SleepWalk: A Three-Tier Benchmark for Stress-Testing Instruction-Guided Vision-Language Navigation

• URL: http://arxiv.org/abs/2605.10376v1
• Date: Mon, 11 May 2026 11:20:14 GMT
• Title: SleepWalk: A Three-Tier Benchmark for Stress-Testing Instruction-Guided Vision-Language Navigation
• Abstract summary: SleepWalk is a benchmark for evaluating instruction-grounded trajectory prediction in single-scene 3D worlds.<n>We evaluate three frontier Vision-Language Models on 2,472 curated 3D environments with nine instructions per scene.
• Score: 18.453985392979785
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Vision-Language Models (VLMs) have advanced rapidly in multimodal perception and language understanding, yet it remains unclear whether they can reliably ground language into spatially coherent, plausibly executable actions in 3D digital environments. We introduce SleepWalk, a benchmark for evaluating instruction-grounded trajectory prediction in single-scene 3D worlds generated from textual scene descriptions and filtered for navigability. Unlike prior navigation benchmarks centered on long-range exploration across rooms, SleepWalk targets localized, interaction-centric embodied reasoning: given rendered visual observations and a natural-language instruction, a model must predict a trajectory that respects scene geometry, avoids collisions, and terminates at an action-compatible location. The benchmark covers diverse indoor and outdoor environments and organizes tasks into three tiers of spatial and temporal difficulty, enabling fine-grained analysis of grounding under increasing compositional complexity. Using a standardized pointwise judge-based evaluation protocol, we evaluate three frontier VLMs on 2,472 curated 3D environments with nine instructions per scene. Results reveal systematic failures in grounded spatial reasoning, especially under occlusion, interaction constraints, and multi-step instructions: performance drops as the difficulty level of the tasks increase. In general, current VLMs can somewhat produce trajectories that are simultaneously spatially coherent, plausibly executable, and aligned with intended actions. By exposing failures in a controlled yet scalable setting, SleepWalk provides a critical benchmark for advancing grounded multimodal reasoning, embodied planning, vision-language navigation, and action-capable agents in 3D environments.

Related papers

• SpatialAct: Probing Spatial Reasoning-to-Action Capabilities of VLM Agents in 3D Scenes [15.076506371158876]
  Humans can effortlessly perceive spatial layouts, form cognitive representations, and translate such reasoning into actions in everyday 3D environments.<n>Recent vision-language models (VLMs) have shown promising performance on observation-conditioned spatial perception and reasoning tasks.<n>We introduce textbfSpatialAct, a simulator-grounded benchmark for probing textitaction-conditioned spatial reasoning in 3D scenes.
  arXiv  Detail & Related papers  (2026-05-29T10:59:26Z)
• Bridging the 2D-3D Gap: A Hierarchical Semantic-Geometric Map for Vision Language Navigation [32.02017382315305]
  Vision-Language Navigation (VLN) enables embodied agents to reach target locations in unseen environments by following language instructions.<n>Despite recent progress with vision-language models (VLMs), a critical semantic-geometric gap remains.<n>We propose a Hierarchical Semantic-Geometric Map (HSGM) that transforms 3D geometric information into a structured representation compatible withVLMs.
  arXiv  Detail & Related papers  (2026-05-25T08:53:21Z)
• PokeGym: A Visually-Driven Long-Horizon Benchmark for Vision-Language Models [33.89873575295867]
  We introduce PokeGym, a visually-driven benchmark instantiated within Pokemon Legends: Z-A, a visually complex 3D open-world Role-Playing Game.<n> PokeGym enforces strict code-level isolation: agents operate solely on raw RGB observations while an independent evaluator verifies success via memory scanning.<n>The benchmark comprises 30 tasks (30-220 steps) spanning navigation, interaction, and mixed scenarios, with three instruction granularities (Visual-Guided, Step-Guided, Goal-Only) to systematically deconstruct visual grounding, semantic reasoning, and autonomous exploration capabilities.
  arXiv  Detail & Related papers  (2026-04-09T15:12:36Z)
• HUGE-Bench: A Benchmark for High-Level UAV Vision-Language-Action Tasks [73.62053624331227]
  Huge-Bench is a benchmark for High-Level UAV Vision-Language-Action tasks.<n>It comprises 4 real-world digital twin scenes, 8 high-level tasks, and 2.56M meters of trajectories.<n>We introduce process-oriented and collision-aware metrics to assess process fidelity, terminal accuracy, and safety.
  arXiv  Detail & Related papers  (2026-03-20T10:08:42Z)
• 3DGSNav: Enhancing Vision-Language Model Reasoning for Object Navigation via Active 3D Gaussian Splatting [12.057873540714098]
  3DGSNav is a novel framework that embeds 3D Gaussian Splatting (3DGS) as persistent memory for vision-language models (VLMs) to enhance spatial reasoning.<n>3DGSNav incrementally constructs a 3DGS representation of the environment, enabling trajectory-guided free-viewpoint rendering of frontier-aware first-person views.<n>During navigation, a real-time object detector filters potential targets, while VLM-driven active viewpoint switching performs target re-verification.
  arXiv  Detail & Related papers  (2026-02-12T16:41:26Z)
• TrackTeller: Temporal Multimodal 3D Grounding for Behavior-Dependent Object References [33.32268865872059]
  We study temporal language-based 3D grounding, where the objective is to identify the referred object in the current frame by leveraging multi-frame observations.<n>We propose TrackTeller, a temporal multimodal grounding framework that integrates LiDAR-image fusion, language-conditioned decoding, and temporal reasoning in a unified architecture.
  arXiv  Detail & Related papers  (2025-12-25T12:02:56Z)
• OccVLA: Vision-Language-Action Model with Implicit 3D Occupancy Supervision [31.929268076595122]
  OccVLA is a novel framework that integrates 3D occupancy representations into a unified multimodal reasoning process.<n>OccVLA achieves state-of-the-art results on the nuScenes benchmark for trajectory planning and demonstrates superior performance on 3D visual question-answering tasks.
  arXiv  Detail & Related papers  (2025-09-06T03:47:21Z)
• ReSem3D: Refinable 3D Spatial Constraints via Fine-Grained Semantic Grounding for Generalizable Robotic Manipulation [12.059517583878756]
  We propose ReSem3D, a unified manipulation framework for semantically diverse environments.<n>We show that ReSem3D performs diverse manipulation tasks under zero-shot conditions, exhibiting strong adaptability and generalization.
  arXiv  Detail & Related papers  (2025-07-24T10:07:31Z)
• Move to Understand a 3D Scene: Bridging Visual Grounding and Exploration for Efficient and Versatile Embodied Navigation [54.04601077224252]
  Embodied scene understanding requires not only comprehending visual-spatial information but also determining where to explore next in the 3D physical world.<n>underlinetextbf3D vision-language learning enables embodied agents to effectively explore and understand their environment.<n>model's versatility enables navigation using diverse input modalities, including categories, language descriptions, and reference images.
  arXiv  Detail & Related papers  (2025-07-05T14:15:52Z)
• Agentic 3D Scene Generation with Spatially Contextualized VLMs [67.31920821192323]
  We introduce a new paradigm that enables vision-language models to generate, understand, and edit complex 3D environments.<n>We develop an agentic 3D scene generation pipeline in which the VLM iteratively reads from and updates the spatial context.<n>Results show that our framework can handle diverse and challenging inputs, achieving a level of generalization not observed in prior work.
  arXiv  Detail & Related papers  (2025-05-26T15:28:17Z)
• Dynam3D: Dynamic Layered 3D Tokens Empower VLM for Vision-and-Language Navigation [61.21302433849139]
  Vision-and-Language Navigation (VLN) is a core task where embodied agents leverage their spatial mobility to navigate in 3D environments.<n>We propose Dynam3D, a dynamic layered 3D representation model that leverages language-aligned, generalizable, and hierarchical 3D representations as visual input to train 3D-VLM in navigation action prediction.<n>Our Dynam3D is capable of online encoding and localization of 3D instances, and dynamically updates them in changing environments to provide large-scale exploration and long-term memory capabilities for navigation.
  arXiv  Detail & Related papers  (2025-05-16T15:46:27Z)
• MMScan: A Multi-Modal 3D Scene Dataset with Hierarchical Grounded Language Annotations [55.022519020409405]
  This paper builds the first largest ever multi-modal 3D scene dataset and benchmark with hierarchical grounded language annotations, MMScan.<n>The resulting multi-modal 3D dataset encompasses 1.4M meta-annotated captions on 109k objects and 7.7k regions as well as over 3.04M diverse samples for 3D visual grounding and question-answering benchmarks.
  arXiv  Detail & Related papers  (2024-06-13T17:59:30Z)
• Volumetric Environment Representation for Vision-Language Navigation [66.04379819772764]
  Vision-language navigation (VLN) requires an agent to navigate through a 3D environment based on visual observations and natural language instructions. We introduce a Volumetric Environment Representation (VER), which voxelizes the physical world into structured 3D cells. VER predicts 3D occupancy, 3D room layout, and 3D bounding boxes jointly.
  arXiv  Detail & Related papers  (2024-03-21T06:14:46Z)

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.