LENS: Multi-level Evaluation of Multimodal Reasoning with Large Language Models

• URL: http://arxiv.org/abs/2505.15616v1
• Date: Wed, 21 May 2025 15:06:59 GMT
• Title: LENS: Multi-level Evaluation of Multimodal Reasoning with Large Language Models
• Authors: Ruilin Yao, Bo Zhang, Jirui Huang, Xinwei Long, Yifang Zhang, Tianyu Zou, Yufei Wu, Shichao Su, Yifan Xu, Wenxi Zeng, Zhaoyu Yang, Guoyou Li, Shilan Zhang, Zichan Li, Yaxiong Chen, Shengwu Xiong, Peng Xu, Jiajun Zhang, Bowen Zhou, David Clifton, Luc Van Gool,
• Abstract summary: Lens is a benchmark with 3.4K contemporary images and 60K+ human-authored questions covering eight tasks and 12 daily scenarios.<n>This dataset intrinsically supports to evaluate MLLMs to handle image-invariable prompts, from basic perception to compositional reasoning.<n>We evaluate 15+ frontier MLLMs such as Qwen2.5-VL-72B, InternVL3-78B, GPT-4o and two reasoning models QVQ-72B-preview and Kimi-VL.
• Score: 59.0256377330646
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Multimodal Large Language Models (MLLMs) have achieved significant advances in integrating visual and linguistic information, yet their ability to reason about complex and real-world scenarios remains limited. The existing benchmarks are usually constructed in the task-oriented manner without guarantee that different task samples come from the same data distribution, thus they often fall short in evaluating the synergistic effects of lower-level perceptual capabilities on higher-order reasoning. To lift this limitation, we contribute Lens, a multi-level benchmark with 3.4K contemporary images and 60K+ human-authored questions covering eight tasks and 12 daily scenarios, forming three progressive task tiers, i.e., perception, understanding, and reasoning. One feature is that each image is equipped with rich annotations for all tasks. Thus, this dataset intrinsically supports to evaluate MLLMs to handle image-invariable prompts, from basic perception to compositional reasoning. In addition, our images are manully collected from the social media, in which 53% were published later than Jan. 2025. We evaluate 15+ frontier MLLMs such as Qwen2.5-VL-72B, InternVL3-78B, GPT-4o and two reasoning models QVQ-72B-preview and Kimi-VL. These models are released later than Dec. 2024, and none of them achieve an accuracy greater than 60% in the reasoning tasks. Project page: https://github.com/Lens4MLLMs/lens. ICCV 2025 workshop page: https://lens4mllms.github.io/mars2-workshop-iccv2025/

Related papers

• Analyze-Prompt-Reason: A Collaborative Agent-Based Framework for Multi-Image Vision-Language Reasoning [3.588567067449924]
  We present a Collaborative Agent-Based Framework for Multi-Image Reasoning.<n>Our approach tackles the challenge of interleaved multimodal reasoning across diverse datasets and task formats.<n>We evaluate our method on 18 diverse datasets from the 2025 MIRAGE Challenge.
  arXiv  Detail & Related papers  (2025-08-01T06:39:15Z)
• Do You See Me : A Multidimensional Benchmark for Evaluating Visual Perception in Multimodal LLMs [9.951669153984708]
  "Do You See Me" is a scalable benchmark with 1,758 images and 2,612 questions.<n>Humans achieve 96.49% accuracy, while top MLLMs average below 50%.<n>This underscores an urgent need for MLLMs with truly robust visual perception.
  arXiv  Detail & Related papers  (2025-05-28T13:31:32Z)
• Can Multimodal Large Language Models Understand Spatial Relations? [16.76001474065412]
  We introduce SpatialMQA, a human-annotated spatial relation reasoning benchmark based on COCO 2017.<n>Results indicate that the current state-of-the-art MLLM achieves only 48.14% accuracy, far below the human-level accuracy of 98.40%.
  arXiv  Detail & Related papers  (2025-05-25T07:37:34Z)
• VOILA: Evaluation of MLLMs For Perceptual Understanding and Analogical Reasoning [63.0285363282581]
  Multimodal Large Language Models (MLLMs) have become a powerful tool for integrating visual and textual information.<n>We introduce VOILA, a benchmark designed to evaluate MLLMs' perceptual understanding and abstract relational reasoning.<n>We reveal that current MLLMs struggle to comprehend inter-image relationships and exhibit limited capabilities in high-level relational reasoning.
  arXiv  Detail & Related papers  (2025-02-25T23:36:19Z)
• MM-Vet v2: A Challenging Benchmark to Evaluate Large Multimodal Models for Integrated Capabilities [146.4724093405187]
  We introduce MM-Vet v2, which includes a new "image-text sequence understanding" capability called "image-text sequence understanding"<n>Using MM-Vet v2 to benchmark large multimodal models, we found that Claude 3.5 Sonnet is the best model with a score of 71.8, slightly outperforming GPT-4o which scored 71.0.
  arXiv  Detail & Related papers  (2024-08-01T17:59:54Z)
• Q-Bench+: A Benchmark for Multi-modal Foundation Models on Low-level Vision from Single Images to Pairs [71.07108539262721]
  We design benchmark settings to emulate human language responses related to low-level vision. We extend the low-level perception-related question-answering and description evaluations of MLLMs from single images to image pairs. We demonstrate that several MLLMs have decent low-level visual competencies on single images, but only GPT-4V exhibits higher accuracy on pairwise comparisons than humans.
  arXiv  Detail & Related papers  (2024-02-11T06:44:11Z)
• SEED-Bench-2: Benchmarking Multimodal Large Language Models [67.28089415198338]
  Multimodal large language models (MLLMs) have recently demonstrated exceptional capabilities in generating not only texts but also images given interleaved multimodal inputs. SEED-Bench-2 comprises 24K multiple-choice questions with accurate human annotations, which spans 27 dimensions. We evaluate the performance of 23 prominent open-source MLLMs and summarize valuable observations.
  arXiv  Detail & Related papers  (2023-11-28T05:53:55Z)
• Pink: Unveiling the Power of Referential Comprehension for Multi-modal LLMs [49.88461345825586]
  This paper proposes a new framework to enhance the fine-grained image understanding abilities of MLLMs. We present a new method for constructing the instruction tuning dataset at a low cost by leveraging annotations in existing datasets. We show that our model exhibits a 5.2% accuracy improvement over Qwen-VL and surpasses the accuracy of Kosmos-2 by 24.7%.
  arXiv  Detail & Related papers  (2023-10-01T05:53:15Z)

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.