🚨CVPR 2026 Accepted
ChartNet: A Million-Scale, High-Quality Multimodal Dataset for Robust Chart Understanding
in collaboration with @MIT & @IBM
1.5M-sample open-source dataset for robust chart understanding
Each sample aligns a chart image, plotting code, CSV/table data, natural-language summary, and QA with reasoning
Full breakdown👇:
#CVPR26 #CVPR2026
✈️We're heading to #CVPR2026
🏄Booth #701 with free swags
📡DataMFM workshop with insights and prizes
🪩A Happy Hour with food, drinks, and relaxed chats
DataMFM workshop
June 3rd, Room 111, 1:00–6:00 PM
Happy Hour (spots are limited)
June 6th
Don’t forget to pass by and bring your perspective
DataMFM: https://t.co/2oC6hkyNVN
Happy hour, register now: https://t.co/OXv5gmlvcg
@_catwu Huge!
Quick question: how does verification state survive the fan-out? When a branch returns a partial result, does Claude treat that as a local retry, a dependency update for sibling agents, or a replan of the whole workflow?
Great work! Thanks for sharing!
@astrogu_@lateinteraction@MIT_CSAIL@nlp_mit@samrmadden@qizhengz_alex@Stanford Amazing work! Love the cache policy angle
Chart workflows have this annoying split-brain thing where the plot, CSV, and code all know the truth, but different parts
Curious how PEEK handles contexts that are partly visual and partly executable
@vikataravi bleed suppression loss is very smart! Thanks for sharing!
we’ve hit the chart version of this with legends, axes, and tiny series overlaps. pixel grounding helps, but tying the visual back to CSV/code...
curious how M3Grounder behaves on chart-heavy docs!
@_cagarwal@CVPR So we've been paying O(n) for visual depth when O(1) would often do. That's a big efficiency win, huge implications for real-time VLMs!
ChartNet: a 1.5M-sample open-source dataset for chart understanding
With an aligned chart image, plotting code, CSV/table data, summary, and QA with reasoning,
Improves chart reconstruction, data extraction, summarization, and chart QA across model sizes.
Huge thanks to respected collaborators @kondic_jovana, @RogerioFeris, @AudeOliva, @ZihanWang123, and everyone involved!
The dataset is out on Hugging Face.
Test it, break it, and we'll see you at #CVPR2026 to discuss!
🚨CVPR 2026 Accepted
ChartNet: A Million-Scale, High-Quality Multimodal Dataset for Robust Chart Understanding
in collaboration with @MIT & @IBM
1.5M-sample open-source dataset for robust chart understanding
Each sample aligns a chart image, plotting code, CSV/table data, natural-language summary, and QA with reasoning
Full breakdown👇:
#CVPR26 #CVPR2026