SPARC-Qwen3.5-0.8B-VTFT

Qwen3.5-0.8B fully fine-tuned for embodied spatial reasoning using VQA data generated from SPARC annotations. VTFT denotes vision-tower fine-tuning.

Training data

The training mixture contains SPARC-generated VQA data from ours_adaptive_det_soft_snr_sp8, FSD, RoboPoint, and LLaVA-OneVision2. SPARC samples use an annotation-quality threshold of 0.97, are sorted by score, and are capped at 700 samples per object. This is the same mixture as the 4B release.

Release	SPARC VQA (filtered)	FSD	RoboPoint	LLaVA-OneVision2	EO-1.5M
Qwen3.5-4B	Yes	Yes	Yes	Yes	No
Qwen3.5-0.8B-VTFT	Yes	Yes	Yes	Yes	No
Qwen3.5-9B-EO	Yes	Yes	Yes	Yes	Yes

The SPARC training data is the filtered release subset, which needs no further SPARC filtering. Its filtering script and release mixture manifest are in irl-kit/SPARC-VQA-Raw. FSD, RoboPoint, LLaVA-OneVision2, and EO-1.5M remain their respective upstream datasets.

Prompting

Prompt formatting is important for these models. Use the bundled chat_template.jinja through processor.apply_chat_template(..., add_generation_prompt=True), with one user turn containing the image(s) followed by the text question. Disable thinking/reasoning mode to match evaluation.

For a single point, append exactly:

Output the point coordinates in JSON format like [{"point_2d": [x, y], "label": "target"}]. Use integer coordinates between 0 and 1000.

For a trajectory or multiple points, append exactly:

Return only a JSON list like [{"point_2d": [x1, y1], "label": "point_1"}, {"point_2d": [x2, y2], "label": "point_2"}, ...]. Use integer coordinates between 0 and 1000.

Training

Both the vision encoder and vision projector are trainable. The model was fully fine-tuned for one epoch with a learning rate of 2e-5 and a maximum sequence length of 5600.

Evaluation

This is the strongest evaluated 0.8B variant: local full-benchmark aggregate 0.605, compared with 0.596 for the frozen-vision 0.8B run. The paper appendix reports a 53.2 pointing/VQA average for this variant.

Model	Aggregate	Where2Place	RefSpatial location	IA-Bench	RoboRefIt testA	VA Bench-P
Qwen3.5-4B	0.698	72.0	59.0	79.0	85.7	65.7
Qwen3.5-0.8B-VTFT	0.605	58.0	47.0	76.7	80.9	48.3
Qwen3.5-9B-EO	0.719	76.0	68.0	78.5	85.2	68.7

Citation

@article{blank2026sparc,
  title={SPARC: Reliable Spatial Annotations from Robot Demonstrations at Scale},
  author={Blank, Nils and others},
  journal={arXiv preprint arXiv:2606.13497},
  year={2026}
}