UCSC-VLAA/VLM-CapCurriculum-VisualReasoning-Data
VLM-CapCurriculum-VisualReasoning (D_vis) Stage-3 visual-reasoning data for the staged post-training recipe in "From Seeing to Thinking: Decoupling Perception and Reasoning Improves Post-Training of Vision-Language Models" (ICML 2026). A 16,195-sample mix of visual math and figure-grounded reasoning, drawn from four open-source corpora and packed alongside the source images. Every row also ships with a precomputed pass_rate so the same data can be ordered by sample difficulty… See the full description on the dataset page: https://huggingface.co/datasets/UCSC-VLAA/VLM-CapCurriculum-VisualReason
| Author / organization | UCSC-VLAA |
|---|---|
| License (as declared) | apache-2.0 |
| Downloads | 177 (as reported by Hugging Face) |
| Likes | 0 |
| Created | 15 May 2026 |
| Last updated at source | 20 May 2026 |
Popularity, last 3 days
177 now (-15 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Connected in the knowledge graph
Recent changes
- 16 Oct 2026 Description: VLM-CapCurriculum-VisualReasoning (D_vis) Stage-3 visual-reasoning data for the staged post-training recipe in "From Seeing to Thinking: Decoupling Perception and Reasoning Improves Post-Training of Vision-Language Models" (ICML 2026). A 16 → VLM-CapCurriculum-VisualReasoning (D_vis) Stage-3 visual-reasoning data for the staged post-training recipe in "From Seeing to Thinking: Decoupling Perception and Reasoning Improves Post-Training of Vision-Language Models" (ICML 2026). A 16
Changes detected by Silytics between refreshes of the source.
Related papers
More from UCSC-VLAA
- UCSC-VLAA/Recap-DataComp-1BDataset
- UCSC-VLAA/HQ-EditDataset
- UCSC-VLAA/GPT-Image-Edit-1.5MDataset
- UCSC-VLAA/gpt-image-edit-benchmark-resultsDataset
- UCSC-VLAA/MedReasonDataset
- UCSC-VLAA/Complex-EditDataset
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.