VOILA: Value-of-Information Guided Fidelity Selection for Cost-Aware Multimodal Question Answering
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910009916391424 |
|---|---|
| author | Bhope, Rahul Atul Jayaram, K. R. Muthusamy, Vinod Kumar, Ritesh Isahagian, Vatche Venkatasubramanian, Nalini |
| author_facet | Bhope, Rahul Atul Jayaram, K. R. Muthusamy, Vinod Kumar, Ritesh Isahagian, Vatche Venkatasubramanian, Nalini |
| contents | Despite significant costs from retrieving and processing high-fidelity visual inputs, most multimodal vision-language systems operate at fixed fidelity levels. We introduce VOILA, a framework for Value-Of-Information-driven adaptive fidelity selection in Visual Question Answering (VQA) that optimizes what information to retrieve before model execution. Given a query, VOILA uses a two-stage pipeline: a gradient-boosted regressor estimates correctness likelihood at each fidelity from question features alone, then an isotonic calibrator refines these probabilities for reliable decision-making. The system selects the minimum-cost fidelity maximizing expected utility given predicted accuracy and retrieval costs. We evaluate VOILA across three deployment scenarios using five datasets (VQA-v2, GQA, TextVQA, LoCoMo, FloodNet) and six Vision-Language Models (VLMs) with 7B-235B parameters. VOILA consistently achieves 50-60% cost reductions while retaining 90-95% of full-resolution accuracy across diverse query types and model architectures, demonstrating that pre-retrieval fidelity selection is vital to optimize multimodal inference under resource constraints. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_03007 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | VOILA: Value-of-Information Guided Fidelity Selection for Cost-Aware Multimodal Question Answering Bhope, Rahul Atul Jayaram, K. R. Muthusamy, Vinod Kumar, Ritesh Isahagian, Vatche Venkatasubramanian, Nalini Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Despite significant costs from retrieving and processing high-fidelity visual inputs, most multimodal vision-language systems operate at fixed fidelity levels. We introduce VOILA, a framework for Value-Of-Information-driven adaptive fidelity selection in Visual Question Answering (VQA) that optimizes what information to retrieve before model execution. Given a query, VOILA uses a two-stage pipeline: a gradient-boosted regressor estimates correctness likelihood at each fidelity from question features alone, then an isotonic calibrator refines these probabilities for reliable decision-making. The system selects the minimum-cost fidelity maximizing expected utility given predicted accuracy and retrieval costs. We evaluate VOILA across three deployment scenarios using five datasets (VQA-v2, GQA, TextVQA, LoCoMo, FloodNet) and six Vision-Language Models (VLMs) with 7B-235B parameters. VOILA consistently achieves 50-60% cost reductions while retaining 90-95% of full-resolution accuracy across diverse query types and model architectures, demonstrating that pre-retrieval fidelity selection is vital to optimize multimodal inference under resource constraints. |
| title | VOILA: Value-of-Information Guided Fidelity Selection for Cost-Aware Multimodal Question Answering |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2602.03007 |