VLN-MME: Diagnosing MLLMs as Language-guided Visual Navigation agents

• URL: http://arxiv.org/abs/2512.24851v2
• Date: Tue, 06 Jan 2026 11:00:10 GMT
• Title: VLN-MME: Diagnosing MLLMs as Language-guided Visual Navigation agents
• Authors: Xunyi Zhao, Gengze Zhou, Qi Wu,
• Abstract summary: We introduce a unified and evaluation framework to probe MLLMs as zero-shot agents.<n>We simplify the evaluation with a highly modular and accessible design.<n>We observe that enhancing our baseline agent with Chain-of-Thought (CoT) reasoning and self-language leads to an unexpected performance decrease.
• Score: 12.383467162169703
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multimodal Large Language Models (MLLMs) have demonstrated remarkable capabilities across a wide range of vision-language tasks. However, their performance as embodied agents, which requires multi-round dialogue spatial reasoning and sequential action prediction, needs further exploration. Our work investigates this potential in the context of Vision-and-Language Navigation (VLN) by introducing a unified and extensible evaluation framework to probe MLLMs as zero-shot agents by bridging traditional navigation datasets into a standardized benchmark, named VLN-MME. We simplify the evaluation with a highly modular and accessible design. This flexibility streamlines experiments, enabling structured comparisons and component-level ablations across diverse MLLM architectures, agent designs, and navigation tasks. Crucially, enabled by our framework, we observe that enhancing our baseline agent with Chain-of-Thought (CoT) reasoning and self-reflection leads to an unexpected performance decrease. This suggests MLLMs exhibit poor context awareness in embodied navigation tasks; although they can follow instructions and structure their output, their 3D spatial reasoning fidelity is low. VLN-MME lays the groundwork for systematic evaluation of general-purpose MLLMs in embodied navigation settings and reveals limitations in their sequential decision-making capabilities. We believe these findings offer crucial guidance for MLLM post-training as embodied agents.

Related papers

• Can LLMs See Without Pixels? Benchmarking Spatial Intelligence from Textual Descriptions [18.455501447828343]
  Spatial Intelligence (SI) has predominantly relied on Vision-Language Models (VLMs)<n>We introduce SiT-Bench, a novel benchmark designed to evaluate the SI performance of Large Language Models (LLMs) without pixel-level input.<n>We find that explicit spatial reasoning significantly boosts performance, suggesting that LLMs possess latent world-modeling potential.
  arXiv  Detail & Related papers  (2026-01-07T05:13:52Z)
• Empowering Reliable Visual-Centric Instruction Following in MLLMs [30.078761482873745]
  Existing benchmarks for evaluating MLLMs' instruction-following capability primarily focus on verbal instructions.<n>We introduce VC-IFEval, a new benchmark accompanied by a systematically constructed dataset.<n>Our benchmark systematically incorporates vision-dependent constraints into instruction design, enabling a more rigorous and fine-grained assessment.
  arXiv  Detail & Related papers  (2026-01-06T17:23:33Z)
• Seeing Beyond Words: Self-Supervised Visual Learning for Multimodal Large Language Models [53.06230963851451]
  JARVIS is a JEPA-inspired framework for self-supervised visual enhancement in MLLMs.<n>We introduce JARVIS, a JEPA-inspired framework for self-supervised visual enhancement in MLLMs.
  arXiv  Detail & Related papers  (2025-12-17T19:01:34Z)
• NaViL: Rethinking Scaling Properties of Native Multimodal Large Language Models under Data Constraints [100.02131897927484]
  This paper focuses on the native training of Multimodal Large Language Models (MLLMs) in an end-to-end manner.<n>We propose a native MLLM called NaViL, combined with a simple and cost-effective recipe.<n> Experimental results on 14 multimodal benchmarks confirm the competitive performance of NaViL against existing MLLMs.
  arXiv  Detail & Related papers  (2025-10-09T17:59:37Z)
• Abstractive Visual Understanding of Multi-modal Structured Knowledge: A New Perspective for MLLM Evaluation [48.462734327375536]
  Multi-modal large language models (MLLMs) incorporate heterogeneous modalities into LLMs, enabling a comprehensive understanding of diverse scenarios and objects.<n>Despite the proliferation of evaluation benchmarks and leaderboards for MLLMs, they predominantly overlook the critical capacity of MLLMs to comprehend world knowledge with structured abstractions that appear in visual form.<n>We propose M3STR, an innovative benchmark grounded in the Multi-Modal Map for STRuctured understanding.<n>Our findings reveal persistent deficiencies in processing abstractive visual information with structured knowledge, thereby charting a pivotal trajectory for advancing MLLMs' holistic reasoning capacities.
  arXiv  Detail & Related papers  (2025-06-02T04:00:35Z)
• Keeping Yourself is Important in Downstream Tuning Multimodal Large Language Model [63.14883657299359]
  Multi-modal Large Language Models (MLLMs) integrate visual and linguistic reasoning to address complex tasks such as image captioning and visual question answering.<n> tuning MLLMs for downstream tasks encounters two key challenges: Task-Expert, where distribution shifts between pre-training and target datasets constrain target performance, and OpenWorld Stabilization, where catastrophic forgetting erases the model general knowledge.
  arXiv  Detail & Related papers  (2025-03-06T15:29:13Z)
• Analyzing Finetuning Representation Shift for Multimodal LLMs Steering [56.710375516257876]
  We propose to map hidden states to interpretable visual and textual concepts.<n>This enables us to more efficiently compare certain semantic dynamics, such as the shift from an original and fine-tuned model.<n>We also demonstrate the use of shift vectors to capture these concepts changes.
  arXiv  Detail & Related papers  (2025-01-06T13:37:13Z)
• AgentPS: Agentic Process Supervision for Content Moderation with Multimodal LLMs [9.35901507816989]
  We introduce AgentPS, a framework that integrates Agentic Process Supervision into large language models.<n>We show that AgentPS achieves substantial improvements over baseline MLLMs on public benchmarks and proprietary datasets.<n>These results establish AgentPS as a scalable and effective solution for complex multimodal classification in large-scale industrial applications.
  arXiv  Detail & Related papers  (2024-12-15T04:58:00Z)
• Elevating Visual Perception in Multimodal LLMs with Visual Embedding Distillation [109.5893580175657]
  In recent times, the standard practice for developing MLLMs is to feed features from vision encoder(s) into the LLM and train with natural language supervision.<n>This approach often causes models to lean towards language comprehension and undermine the rich visual perception signals present in the data.<n>We propose VisPer-LM, the first approach that infuses visual perception knowledge from expert vision encoders into the LLM's hidden representations.
  arXiv  Detail & Related papers  (2024-12-12T18:55:18Z)
• NavGPT-2: Unleashing Navigational Reasoning Capability for Large Vision-Language Models [30.685419129265252]
  We bridge the divide between VLN-specialized models and LLM-based navigation paradigms. We exploit a way to incorporate LLMs and navigation policy networks for effective action predictions and navigational reasoning.
  arXiv  Detail & Related papers  (2024-07-17T07:44:26Z)
• Dense Connector for MLLMs [89.50595155217108]
  We introduce the Dense Connector - a plug-and-play vision-language connector that significantly enhances existing MLLMs. Building on this, we also propose the Efficient Dense Connector, which achieves performance comparable to LLaVA-v1.5 with only 25% of the visual tokens. Our model, trained solely on images, showcases remarkable zero-shot capabilities in video understanding as well.
  arXiv  Detail & Related papers  (2024-05-22T16:25:03Z)
• TINA: Think, Interaction, and Action Framework for Zero-Shot Vision Language Navigation [11.591176410027224]
  This paper presents a Vision-Language Navigation (VLN) agent based on Large Language Models (LLMs) We propose the Thinking, Interacting, and Action framework to compensate for the shortcomings of LLMs in environmental perception. Our approach also outperformed some supervised learning-based methods, highlighting its efficacy in zero-shot navigation.
  arXiv  Detail & Related papers  (2024-03-13T05:22:39Z)

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.