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

cs.CVOct 27, 2026

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

Authors: Lai Wei, Yatai Li, Liangbo He, Jun Lan, Renjie Gu, Linhan Cao, Junkai Lin, Huijia Zhu,

Organizations: School of Computer Science, Shanghai Jiao Tong University · Zhongguancun Academy · Ant Group · Fudan University ·

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.

Figures & tables

Figure 1: Performance comparison of ZwZ2 and other advanced models.

Figure 2: Overview of our multi-region question synthesis pipeline. We select related regions from a single image and construct a privileged crop view to generate compositional questions. The illustrative example connects a flight number on a boarding pass to its departure gate and then to a service icon (T218 $\rightarrow$ C6 $\rightarrow$ plug). Cross-model answer anchoring and drop-one region tests assess answer reliability and evidence dependence, followed by multiple-choice conversion and difficulty filtering. During training, the policy receives the full image with region outlines, while privileged crops are used only for data construction.

Domains	Benchmarks	Base models			ZwZ2			Advanced models
Domains	Benchmarks	4B	9B	27B	4B	9B	27B	GPT	GLM	DeepSeek
	VStar	82.20	83.77	87.96	89.00	90.58	91.62	75.92	83.77	78.53
	HR-4K	82.37	84.88	84.25	88.62	89.75	89.88	83.87	84.25	79.13
	HR-8K	78.37	82.50	80.00	83.63	87.62	88.00	79.13	80.63	75.00
	MME-en	70.60	71.00	72.62	75.48	75.33	77.41	58.50	67.88	50.75
	MME-cn	69.33	69.53	68.68	73.89	73.45	74.97	56.90	68.31	50.04

Table 1: Main results on 25 benchmarks (accuracy, %). Domain averages and Overall are unweighted means within each domain and over all 25 benchmarks, respectively.

Models	Fine-grained perception			Spatial perception			Embodied spatial reasoning			Long-context understanding		Avg.
Models	HR-4K	HR-8K	Chart	O3	CountQA	MMSI	MindCube	ERQA	ViewSpatial	MileBench	MemLens	Avg.
Base	82.37	78.37	5.00	53.91	36.91	30.70	47.62	45.75	37.43	60.08	40.62	47.16
MMHops (31K)	77.75	73.12	3.00	45.80	30.82	30.50	42.10	43.75	40.90	66.29	15.97	42.73
TVG (19.4K)	78.25	73.62	13.00	53.04	33.97	29.50	28.86	46.00	42.16	64.56	37.29	45.48
F 2 RVLM (37K)	80.13	77.62	10.00	54.20	32.72	31.90	34.38	44.75	40.56	64.58	39.80	46.42
CRIT (30K)	82.25	77.88	6.00	45.22	30.56	29.90	43.43	45.00	43.98	68.13	40.40	46.61

Table 2: Training Qwen3.5-4B on different synthetic datasets (accuracy, %). Parentheses give the number of training examples. HR denotes HR-Bench; Chart and O3 denote Chartography and O3-Bench. MMSI, MindCube, and ViewSpatial denote MMSI-Bench, MindCube-Tiny, and ViewSpatial-Bench, respectively.

Settings	Fine-grained perception			Spatial perception			Embodied spatial reasoning			Long-context understanding		Avg.
Settings	HR-4K	HR-8K	Chart	O3	CountQA	MMSI	MindCube	ERQA	ViewSpatial	MileBench	MemLens	Avg.
Base	82.37	78.37	5.00	53.91	36.91	30.70	47.62	45.75	37.43	60.08	40.62	47.16
Without privileged views	85.12	82.50	13.00	53.62	40.51	33.00	50.19	52.25	43.87	70.22	40.87	51.38
Without drop-one	85.75	79.75	8.00	58.26	41.30	34.10	47.14	51.50	43.84	70.16	41.48	51.03
Ours (7K)	86.62	82.87	9.00	62.90	45.09	34.60	52.29	47.50	45.68	70.47	42.14	52.65

Table 3: Data-synthesis ablations on Qwen3.5-4B (accuracy, %). Each trained variant uses 7K examples. The 7K reference model is shared by both comparisons and is reported only once; it uses privileged views and drop-one verification. Avg. is the unweighted mean over 11 benchmarks.

Appendix figures & tables4 assets

Supplementary material from the paper’s appendix.

Appendix

Stage	Model	Visual input	Calls	Temp.
Evidence discovery	Qwen3.5-397B-A17B	source image	1 per image	0.4
Evidence grouping	Qwen3.5-397B-A17B	none (text only)	1 per image	0.3
Question realization	Qwen3.5-397B-A17B	$V_{G}$	1 per group	0.5
Answer anchoring	Qwen3.5-397B-A17B, Kimi-K2.6	$V_{G}$	up to 4 + 4 per candidate	0.4
Equivalence judging	GLM-5.2	none (text only)	when needed	0.0
Drop-one testing	Qwen3.5-397B-A17B	$V_{G}^{-i}$	1–3 per region	0.4

Appendix

Table 4: Model calls in the EvidenceConnect pipeline.

V_{G}

denotes the privileged layout,

V_{G}^{-i}

the layout with region

r_{i}

masked, and

V_{I}

the policy view.

Figure 7: Composition of the 20.4K training examples by the number of selected regions per example and by difficulty level.

	4B			9B
Layer	Qwen3.5	ZwZ2	$\Delta$	Qwen3.5	ZwZ2	$\Delta$
4	11.625	11.877	+0.252	10.622	10.699	+0.077
8	10.355	10.402	+0.047	10.069	10.051	-0.018
12	11.568	11.692	+0.124	11.073	11.050	-0.023
16	10.400	10.650	+0.251	11.080	11.256	+0.175
20	11.132	11.386	+0.255	11.493	11.777	+0.284

Appendix

Table 5: Top-

k

evidence IoU on TreeBench. Mean IoU is reported as a percentage;

\Delta

is ZwZ2 minus Qwen3.5 in percentage points. Each entry averages the same 405 paired examples. Differences are computed before rounding.

Figure 8: Qualitative comparisons between Qwen3.5-4B and ZwZ2-4B. Three selected examples show correct predictions by ZwZ2-4B where the base model fails: (a) identifying a category present in every image; (b) associating recognized text with the queried location; and (c) counting all instances in a structured arrangement. Responses are excerpted from the original outputs, with omissions marked by [...] and answer fields reformatted. The enlarged image region in (b) is provided for visualization only.

Multimodal Understanding

Multimodal understanding studies how AI systems interpret and reason across text, images, video, audio, and other signals.

See all 479 papers

MemEye: A Visual-Centric Evaluation Framework for Multimodal Agent Memory
May 14, 2026Minghao Guo, Qingyue Jiao, Zeru Shi +14Multimodal Agent Evaluation Multimodal Memory
Rutgers · Notre Dame · Princeton +1
Imaginative Perception Tokens Enhance Spatial Reasoning in Multimodal Language Models
Jun 2, 2026Mahtab Bigverdi, Linjie Li, Weikai Huang +8Multimodal Understanding Spatial Reasoning
University of Washington · 2Allen Institute for AI · 3Microsoft +1
Don't Blink: Evidence Collapse during Multimodal Reasoning
Apr 5, 2026Suresh Raghu, Satwik PandeySelective Prediction for VLMs Vision-Language Models
Independent Researcher, New Delhi, India

May 14, 2026J/K move · Enter open · S save

MemEye: A Visual-Centric Evaluation Framework for Multimodal Agent Memory

Minghao Guo, Qingyue Jiao, Zeru Shi, Yihao Quan+13

Rutgers · Notre Dame · Princeton+1

Multimodal Agent Evaluation Multimodal Memory AI Agent Evaluation Long-Horizon Agent Evaluation Visually Grounded Reasoning Multimodal Model Evaluation

Open paper AskPDF arXiv

Long-term agent memory is increasingly multimodal, yet existing evaluations rarely test whether agents preserve the visual evidence needed for later reasoning. In prior work, many visually grounded questions can be answered using only captions or textual traces, allowing answers to be inferred without preserving the fine-grained visual evidence. Meanwhile, harder cases that require reasoning over changing visual states are largely absent. Therefore, we introduce MemEye, a framework that evaluates memory capabilities from two dimensions: one measures the granularity of decisive visual evidence (from scene-level to pixel-level evidence), and the other measures how retrieved evidence must be used (from single evidence to evolutionary synthesis). Under this framework, we construct a new benchmark across 8 life-scenario tasks, with ablation-driven validation gates for assessing answerability, shortcut resistance, visual necessity, and reasoning structure. By evaluating 13 memory methods across 4 VLM backbones, we show that current architectures still struggle to preserve fine-grained visual details and reason about state changes over time. Our findings show that long-term multimodal memory depends on evidence routing, temporal tracking, and detail extraction.

Abstract

Multimodal Understanding

Explore similar work

MemEye: A Visual-Centric Evaluation Framework for Multimodal Agent Memory

Imaginative Perception Tokens Enhance Spatial Reasoning in Multimodal Language Models

Don't Blink: Evidence Collapse during Multimodal Reasoning