Paper Detail

Zooming without Zooming v2: From Seeing Details to Connecting Evidence for Generalizable Multimodal Perception

Lai Wei, Yatai Li, Liangbo He, Jun Lan, Renjie Gu, Linhan Cao, Junkai Lin, Huijia Zhu, Weiqiang Wang, Linghe Kong, Yue Wang, Zhuosheng Zhang, Weiran Huang

huggingface Score 16.5

Published 2026-10-27 · First seen 2026-10-28

General AI

Abstract

Multimodal perception forms the foundation of multimodal intelligence. In particular, spatial perception and multimodal long-context understanding are two pivotal tasks that underpin embodied AI and multimodal agents. However, synthesizing targeted, high-quality training data for these tasks remains difficult and costly. Our key insight is that these tasks share an underlying capability, which we call connecting evidence: identifying relevant visual facts, binding them to the correct entities, recovering their relations, and composing them into an answer. Crucially, targeted training data are far easier to synthesize for this capability than for these tasks themselves. To strengthen this capability, we propose EvidenceConnect, an easy-to-implement data synthesis framework that uses zoomed views of related regions within a single image to generate questions requiring comparison, aggregation, or chaining of visual evidence. To ensure quality, a reference answer is retained only when independent VLMs agree on it, and region-ablation tests discard questions whose answer survives the removal of selected regions. We then train base VLMs on the resulting data with RLVR, obtaining Zooming without Zooming v2 (ZwZ2). Although the synthesized data are not tailored to any downstream benchmark, ZwZ2 improves fine-grained and spatial perception, with further gains on embodied spatial reasoning, multimodal long-context understanding, multimodal agent tasks, and general visual reasoning. These results establish EvidenceConnect as an effective framework for supervising evidence connection and suggest that strengthening this shared capability leads to more generalizable multimodal perception.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
now
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@misc{wei2026zooming,
  title = {Zooming without Zooming v2: From Seeing Details to Connecting Evidence for Generalizable Multimodal Perception},
  author = {Lai Wei and Yatai Li and Liangbo He and Jun Lan and Renjie Gu and Linhan Cao and Junkai Lin and Huijia Zhu and Weiqiang Wang and Linghe Kong and Yue Wang and Zhuosheng Zhang and Weiran Huang},
  year = {2026},
  abstract = {Multimodal perception forms the foundation of multimodal intelligence. In particular, spatial perception and multimodal long-context understanding are two pivotal tasks that underpin embodied AI and multimodal agents. However, synthesizing targeted, high-quality training data for these tasks remains difficult and costly. Our key insight is that these tasks share an underlying capability, which we call connecting evidence: identifying relevant visual facts, binding them to the correct entities, r},
  url = {https://huggingface.co/papers/2610.37037},
  keywords = {multimodal perception, spatial perception, multimodal long-context understanding, embodied AI, connecting evidence, EvidenceConnect, VLMs, RLVR, Zooming without Zooming v2, visual reasoning, code available, huggingface daily},
  eprint = {2610.37037},
  archiveprefix = {arXiv},
}

Metadata

{}