| --- |
| license: other |
| task_categories: |
| - visual-question-answering |
| language: |
| - en |
| --- |
| |
| # SynthFormulaNet_cleaned |
| |
| The **SynthFormulaNet__x** family of the ElliotVL supervised-fine-tuning pool, **after VLM cleaning**. |
| |
| | | | |
| |---|---| |
| | images | 477,803 | |
| | QA turns | 856,390 | |
| | answers rewritten by the cleaning pass | 103,773 | |
| | QA created by the cleaning pass (`new_qa`) | 405,724 (47.4%) | |
| | shards | 6 | |
|
|
| ## How this was cleaned |
|
|
| A vision-language model read each image together with its QA and judged the item. The pass is |
| **not a filter that only removes rows** — it rewrites answers it finds wrong but salvageable, |
| drops what it cannot salvage, and adds QA where the image supports more than the source provided. |
| Each row carries the judge's own record in `clean_meta`, including the cleaner identity, the |
| policy it applied, and its per-item scores for legibility, richness and coverage. |
|
|
| **A large share of the QA here was written by the cleaning pass, not by the original dataset.** |
| Across the pool that share runs from roughly half to over 80% of a family's turns, and it is |
| reported in the table above. Those rows have an empty `org_answer` because no pre-clean original |
| exists. Treat them as model-generated supervision: they were judged against the image, but they |
| are not human annotation, and model-written QA is where formatting defects are most likely. |
|
|
| The effect on the answers that *were* carried over is substantive rather than cosmetic. In this pool the pass corrects values, not just |
| wording — a curve's critical point restated from `4.00` to `2.00`, a computed ratio from `1` to |
| `Approximately 1.33` — and for some families it removed the content entirely, which is why those |
| families are absent here rather than published empty. |
|
|
| ## `answer` vs `org_answer` |
| |
| - `answer` — the cleaned answer. **Train on this.** |
| - `org_answer` — the pre-cleaning answer from the same `(image, question)` in the uncleaned pool. |
| It is empty for QA the cleaning pass added, which have no original. |
|
|
| Where the two differ, the difference is the correction. Keeping both makes every edit auditable |
| instead of asking you to take the cleaning on trust. |
|
|
|
|
|
|
| ## Schema |
|
|
| - `image` — HF `Image()`, renders directly in the dataset viewer |
| - `image_sha256` — content hash of the image |
| - `n_qa` — number of turns attached to this image |
| - `qa` — list of {`question`, `answer`, `org_answer`}, **in source order** |
| - `family` / `clean_meta` — provenance and the judge's per-row record |
|
|
| A row with more than one entry in `qa` is a **conversation over a single image**, not a set of |
| independent questions — the turns depend on each other and the image belongs to all of them. |
| Keep them together and in order when training. |
|
|
| Parquet row groups are written at ~32 MB so the dataset viewer pages without stalling. |
|
|
| ## What is NOT claimed |
|
|
| No rejection sampling was run against this family: there is no accept/reject verdict per row, and |
| `answer` is the cleaned reference rather than a model generation that a string verifier confirmed. |
| For that, use the `*_RS_think` / `*_rejected` datasets in [elliot-mllm](https://huggingface.co/elliot-mllm). |
|
|
|
|
| ## Examples |
| 10 rows taken straight from this dataset. Images are the **original files** as stored here - no downscaling, no recompression - and every turn is shown in full. |
|
|
| ### Example 1 |
|
|
|  |
|
|
| | # | question | answer | |
| |---|---|---| |
| | 1 | Convert formula to latex. | <formula><loc_0><loc_0><loc_500><loc_500>\bigwedge _ { i \in I } \overline { x } ^ { i } \leq \overline { x } ^ { j } \leq \overline { a } ^ { j } = a .</formula> | |
|
|
|
|
| ### Example 2 — 2 turns on one image |
|
|
|  |
|
|
| | # | question | answer | |
| |---|---|---| |
| | 1 | What is the full mathematical inequality written in the image? | g(a_1) < g(a_2) | |
| | 2 | Which function is being compared in the expression? | g | |
|
|
|
|
| ### Example 3 — 4 turns on one image |
|
|
|  |
|
|
| | # | question | answer | |
| |---|---|---| |
| | 1 | Convert formula to latex. | <formula><loc_0><loc_0><loc_500><loc_500>4</formula> | |
| | 2 | What color is the digit? | Black | |
| | 3 | What is the background color? | White | |
| | 4 | What font style is the digit? | Serif | |
|
|
|
|
| ### Example 4 |
|
|
|  |
|
|
| | # | question | answer | |
| |---|---|---| |
| | 1 | Convert formula to latex. | <formula><loc_0><loc_0><loc_500><loc_500>F ( x ) = { \frac { 1 } { q + 1 } } x ^ { q + 1 }</formula> | |
|
|
|
|
| ### Example 5 — 2 turns on one image |
|
|
|  |
|
|
| | # | question | answer (cleaned) | org_answer (pre-judge) | |
| |---|---|---|---| |
| | 1 | <formula> | v(x,t) | <loc_0><loc_0><loc_500><loc_500>v ( x , t )</formula> | |
| | 2 | What is the color of the mathematical expression? | Black | _(unchanged)_ | |
|
|
|
|
| ### Example 6 |
|
|
|  |
|
|
| | # | question | answer | |
| |---|---|---| |
| | 1 | Describe the visual appearance of the character in the image. | It is a black, lowercase, italic letter 'd' on a white background. | |
|
|
|
|
| ### Example 7 |
|
|
|  |
|
|
| | # | question | answer | |
| |---|---|---| |
| | 1 | Convert formula to latex. | <formula><loc_0><loc_0><loc_500><loc_500>v ^ { ( 2 ) } ( z ) = \begin{cases} \begin{pmatrix} 1 & - s e ^ { - 2 t f ( z ) } \delta ^ { 2 } ( z ) \\ 0 & 1 \end{pmatrix} \begin{pmatrix} 1 & 0 \\ s e ^ { 2 t f ( z ) } \delta ^ { - 2 } ( z ) & 1 \end{pmatrix} , \quad & z \in \Sigma _ { 1 } , \\ \begin{pmatrix} 1 & 0 \\ \frac { s } { 1 - s ^ { 2 } } e ^ { 2 t f ( z ) } \delta _ { - } ^ { - 2 } & 1 - s ^ { 2 } \end{pmatrix} \begin{pmatrix} 1 & - \frac { s } { 1 - s ^ { 2 } } e ^ { - 2 t f ( z ) } \delta _ { + } ^ { 2 } ( z ) \\ 0 & 1 \end{pmatrix} , \quad & z \in \Sigma _ { 2 } . \end{cases}</formula> | |
|
|
|
|
| ### Example 8 — 3 turns on one image |
|
|
|  |
|
|
| | # | question | answer | |
| |---|---|---| |
| | 1 | Convert formula to latex. | <formula><loc_0><loc_0><loc_500><loc_500>\begin{array} { r l } { [ c ] \phi _ { 0 } } & = 1 } \\ { \phi _ { 1 } } & = 2 \sin ( n x / 2 ) \sin ( y ) } \\ { \phi _ { 2 } } & = 2 \sin ( n x / 2 ) \sin ( 2 y ) } \end{array}</formula> | |
| | 2 | What are the colors of the text and the background? | Black text on a white background. | |
| | 3 | How many equations are listed in the image? | Three. | |
|
|
|
|
| ### Example 9 |
|
|
|  |
|
|
| | # | question | answer (cleaned) | org_answer (pre-judge) | |
| |---|---|---|---| |
| | 1 | <formula> | P ^ { - 1 } & = \left ( R + \frac { 1 } { N - 1 } H A \left ( H A \right ) ^ { T } \right ) ^ { - 1 } \\ & = R ^ { - 1 } \left [ I - \frac { 1 } { N - 1 } \left ( H A \right ) \left ( I + \left ( H A \right ) ^ { T } R ^ { - 1 } \frac { 1 } { N - 1 } \left ( H A \right ) \right ) ^ { - 1 } \left ( H A \right ) ^ { T } R ^ { - 1 } \right ] | <loc_0><loc_0><loc_500><loc_500>P ^ { - 1 } & = \left ( R + \frac { 1 } { N - 1 } H A \left ( H A \right ) ^ { T } \right ) ^ { - 1 } \\ & = R ^ { - 1 } \left [ I - \frac { 1 } { N - 1 } \left ( H A \right ) \left ( I + \left ( H A \right ) ^ { T } R ^ { - 1 } \frac { 1 } { N - 1 } \left ( H A \right ) \right ) ^ { - 1 } \left ( H A \right ) ^ { T } R ^ { - 1 } \right ] ,</formula> | |
| |
| |
| ### Example 10 — 2 turns on one image |
| |
|  |
| |
| | # | question | answer | |
| |---|---|---| |
| | 1 | Convert formula to latex. | <formula><loc_0><loc_0><loc_500><loc_500>\mu \frac { d m ^ { 2 } ( \mu ) } { d \mu } = \gamma _ { m } ( \lambda ) m ^ { 2 }</formula> | |
| | 2 | What are the colors of the text and the background in the image? | The text is black and the background is white. | |
|
|
|
|
