AVA, a training-free framework that wraps a frozen RS MLLM with three modules that improves average accuracy by up to 2.40 points over the strongest same-setting baselines, and raises the visual-dependency score from 0.28 to 0.49, is presented.
Abstract
Remote sensing visual question answering (RS-VQA) requires answers that truly depend on image evidence, yet current remote sensing multimodal large language models (RS MLLMs) are still evaluated mainly by accuracy, which cannot reveal whether a correct answer was produced without using the image. This setting introduces three challenges: demonstrations retrieved across different ground sample distances (GSD), sensors, and city textures may be poorly matched; real queries often involve long visual contexts from multi-scale tiles, multi-temporal frames, and retrieved knowledge, where evidence can be diluted and errors can propagate through multi-step reasoning; and prior-only answering remains invisible to standard metrics. We present AVA, a training-free framework that wraps a frozen RS MLLM with three modules. GDR performs GSD-bucketed, scene-graph-based demonstration retrieval and filters demonstrations whose answers are unsupported by their images. MACE decomposes each query into an evidence-typed sub-question DAG, routes sub-questions to suitable perception tools, and records verified provenance in a shared evidence ledger with bounded re-perception. CVA audits the final answer through counterfactual perturbations of cited evidence, including ablation and swap, and computes a visual-dependency score for confidence calibration and evidence-forcing re-query. On RSVQA-LR/HR, VRSBench, and LHRS-Bench, AVA improves average accuracy by up to 2.40 points over the strongest same-setting baselines, raises the visual-dependency score from 0.28 to 0.49, reduces the prior-only answer rate from 46.2 percent to 21.7 percent, and lowers the hallucination rate from 17.1 percent to 9.8 percent, with 2.6x the cost of a single backbone forward pass and no fine-tuning.
Reliable language-guided perception is essential for embodied agents, where questions about objects, events, and their locations must be answered from long observation histories. Unlike conventional visual question answering, realistic robotic settings require reasoning over extended temporal horizons and persistent sc...
Victoria Khoruzhaya, М. В. Калмыков, Tuong Do Minh et al.· 0 citations
Advancing spatial intelligence in Multimodal Large Language Models (MLLMs) is bottlenecked by the scarcity of complex, scalable 3D question-answer (QA) data. While manual annotation is labor-intensive, directly utilizing LLMs to synthesize these QA pairs often fails due to their inherent deficiencies in spatial and geo...
The HBB inference profiles of AgentRSVG, whose controller maintains cross-scale hypotheses, directs retrieval and zoom toward unresolved query constraints, and uses a grounding verifier to stop or continue are studied to identify a strong operating point for budget-constrained agentic grounding.
Chong-Yang Li· Proceedings of the 1st Inter...· 0 citations
This work introduces ReVA, a new dataset for remote sensing video question answering, designed to assess spatiotemporal, scene-centric, and reasoning-oriented capabilities of MLLMs, and develops a semi-automatic annotation pipeline that leverages Text LLMs and MLLMs for question-answer generation with human verificatio...
Zhen Yao, Li-Kai Wang, Yu-Ming Yang et al.· arXiv.org· 3 citations
Real-world reasoning rarely reduces to static question answering: agents must actively gather information from tools and sensors that are often noisy and unreliable. Yet most existing active reasoning benchmarks assume that environmental feedback is trustworthy, or introduce noise without exposing an explicit, calibrat...
B. Nguyễn, Zhao-Nan Li, M. Nguyen et al.· 0 citations
QVR3D is introduced, a query-guided visual reasoning framework that operates on RGB-D observations using off-the-shelf foundation models that integrates instance confidence, attribute consistency, and relationship consistency and supports both single- and multi-object grounding without task-specific adaptation.
Jung-Hao Kao, Jing Huang, Jia-Wei Liao et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.