Image de couverture de livre

Seeing Isn't Knowing: Do VLMs Know When Not to Answer Spatial Questions (and Why)?

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Zhang, Yue, Wang, Zun, Lin, Han, Bitton, Yonatan, Szpektor, Idan, Bansal, Mohit
Format:	Preprint
Publié:	2026
Sujets:	Computer Vision and Pattern Recognition Artificial Intelligence Computation and Language
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Error-Driven Scene Editing for 3D Grounding in Large Language Models
par: Zhang, Yue, et autres
Publié: (2025)

Knowing the Answer Isn't Enough: Fixing Reasoning Path Failures in LVLMs
par: Wang, Chaoyang, et autres
Publié: (2025)

Bridging the Visual Gap: Fine-Tuning Multimodal Models with Knowledge-Adapted Captions
par: Yanuka, Moran, et autres
Publié: (2024)

Mismatch Quest: Visual and Textual Feedback for Image-Text Misalignment
par: Gordon, Brian, et autres
Publié: (2023)

Unblocking Fine-Grained Evaluation of Detailed Captions: An Explaining AutoRater and Critic-and-Revise Pipeline
par: Gordon, Brian, et autres
Publié: (2025)

Visual Riddles: a Commonsense and World Knowledge Challenge for Large Vision and Language Models
par: Bitton-Guetta, Nitzan, et autres
Publié: (2024)

Seeing Isn't Believing: Uncovering Blind Spots in Evaluator Vision-Language Models
par: Khan, Mohammed Safi Ur Rahman, et autres
Publié: (2026)

Knowing When Not to Answer: Evaluating Abstention in Multimodal Reasoning Systems
par: Madhusudhan, Nishanth, et autres
Publié: (2026)

Contrastive Sequential-Diffusion Learning: Non-linear and Multi-Scene Instructional Video Synthesis
par: Ramos, Vasco, et autres
Publié: (2024)

When and How Much to Imagine: Adaptive Test-Time Scaling with World Models for Visual Spatial Reasoning
par: Yu, Shoubin, et autres
Publié: (2026)

TALC: Time-Aligned Captions for Multi-Scene Text-to-Video Generation
par: Bansal, Hritik, et autres
Publié: (2024)

3DLLM-Mem: Long-Term Spatial-Temporal Memory for Embodied 3D Large Language Model
par: Hu, Wenbo, et autres
Publié: (2025)

DreamRunner: Fine-Grained Compositional Story-to-Video Generation with Retrieval-Augmented Motion Adaptation
par: Wang, Zun, et autres
Publié: (2024)

Video-STaR: Self-Training Enables Video Instruction Tuning with Any Supervision
par: Zohar, Orr, et autres
Publié: (2024)

Beyond Words: Multimodal LLM Knows When to Speak
par: Liao, Zikai, et autres
Publié: (2025)

GrAInS: Gradient-based Attribution for Inference-Time Steering of LLMs and VLMs
par: Nguyen, Duy, et autres
Publié: (2025)

Planning with Sketch-Guided Verification for Physics-Aware Video Generation
par: Huang, Yidong, et autres
Publié: (2025)

AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
par: Wang, Zun, et autres
Publié: (2026)

RefVNLI: Towards Scalable Evaluation of Subject-driven Text-to-image Generation
par: Slobodkin, Aviv, et autres
Publié: (2025)

Seeing Isn't Orienting: A Cognitively Grounded Benchmark Reveals Systematic Orientation Failures in MLLMs
par: Tasnim, Nazia, et autres
Publié: (2025)

Generating Coherent Sequences of Visual Illustrations for Real-World Manual Tasks
par: Bordalo, João, et autres
Publié: (2024)

Vision-and-Language Navigation Today and Tomorrow: A Survey in the Era of Foundation Models
par: Zhang, Yue, et autres
Publié: (2024)

Seeing Isn't Orienting: A Cognitively Grounded Benchmark Reveals Systematic Orientation Failures in MLLMs Supplementary
par: Tasnim, Nazia, et autres
Publié: (2026)

Seeing Isn't Believing: Context-Aware Adversarial Patch Synthesis via Conditional GAN
par: Kazoom, Roie, et autres
Publié: (2025)

See It from My Perspective: How Language Affects Cultural Bias in Image Understanding
par: Ananthram, Amith, et autres
Publié: (2024)

VideoLLM Knows When to Speak: Enhancing Time-Sensitive Video Comprehension with Video-Text Duet Interaction Format
par: Wang, Yueqian, et autres
Publié: (2024)

When Pretty Isn't Useful: Investigating Why Modern Text-to-Image Models Fail as Reliable Training Data Generators
par: Adamkiewicz, Krzysztof, et autres
Publié: (2026)

Are VLMs Seeing or Just Saying? Uncovering the Illusion of Visual Re-examination
par: Shi, Chufan, et autres
Publié: (2026)

LLMs Know More Than They Show: On the Intrinsic Representation of LLM Hallucinations
par: Orgad, Hadas, et autres
Publié: (2024)

Seeing Isn't Always Believing: Analysis of Grad-CAM Faithfulness and Localization Reliability in Lung Cancer CT Classification
par: Panboonyuen, Teerapong
Publié: (2026)

Inverse Scaling: When Bigger Isn't Better
par: McKenzie, Ian R., et autres
Publié: (2023)

Rephrase, Augment, Reason: Visual Grounding of Questions for Vision-Language Models
par: Prasad, Archiki, et autres
Publié: (2023)

"I Know It When I See It": Mood Spaces for Connecting and Expressing Visual Concepts
par: Yang, Huzheng, et autres
Publié: (2025)

Reasmory: 3D Reconstruction as Explicit Memory for VLMs Spatial Reasoning
par: He, Jixuan, et autres
Publié: (2026)

Decompose and Compare Consistency: Measuring VLMs' Answer Reliability via Task-Decomposition Consistency Comparison
par: Yang, Qian, et autres
Publié: (2024)

When One Moment Isn't Enough: Multi-Moment Retrieval with Cross-Moment Interactions
par: Cao, Zhuo, et autres
Publié: (2025)

Right this way: Can VLMs Guide Us to See More to Answer Questions?
par: Liu, Li, et autres
Publié: (2024)

CAPTURe: Evaluating Spatial Reasoning in Vision Language Models via Occluded Object Counting
par: Pothiraj, Atin, et autres
Publié: (2025)

When Seeing Overrides Knowing: Disentangling Knowledge Conflicts in Vision-Language Models
par: Ortu, Francesco, et autres
Publié: (2025)

SAME: Learning Generic Language-Guided Visual Navigation with State-Adaptive Mixture of Experts
par: Zhou, Gengze, et autres
Publié: (2024)