Can LLMs See Without Pixels? Benchmarking Spatial Intelligence from Textual Descriptions

• URL: http://arxiv.org/abs/2601.03590v1
• Date: Wed, 07 Jan 2026 05:13:52 GMT
• Title: Can LLMs See Without Pixels? Benchmarking Spatial Intelligence from Textual Descriptions
• Authors: Zhongbin Guo, Zhen Yang, Yushan Li, Xinyue Zhang, Wenyu Gao, Jiacheng Wang, Chengzhi Li, Xiangrui Liu, Ping Jian,
• Abstract summary: 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.
• Score: 18.455501447828343
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent advancements in Spatial Intelligence (SI) have predominantly relied on Vision-Language Models (VLMs), yet a critical question remains: does spatial understanding originate from visual encoders or the fundamental reasoning backbone? Inspired by this question, we introduce SiT-Bench, a novel benchmark designed to evaluate the SI performance of Large Language Models (LLMs) without pixel-level input, comprises over 3,800 expert-annotated items across five primary categories and 17 subtasks, ranging from egocentric navigation and perspective transformation to fine-grained robotic manipulation. By converting single/multi-view scenes into high-fidelity, coordinate-aware textual descriptions, we challenge LLMs to perform symbolic textual reasoning rather than visual pattern matching. Evaluation results of state-of-the-art (SOTA) LLMs reveals that while models achieve proficiency in localized semantic tasks, a significant "spatial gap" remains in global consistency. Notably, we find that explicit spatial reasoning significantly boosts performance, suggesting that LLMs possess latent world-modeling potential. Our proposed dataset SiT-Bench serves as a foundational resource to foster the development of spatially-grounded LLM backbones for future VLMs and embodied agents. Our code and benchmark will be released at https://github.com/binisalegend/SiT-Bench .

Related papers

• TagaVLM: Topology-Aware Global Action Reasoning for Vision-Language Navigation [70.23578202012048]
  Vision-Language Navigation (VLN) presents a unique challenge for Large Vision-Language Models (VLMs) due to their inherent architectural mismatch.<n>We propose TagaVLM (Topology-Aware Global Action reasoning), an end-to-end framework that explicitly injects topological structures into the VLM backbone.<n>To enhance topological node information, an Interleaved Navigation Prompt strengthens node-level visual-text alignment.<n>With the embedded topological graph, the model is capable of global action reasoning, allowing for robust path correction.
  arXiv  Detail & Related papers  (2026-03-03T13:28:07Z)
• VLN-MME: Diagnosing MLLMs as Language-guided Visual Navigation agents [12.383467162169703]
  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.
  arXiv  Detail & Related papers  (2025-12-31T13:21:21Z)
• From Indoor to Open World: Revealing the Spatial Reasoning Gap in MLLMs [65.04549036809557]
  We introduce a benchmark built from pedestrian-perspective videos captured with synchronized stereo cameras, LiDAR, and IMU/GPS sensors.<n>This dataset provides metrically precise 3D information, enabling the automatic generation of spatial reasoning questions.<n> Evaluations reveal that the performance gains observed in structured indoor benchmarks vanish in open-world settings.
  arXiv  Detail & Related papers  (2025-12-22T18:58:12Z)
• Scaling Spatial Reasoning in MLLMs through Programmatic Data Synthesis [8.60591720958037]
  Vision-Language Models (VLMs) are scalable but structurally rigid, while manual annotation is linguistically diverse but unscalable.<n>We introduce SP-RITE, a novel framework that overcomes this dilemma leveraging simulators and large models.<n>We have curated a dataset encompassing 3 simulators, 11k+ scenes, and 300k+ image/video instruction-tuning pairs.<n>We demonstrate that a VLM trained on our data achieves significant performance gains on multiple spatial benchmarks.
  arXiv  Detail & Related papers  (2025-12-18T06:30:08Z)
• Response Wide Shut? Surprising Observations in Basic Vision Language Model Capabilities [54.94982467313341]
  Vision-language Models (VLMs) have emerged as general-purpose tools for addressing a variety of complex computer vision problems.<n>We set out to understand the limitations of SoTA VLMs on fundamental visual tasks by constructing a series of tests that probe which components of design, specifically, may be lacking.
  arXiv  Detail & Related papers  (2025-07-10T15:26:41Z)
• AlignVLM: Bridging Vision and Language Latent Spaces for Multimodal Document Understanding [79.43306110124875]
  AlignVLM is a vision-text alignment method that maps visual features to a weighted average of text embeddings.<n>Our experiments show that AlignVLM achieves state-of-the-art performance compared to prior alignment methods.
  arXiv  Detail & Related papers  (2025-02-03T13:34:51Z)
• 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)
• MC-Bench: A Benchmark for Multi-Context Visual Grounding in the Era of MLLMs [61.56904387052982]
  This paper proposes a new visual grounding task called multi-context visual grounding.<n>It aims to localize instances of interest across multiple images based on open-ended text prompts.<n>We benchmark over 20 state-of-the-art MLLMs and foundation models with potential multi-context visual grounding capabilities.
  arXiv  Detail & Related papers  (2024-10-16T07:52:57Z)
• Response Wide Shut: Surprising Observations in Basic Vision Language Model Capabilities [30.176918208200604]
  Vision-Language Models (VLMs) have emerged as general purpose tools for addressing a variety of complex computer vision problems. These models have been shown to be highly capable, but also lacking some basic visual understanding skills. This paper sets out to understand the limitations of SoTA VLMs on fundamental visual tasks.
  arXiv  Detail & Related papers  (2024-08-13T08:26:32Z)
• SC-Tune: Unleashing Self-Consistent Referential Comprehension in Large Vision Language Models [19.005364038603204]
  We introduce a novel fine-tuning paradigm named Self-Consistency Tuning (SC-Tune) SC-Tune features the synergistic learning of a cyclic describer-locator system. We demonstrate that SC-Tune significantly elevates performance across a spectrum of object-level vision-language benchmarks.
  arXiv  Detail & Related papers  (2024-03-20T03:00:21Z)

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.