Datasetpublic

julian8897/finmix-autoscientist-10k

FinMix AutoScientist 10k A deterministic, upload-ready 10,000-row subset of FinMix v1, created for fast finance adaptation runs in the Adaption AutoScientist challenge. Use with Adaption Adaptive Data Import this Hugging Face dataset and map: Prompt: prompt Context: context Completion: completion Leave task_type, source, and group_key unmapped. They are retained for provenance and auditing. Fields Field Description prompt Financial… See the full description on the dataset page: https://huggingface.co/datasets/julian8897/finmix-autoscientist-10k.

sourceHugging Facecc-by-4.0updated 2mo agoView on Hugging Face

0likes104downloads

Dataset Card

FinMix AutoScientist 10k

A deterministic, upload-ready 10,000-row subset of FinMix v1, created for fast finance adaptation runs in the Adaption AutoScientist challenge.

Use with Adaption Adaptive Data

Import this Hugging Face dataset and map:

—Prompt: prompt
—Context: context
—Completion: completion

Leave task_type, source, and group_key unmapped. They are retained for provenance and auditing.

Fields

Field	Description
`prompt`	Financial question or instruction
`context`	System-level task and response guidance
`completion`	Expected answer
`task_type`	Controlled FinMix task label
`source`	Upstream dataset identifier
`group_key`	Source document or filing group

Composition

Task	Rows
`qa`	2,288
`qa_with_context`	2,288
`sentiment_analysis`	2,188
`numeric_reasoning`	1,289
`topic_classification`	998
`qa_conversation`	749
`ner_sentiment_analysis`	100
`risk_analysis`	100
Total	10,000

The subset contains 8,908 complete groups. Sampling approximately preserves the full training task mixture while guaranteeing at least 100 examples per task. Every selected group_key is retained in full, so no filing or source document is partially sampled. FinMix validation and test rows are excluded.

Reproduction

bash

uv run scripts/subsample_dataset.py \
  --input data/finmix-v1/train.jsonl \
  --output data/finmix-autoscientist-10k/train.jsonl \
  --size 10000 \
  --min-per-task 100 \
  --seed finmix-autoscientist-10k

The generated train.jsonl has SHA-256:

text

70c8c12c3d9e80a1c7d8f54c6b0a808aa8471ce12ff3183ef0dedb31e630e067

See train.stats.json for the full task and source composition.

Licence and attribution

The subset inherits the FinMix v1 CC-BY-4.0 aggregate licence. Individual components remain under their upstream licences. Per-source attribution is provided in NOTICE, and every row retains its source identifier.

Dataset authors: Julian Chan and Sherri Chuah, equal contribution.