mlsanityData to 7 Oct 2026

Evaluating the Impact of Post-Training Quantization on Reliable VQA with Multimodal LLMs

arXiv 2602.13289 ↗First posted 8 Feb 2026cs.CV cs.AIcited by 0

On the map: Multimodal LLMs / vision-language models → Visual token pruning and efficiency

Abstract (arXiv)

Multimodal Large Language Models (MLLM) are increasingly deployed in domains where both reliability and efficiency are critical. However, current models remain overconfident, producing highly certain but incorrect answers. At the same time, their large size limits deployment on edge devices, necessitating compression. We study the intersection of these two challenges by analyzing how Post-Training Quantization (PTQ) compression affects both accuracy and reliability in Visual Question Answering (VQA). We evaluate two MLLMs, Qwen2-VL-7B and Idefics3-8B, quantized with data-free (HQQ) and data-aware (MBQ) methods across multiple bit widths. To counteract the reduction in reliability caused by quantization, we adapt the Selector confidence estimator for quantized multimodal settings and test its robustness across various quantization levels and out-of-distribution (OOD) scenarios. We find that PTQ degrades both accuracy and reliability. Data-aware methods soften the effect thereof. The Selector substantially mitigates the reliability impact. The combination of int4 MBQ and the Selector achieves the best efficiency-reliability trade-off, closing in on uncompressed performance at approx. 75% less memory demand. Overall, we present the first systematic study linking quantization and reliability in multimodal settings.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Visual token pruning and efficiency, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 21 of its 22 references that are in the map. First posted on arXiv 8 Feb 2026.

Builds on

21 of 22 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024MBQ: Modality-Balanced Quantization for Large Vision-Language Models27 Dec

Other references · 20

  1. 2017Attention Is All You Needcited by 59,932
  2. 2023Visual Instruction Tuningcited by 7,792
  3. 2023BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Modelscited by 6,069
  4. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,151
  5. 2016Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answeringcited by 2,498
  6. 2015VQA: Visual Question Answeringcited by 2,489
  7. 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,362
  8. 2019Towards VQA Models That Can Readcited by 1,696
  9. 2019OK-VQA: A Visual Question Answering Benchmark Requiring External Knowledgecited by 1,114
  10. 2023ShareGPT4V: Improving Large Multi-Modal Models with Better Captionscited by 1,000
  11. 2018VizWiz Grand Challenge: Answering Visual Questions from Blind Peoplecited by 852
  12. 2017Selective Classification for Deep Neural Networkscited by 799
8 more
  1. 2021A Survey of Quantization Methods for Efficient Neural Network Inferencecited by 798
  2. 2020How Can We Know When Language Models Know? On the Calibration of Language Models for Question Answeringcited by 439
  3. 2024Building and better understanding vision-language models: insights and future directionscited by 169
  4. 2022Reliable Visual Question Answering: Abstain Rather Than Answer Incorrectlycited by 86
  5. 2024Evaluating Quantized Large Language Modelscited by 75
  6. 2021Human-Adversarial Visual Question Answeringcited by 57
  7. 2024A Comprehensive Evaluation of Quantization Strategies for Large Language Modelscited by 47
  8. 2023Improving Selective Visual Question Answering by Learning from Your Peerscited by 24

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Towards Understanding Best Practices for Quantization of Vision-Language Models
  2. 2025Benchmarking Post-Training Quantization in LLMs: Comprehensive Taxonomy, Unified Evaluation, and Comparative Analysis
  3. 2025LUQ: Layerwise Ultra-Low Bit Quantization for Multimodal Large Language Models
  4. 2026QAPruner: Quantization-Aware Vision Token Pruning for Multimodal Large Language Models
  5. 2025Evaluating Variance in Visual Question Answering Benchmarks
  6. 2024Learning to Compress Contexts for Efficient Knowledge-based Visual Question Answering
  7. 2024Efficient Large Multi-modal Models via Visual Context Compression
  8. 2026VOILA: Value-of-Information Guided Fidelity Selection for Cost-Aware Multimodal Question Answering