mlsanityData to 14 Oct 2026

MirrorBench: Evaluating Self-centric Intelligence in MLLMs by Introducing a Mirror

arXiv 2604.14785 ↗First posted 16 Apr 2026cs.AIcited by 0

On the map: Multimodal LLMs / vision-language models → Embodied and 3D vision-language models

Abstract (arXiv)

Recent progress in Multimodal Large Language Models (MLLMs) has demonstrated remarkable advances in perception and reasoning, suggesting their potential for embodied intelligence. While recent studies have evaluated embodied MLLMs in interactive settings, current benchmarks mainly target capabilities to perceive, understand, and interact with external objects, lacking a systematic evaluation of self-centric intelligence. To address this, we introduce MirrorBench, a simulation-based benchmark inspired by the classical Mirror Self-Recognition (MSR) test in psychology. MirrorBench extends this paradigm to embodied MLLMs through a tiered framework of progressively challenging tasks, assessing agents from basic visual perception to high-level self-representation. Experiments on leading MLLMs show that even at the lowest level, their performance remains substantially inferior to human performance, revealing fundamental limitations in self-referential understanding. Our study bridges psychological paradigms and embodied intelligence, offering a principled framework for evaluating the emergence of general intelligence in large models. Project page: https://fflahm.github.io/mirror-bench-page/.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Embodied and 3D vision-language models, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 22 of its 34 references that are in the map. First posted on arXiv 16 Apr 2026.

Builds on

22 of 34 references.

Closest ancestors · 4

Same direction, newest first

  1. 2025GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning1 Jul
  2. 2025NavBench: Probing Multimodal Large Language Models for Embodied Navigation1 Jun
  3. 2025EmbodiedBench: Comprehensive Benchmarking Multi-modal Large Language Models for Vision-Driven Embodied Agents13 Feb
  4. 2025EmbodiedEval: Evaluate Multimodal LLMs as Embodied Agents21 Jan

Other references · 18

  1. 2023GPT-4 Technical Reportcited by 18,396
  2. 2023Visual Instruction Tuningcited by 7,869
  3. 2025Qwen2.5-VL Technical Reportcited by 5,587
  4. 2024GPT-4o System Cardcited by 4,588
  5. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,796
  6. 2023Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyondcited by 2,224
  7. 2023MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGIcited by 2,080
  8. 2023MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Modelscited by 1,652
  9. 2025InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiencycited by 1,510
  10. 2023MM-Vet: Evaluating Large Multimodal Models for Integrated Capabilitiescited by 1,239
  11. 2019OK-VQA: A Visual Question Answering Benchmark Requiring External Knowledgecited by 1,121
  12. 2019ALFRED: A Benchmark for Interpreting Grounded Instructions for Everyday Taskscited by 783
6 more
  1. 2018nocaps: novel object captioning at scalecited by 444
  2. 2017Embodied Question Answeringcited by 336
  3. 2024VLABench: A Large-Scale Benchmark for Language-Conditioned Robotics Manipulation with Long-Horizon Reasoning Taskscited by 175
  4. 2024BEHAVIOR-1K: A Human-Centered, Embodied AI Benchmark with 1,000 Everyday Activities and Realistic Simulationcited by 148
  5. 2025RBench-V: A Primary Assessment for Visual Reasoning Models with Multi-modal Outputscited by 17
  6. 2020Robot in the mirror: toward an embodied computational model of mirror self-recognitioncited by 5

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Mirror, Mirror on the Wall: Can VLM Agents Tell Who They Are at All?
  2. 2025ECBench: Can Multi-modal Foundation Models Understand the Egocentric World? A Holistic Embodied Cognition Benchmark
  3. 2025Static and Plugged: Make Embodied Evaluation Simple
  4. 2024MM-SAP: A Comprehensive Benchmark for Assessing Self-Awareness of Multimodal Large Language Models in Perception
  5. 2025Robobench: A Comprehensive Evaluation Benchmark for Multimodal Large Language Models as Embodied Brain
  6. 2025MM-IQ: Benchmarking Human-Like Abstraction and Reasoning in Multimodal Models
  7. 2025ENACT: Evaluating Embodied Cognition with World Modeling of Egocentric Interaction
  8. 2025IQBench: How "Smart'' Are Vision-Language Models? A Study with Human IQ Tests