VBVR-Pro: A Scalable and Verifiable Suite for Native Visual Reasoning

• URL: http://arxiv.org/abs/2608.26105v1
• Date: Wed, 26 Aug 2026 17:59:51 GMT
• Title: VBVR-Pro: A Scalable and Verifiable Suite for Native Visual Reasoning
• Abstract summary: We introduce VBVR-Pro, a testbed that makes native visual reasoning generation controllable.<n> VBVR-Pro turns visual reasoning into a controlled task space of 300 procedurally generated tasks.<n>We identify recurring failure modes of the prevalent VLM-as-a-judge paradigm.
• Score: 133.66021945481876
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Native visual reasoning treats visual generation as the medium of reasoning itself: visual states (i.e. images and videos) are not merely inputs to be understood or outputs to be rendered, but first-class substrates for problem solving beyond language. Yet progress remains bottlenecked by the lack of scalable training tasks, reliable feedback, and controlled comparisons across generative substrates. In this work, we introduce VBVR-Pro, a closed-loop testbed that makes native visual reasoning through generation trainable, verifiable, optimizable, and experimentally controllable. 1) Task scaling. VBVR-Pro turns visual reasoning into a controlled task space of 300 procedurally generated tasks. Models trained on VBVR-Pro show strong transfer beyond the proposed suite across seven external visual reasoning benchmarks such as RISE-Video, MME-CoF-Pro, and BabyVision. 2) Verifiable rewards. VBVR-Pro provides verifiable reward scorers for task-grounded evaluation. Through a systematic study of leading MLLMs as judges, we identify recurring failure modes of the prevalent VLM-as-a-judge paradigm. In contrast, the proposed scorers are grounded in deterministic, task-specific rules, achieve fine-grained alignment with human judgments. Importantly, they serve as reliable reward signals for large-scale multi-task reinforcement learning and demonstrate stronger post-RL performance across visual reasoning tasks. 3) Mechanism study. VBVR-Pro enables controlled modality studies across more than 30 image, video, and interleaved generators. Our analysis shows that video generation remains strongest for tasks requiring persistent spatiotemporal state tracking, while interleaved generation provides a compute-efficient alternative. Critically, ablations and probing suggest the presence of vision-native trajectories that are crucial to visual reasoning. We release all data, models, scorers, and code.

Related papers

• VisualThink-VLA: Visual Intermediate Reasoning for Effective and Low-Latency Vision-Language-Action Policies [58.65913948991329]
  We present VISUALTHINK-VLA, a visual intermediate-reasoning framework for accurate, low-latency VLA policies.<n>Our bootstrapping philosophy is to guide action with effective visual thinking.<n>It bootstraps action prediction through a compact visual-evidence interface that preserves spatial precision while avoiding decoding overhead.
  arXiv  Detail & Related papers  (2026-05-28T14:36:53Z)
• CollabVR: Collaborative Video Reasoning with Vision-Language and Video Generation Models [41.60751475372144]
  Video Generation Models (VGMs) exhibit recurring failure modes on goal-directed tasks.<n>We propose VLM-VGM Collaborative Video Reasoning (CollabVR), a closed-loop framework that couples the VLM with the VGM at step-level critiques.<n>On Gen-ViRe and VBVR-Bench, CollabVR improves both open-source and closed-source VGMs over single-inference, Pass@$k$, and prior test-time scaling baselines at matched compute.
  arXiv  Detail & Related papers  (2026-05-09T06:39:17Z)
• PokeGym: A Visually-Driven Long-Horizon Benchmark for Vision-Language Models [33.89873575295867]
  We introduce PokeGym, a visually-driven benchmark instantiated within Pokemon Legends: Z-A, a visually complex 3D open-world Role-Playing Game.<n> PokeGym enforces strict code-level isolation: agents operate solely on raw RGB observations while an independent evaluator verifies success via memory scanning.<n>The benchmark comprises 30 tasks (30-220 steps) spanning navigation, interaction, and mixed scenarios, with three instruction granularities (Visual-Guided, Step-Guided, Goal-Only) to systematically deconstruct visual grounding, semantic reasoning, and autonomous exploration capabilities.
  arXiv  Detail & Related papers  (2026-04-09T15:12:36Z)
• A Very Big Video Reasoning Suite [155.70016888896927]
  Rapid in video models has largely captured visual quality, leaving their reasoning capabilities underexplored.<n>The Very Big Video Reasoning (VBVR) dataset is an unprecedentedly large-scale resource spanning 200 curated reasoning tasks.<n>VBVR-Bench is a verifiable evaluation framework that moves beyond model-based judging by rule-based, human-aligned scorers.
  arXiv  Detail & Related papers  (2026-02-23T18:59:41Z)
• ReViP: Reducing False Completion in Vision-Language-Action Models with Vision-Proprioception Rebalance [50.05984919728878]
  We present ReViP, a novel VLA framework with Vision-Proprioception Rebalance to enhance visual grounding and robustness under perturbations.<n>Specifically, we use an external VLM as a task-stage observer to extract real-time task-centric visual cues from visual observations.<n>To evaluate false completion, we propose the first False-Completion Benchmark Suite built on LIBERO with controlled settings such as Object-Drop.
  arXiv  Detail & Related papers  (2026-01-23T11:31:07Z)
• Beyond Task-Specific Reasoning: A Unified Conditional Generative Framework for Abstract Visual Reasoning [52.107043437362556]
  Abstract visual reasoning (AVR) enables humans to quickly discover and generalize abstract rules to new scenarios.<n>This paper proposes a novel Unified Conditional Generative Solver (UCGS) to address multiple tasks in a unified framework.<n>UCGS exhibits the ability of zero-shot reasoning, enabling it to perform abstract reasoning on problems from unseen tasks in the testing phase.
  arXiv  Detail & Related papers  (2025-07-15T21:54:51Z)
• See, Think, Confirm: Interactive Prompting Between Vision and Language Models for Knowledge-based Visual Reasoning [60.43585179885355]
  We propose a novel framework named Interactive Prompting Visual Reasoner (IPVR) for few-shot knowledge-based visual reasoning. IPVR contains three stages, see, think and confirm. We conduct experiments on a range of knowledge-based visual reasoning datasets.
  arXiv  Detail & Related papers  (2023-01-12T18:59:50Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.