Be Consistent! Enhancing Robust Visual Reasoning in LVLMs with Consistency Constraints

• URL: http://arxiv.org/abs/2607.21722v1
• Date: Thu, 23 Jul 2026 18:04:51 GMT
• Title: Be Consistent! Enhancing Robust Visual Reasoning in LVLMs with Consistency Constraints
• Abstract summary: Large Vision-Language Models (LVLMs) exhibit strong perceptual capabilities, but remain vulnerable in visual reasoning tasks.<n>We introduce ConVBench, a vision-centric reasoning benchmark in which each image is paired with two logically equivalent questions.<n>We also present ConVLM, which improves LVLM reasoning through Group Relative Policy Optimization (GRPO)-based reinforcement learning with a novel consistency reward.
• Score: 31.911378374313376
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: While Large Vision-Language Models (LVLMs) exhibit strong perceptual capabilities, they remain vulnerable in visual reasoning tasks. Existing benchmarks largely focus on symbolic mathematical or scientific problems and simple vision-centric tasks, offering limited assessment of complex visual reasoning and logical consistency, a critical requirement for reliable reasoning systems. We introduce ConVBench, a complex vision-centric reasoning benchmark in which each image is paired with two logically equivalent questions across six categories: action and state, complex counting, spatial reasoning, causal and intent understanding, commonsense reasoning, and temporal perception. To complement this benchmark, we define two evaluation metrics, logical consistency and robust accuracy, that jointly assess both the correctness and consistency of model responses. We further present ConVLM, which improves LVLM reasoning through Group Relative Policy Optimization (GRPO)-based reinforcement learning with a novel consistency reward. This method leverages automatically generated logically equivalent question-answer pairs and a dual-reward design combining accuracy- and consistency-based signals, encouraging agreement between paired responses. The framework functions effectively with or without strict answer supervision.

Related papers

• Weakly supervised concept Bottleneck Learning for Robust Two stage Object centric visual reasoning [3.6959117433332724]
  Two-stage neuro-symbolic architectures provide an elegant paradigm for visual problem solving.<n>We introduce the Dynamic Orthogonal Concept Bottleneck (D-OCB), an object-centric slot- VAE framework.
  arXiv  Detail & Related papers  (2026-08-23T20:39:49Z)
• Bridging Perception and Reasoning: Token Reweighting for RLVR in Multimodal LLMs [51.60575965819268]
  We propose a plug-and-play Token-Reweighting (ToR) strategy that explicitly models this interdependence.<n>ToR delivers consistent performance gains across multiple multi-modal reasoning benchmarks.
  arXiv  Detail & Related papers  (2026-03-26T06:25:27Z)
• A Survey of Reasoning in Autonomous Driving Systems: Open Challenges and Emerging Paradigms [49.66022971508878]
  We argue that reasoning should be elevated from a modular component to the system's cognitive core.<n>We derive and systematize seven core reasoning challenges, such as the responsiveness-reasoning trade-off and social-game reasoning.<n>We identify a fundamental and unresolved tension between the high-latency, deliberative nature of LLM-based reasoning and the millisecond-scale, safety-critical demands of vehicle control.
  arXiv  Detail & Related papers  (2026-03-11T07:40:53Z)
• GeoReason: Aligning Thinking And Answering In Remote Sensing Vision-Language Models Via Logical Consistency Reinforcement Learning [12.987952829880363]
  GeoReason is a framework designed to synchronize internal thinking with final decisions.<n>We first construct GeoReason-Bench, a logic-driven dataset containing 4,000 reasoning trajectories.<n>We then formulate a two-stage training strategy: (1) Supervised Knowledge Initialization to equip the model with reasoning syntax and domain expertise, and (2) Consistency-Aware Reinforcement Learning to refine deductive reliability.
  arXiv  Detail & Related papers  (2026-01-07T17:26:41Z)
• Stepwise Think-Critique: A Unified Framework for Robust and Interpretable LLM Reasoning [47.867294403474176]
  We propose Stepwise Think-Critique, a unified framework that interleaves reasoning and self-critique at each step within a single model.<n> STC is trained with a hybrid reinforcement learning objective combining reasoning rewards and critique-consistency rewards to jointly optimize reasoning quality and self-evaluation.
  arXiv  Detail & Related papers  (2025-12-17T18:15:17Z)
• STaR: Towards Cognitive Table Reasoning via Slow-Thinking Large Language Models [12.745473719032026]
  We present STaR (slow-thinking for table reasoning), a new framework achieving cognitive table reasoning.<n> STaR explicitly modeling step-by-step thinking and uncertainty-aware inference.<n>Experiments on benchmarks demonstrate that STaR achieves superior performance and enhanced reasoning stability.
  arXiv  Detail & Related papers  (2025-11-14T12:34:17Z)
• Uni-MMMU: A Massive Multi-discipline Multimodal Unified Benchmark [69.8473923357969]
  Unified multimodal models aim to jointly enable visual understanding and generation, yet current benchmarks rarely examine their true integration.<n>We present Uni-MMMU, a comprehensive benchmark that unfolds the bidirectional synergy between generation and understanding across eight reasoning-centric domains.
  arXiv  Detail & Related papers  (2025-10-15T17:10:35Z)
• Selection, Reflection and Self-Refinement: Revisit Reasoning Tasks via a Causal Lens [19.316594303998667]
  Reasoning tasks have long been regarded as rigorous benchmarks for assessing the capabilities of machine learning models.<n>We revisit reasoning tasks from a causal perspective, seeking to understand their behavior in latent space.<n>We introduce a framework, called SR$2$, that incorporates the estimated latent variables as feedback into the selection mechanism.
  arXiv  Detail & Related papers  (2025-10-09T13:45:31Z)
• Perception-Consistency Multimodal Large Language Models Reasoning via Caption-Regularized Policy Optimization [72.30168853571216]
  multimodal large language models excel at tasks that integrate visual perception with symbolic reasoning.<n>CapPO integrates two key mechanisms: (1) a caption-based consistency regularization, which minimizes the divergence between responses conditioned on raw images and those conditioned on captions, and (2) a KL-weighted advantage estimation scheme, which adaptively scales reinforcement signals to strengthen perceptually consistent trajectories.
  arXiv  Detail & Related papers  (2025-09-26T04:32:26Z)
• A Cognitive Paradigm Approach to Probe the Perception-Reasoning Interface in VLMs [3.2228025627337864]
  This paper introduces a structured evaluation framework to dissect the perception-reasoning interface in Vision-Language Models (VLMs)<n>We propose three distinct evaluation paradigms, mirroring human problem-solving strategies.<n>Applying this framework, we demonstrate that CA, leveraging powerful language models for reasoning over rich, independently generated descriptions, achieves new state-of-the-art (SOTA) performance.
  arXiv  Detail & Related papers  (2025-01-23T12:42:42Z)
• Measuring and Improving Chain-of-Thought Reasoning in Vision-Language Models [61.28463542324576]
  Vision-language models (VLMs) have recently demonstrated strong efficacy as visual assistants that can generate human-like outputs. We evaluate existing state-of-the-art VLMs and find that even the best-performing model is unable to demonstrate strong visual reasoning capabilities and consistency. We propose a two-stage training framework aimed at improving both the reasoning performance and consistency of VLMs.
  arXiv  Detail & Related papers  (2023-09-08T17:49:44Z)

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.