspberry Pi Pico 2\n\nUsing the materials in `input/`, create **a 10-sec(...TRUNCATED) | "Deliver only the artifacts and companion files requested in the task statement. The task statement (...TRUNCATED) | "[\n {\n \"id\": \"IF01\",\n \"dimension\": \"Task Fulfillment / Compliance\",\n \"criteri(...TRUNCATED) | "Evaluate artifact/. Read prompt.md, platform_instructions.txt and every rubric.json criterion with (...TRUNCATED) | "{\"G01\": [\"IF01\"], \"G02\": [\"IF02\"], \"G03\": [\"IF03\", \"IF11\"], \"G04\": [\"IF04\", \"IF0(...TRUNCATED) | "{\n \"files\": [\n {\n \"path\": \"input/figures/pico2_p5_image1.jpeg\",\n \"bytes\":(...TRUNCATED) | data/tasks/animations-videos-024 | data/tasks/animations-videos-024/inputs |
animations-videos-025 | dynamic_media | animations-videos | "# Update an older product animation to Pico 2 W\n\nUpdate the 12-second English Pico 2 animation in(...TRUNCATED) | "Deliver only the artifacts and companion files requested in the task statement. The task statement (...TRUNCATED) | "[\n {\n \"id\": \"IF01\",\n \"dimension\": \"Task Fulfillment / Compliance\",\n \"criteri(...TRUNCATED) | "Evaluate artifact/. Read prompt.md, platform_instructions.txt and every rubric.json criterion with (...TRUNCATED) | "{\"G01\": [\"IF01\", \"IF02\", \"IF05\", \"IF07\", \"IF14\", \"IF25\", \"IF26\", \"IF27\"], \"G02\"(...TRUNCATED) | "{\n \"files\": [\n {\n \"path\": \"workspace/README.md\",\n \"bytes\": 1237,\n \(...TRUNCATED) | data/tasks/animations-videos-025 | data/tasks/animations-videos-025/inputs |
animations-videos-026 | dynamic_media | animations-videos | "# 修正冒泡排序动画的状态与标注\n\n请基于随题 input/workspace/ 中的原始动画(...TRUNCATED) | "仅交付题面要求的产物和配套文件。格式、数量、尺寸、时长、可编辑性、(...TRUNCATED) | "[\n {\n \"id\": \"IF01\",\n \"dimension\": \"Task Fulfillment / Compliance\",\n \"criteri(...TRUNCATED) | "Evaluate artifact/. Read prompt.md, platform_instructions.txt and every rubric.json criterion with (...TRUNCATED) | "{\"G01\": [\"IF01\", \"IF17\", \"IF19\", \"IF20\"], \"G02\": [\"IF02\", \"IF03\", \"IF04\", \"IF05\(...TRUNCATED) | "{\n \"files\": [\n {\n \"path\": \"input/workspace/README.md\",\n \"bytes\": 541,\n (...TRUNCATED) | data/tasks/animations-videos-026 | data/tasks/animations-videos-026/inputs |
animations-videos-027 | dynamic_media | animations-videos | "# Position and velocity of an ideal projectile\n\nUsing the materials in `input/`, create **a 15-se(...TRUNCATED) | "Deliver only the artifacts and companion files requested in the task statement. The task statement (...TRUNCATED) | "[\n {\n \"id\": \"IF01\",\n \"dimension\": \"Task Fulfillment / Compliance\",\n \"criteri(...TRUNCATED) | "Evaluate artifact/. Read prompt.md, platform_instructions.txt and every rubric.json criterion with (...TRUNCATED) | "{\"G01\": [\"IF01\", \"IF02\", \"IF03\", \"IF04\", \"IF05\"], \"G02\": [\"IF06\", \"IF07\", \"IF08\(...TRUNCATED) | "{\n \"files\": [\n {\n \"path\": \"input/image_sources.json\",\n \"bytes\": 1722,\n (...TRUNCATED) | data/tasks/animations-videos-027 | data/tasks/animations-videos-027/inputs |
animations-videos-028 | dynamic_media | animations-videos | "# Your first recording in OBS Studio\n\nUsing the materials in `input/`, create **a 25-second MP4**(...TRUNCATED) | "Deliver only the artifacts and companion files requested in the task statement. The task statement (...TRUNCATED) | "[\n {\n \"id\": \"IF01\",\n \"dimension\": \"Task Fulfillment / Compliance\",\n \"criteri(...TRUNCATED) | "Evaluate artifact/. Read prompt.md, platform_instructions.txt and every rubric.json criterion with (...TRUNCATED) | "{\"G01\": [\"IF01\", \"IF06\"], \"G02\": [\"IF02\"], \"G03\": [\"IF03\"], \"G04\": [\"IF04\"], \"G0(...TRUNCATED) | "{\n \"files\": [\n {\n \"path\": \"input/images/obs_overview_01.png\",\n \"bytes\": 2(...TRUNCATED) | data/tasks/animations-videos-028 | data/tasks/animations-videos-028/inputs |
animations-videos-029 | dynamic_media | animations-videos | "# Create a GitHub Issue in five steps\n\nUsing the materials in `input/`, create **a 12-second GIF*(...TRUNCATED) | "Deliver only the artifacts and companion files requested in the task statement. The task statement (...TRUNCATED) | "[\n {\n \"id\": \"IF01\",\n \"dimension\": \"Task Fulfillment / Compliance\",\n \"criteri(...TRUNCATED) | "Evaluate artifact/. Read prompt.md, platform_instructions.txt and every rubric.json criterion with (...TRUNCATED) | "{\"G01\": [\"IF01\"], \"G02\": [\"IF02\", \"IF03\", \"IF05\"], \"G03\": [\"IF04\", \"IF06\"], \"G04(...TRUNCATED) | "{\n \"files\": [\n {\n \"path\": \"input/github_issue.html\",\n \"bytes\": 262339,\n (...TRUNCATED) | data/tasks/animations-videos-029 | data/tasks/animations-videos-029/inputs |
animations-videos-039 | dynamic_media | animations-videos | "# Why not every full moon produces a lunar eclipse\n\nUsing the NASA materials, create a silent Eng(...TRUNCATED) | "Deliver only the artifacts and companion files requested in the task statement. The task statement (...TRUNCATED) | "[\n {\n \"id\": \"IF01\",\n \"dimension\": \"Task Fulfillment / Compliance\",\n \"criteri(...TRUNCATED) | "Evaluate artifact/. Read prompt.md, platform_instructions.txt and every rubric.json criterion with (...TRUNCATED) | "{\"G01\": [\"IF01\", \"IF10\", \"IF14\"], \"G02\": [\"IF02\", \"IF03\", \"IF09\", \"IF11\", \"IF13\(...TRUNCATED) | "{\n \"files\": [\n {\n \"path\": \"input/eclipses.html\",\n \"bytes\": 324053,\n (...TRUNCATED) | data/tasks/animations-videos-039 | data/tasks/animations-videos-039/inputs |
VisualArtifactSpec 1.0
Evaluating Visual Artifacts under Task Specifications
211 tasks · 4 artifact families · 14 task types · 8 application purposes
VisualArtifactSpec (VAS) evaluates how agents turn source materials and task specifications into visual artifacts that serve practical needs. Tasks span creation and modification, with explicit objectives, materials, constraints, and deliverables. Evaluation separates Task Fulfillment from Artifact Performance, assessing both requirement satisfaction and suitability for the intended audience, purpose, and medium.
Illustrations show task coverage, not model results.
Task coverage
| Artifact family | Tasks | Task types |
|---|---|---|
| Presentations | 47 | Original decks, template adaptation, existing-deck revision |
| Static Graphics | 72 | Posters, infographics and visual summaries, banners, multi-page series |
| Animations & Videos | 32 | Animated graphics, animated videos, edited videos |
| Web Interfaces | 60 | Content browsing, data exploration, tools and simulations, stateful workflows |
Categories follow the primary authoring objective rather than the file extension. Each task includes instructions, input attachments, evaluator guidance, and task-specific rubrics. The dataset contains 11,718 rubric items: 7,920 requirement checks and 3,798 quality items, with 3,125 acceptance groups.
Download
Browse the tasks or download the complete dataset below.
pip install huggingface_hub datasets
from huggingface_hub import snapshot_download
snapshot_download(
repo_id="Accio-Lab/VisualArtifactSpec",
repo_type="dataset",
local_dir="vas-bench",
)
Attachments are included under each task's inputs/ directory. Pass vas-bench/data to the local Docker Judge and scoring tools. Generate submissions using your own model and harness.
Dataset layout and programmatic access
data/
benchmark.json
tasks/TASK_ID/
prompt.md
platform_instructions.txt
rubric.json
evaluation_prompt.txt
task.json
inputs.manifest.json
inputs/
tasks.jsonl
prompt.mdandplatform_instructions.txt: agent-facing task and execution requirements.rubric.jsonandevaluation_prompt.txt: evaluator criteria, answer anchors, weights, and instructions.task.json: task metadata, acceptance groups, and objective definitions.inputs.manifest.json: attachment paths and hashes.
Task IDs use category names and three-digit numbers, such as presentations-001 and web-interfaces-001. Attachment filenames and nested paths follow the task's references.
The dataset viewer and Python API read tasks.jsonl:
from datasets import load_dataset
tasks = load_dataset("Accio-Lab/VisualArtifactSpec", split="test")
Fields ending in _json contain JSON strings. Download the repository snapshot to obtain the actual input files. Pin the dataset commit when reporting or comparing evaluations.
Evaluation
Task Fulfillment establishes the base score (0–50). Compliance checks specified obligations; Functionality checks required behaviors where applicable. Each acceptance group represents one independently acceptable output or behavior. All requirements in a group must pass, and groups contribute equally to the shared 50-point pool.
Artifact Performance provides an adjustment (−50 to +50). Eighteen shared criteria assess Content, Organization, Visuals, and Delivery, using task-specific anchors. Ratings from −3 to +3 are mapped to points before weighted aggregation.
Final score = fulfillment base + performance adjustment
Range: −50 to 100 · Pass: 50 or above · Negative totals are retained.
Judgments must cite inspectable evidence from source materials, rendered views, native structures, temporal observations, or recorded interactions. Unresolved evaluations remain unscored.
Scoring details
| Criterion rating | −3 | −2 | −1 | 0 | +1 | +2 | +3 |
|---|---|---|---|---|---|---|---|
| Mapped points | −50 | −25 | −10 | 0 | +10 | +25 | +50 |
Average mapped points within each dimension using rubric item weights, then combine Content (25%), Organization (15%), Visuals (50%), and Delivery (10%).
B = 50 × passed acceptance groups / all acceptance groups
Δ = 0.25 C + 0.15 O + 0.50 V + 0.10 D
S = B + Δ
C/O/V/D are dimension point scores after mapping, not raw ratings. Zero denotes ordinary quality without substantive defects. Positive levels require the progressively stronger evidence specified by each item's anchors.
Keep source materials and submitted artifacts unchanged. Retain item-level judgments and evidence, report completion coverage and evaluation failures, and compare common tasks under matching rubrics, grouping, and scoring rules. Complete saved judgments can be rescored with the code repository's scoring tools.
For observation requirements and rating boundaries, see the Judge protocol.
Citation
If you use VisualArtifactSpec in your research, please cite our report:
@misc{ouyang2026visualartifactspec10evaluatingvisual,
title={VisualArtifactSpec 1.0: Evaluating Visual Artifacts under Task Specifications},
author={Shuyi Ouyang and Xiaojun Tang and Zijian Zou and Xintong Zhang and Mingli Song and Guannan Zhang and Hongwei Xue},
year={2026},
eprint={2611.21382},
archivePrefix={arXiv},
primaryClass={cs.AI},
url={https://arxiv.org/abs/2611.21382},
}
License
Original task instructions, rubrics, and illustrations use CC BY 4.0; the code and Judge harness use Apache-2.0; third-party attachments and components retain their original licenses and attribution requirements.
- Downloads last month
- 6

