LASAR: Towards Spatio-temporal Reasoning with Latent Cognitive Map

• URL: http://arxiv.org/abs/2605.16899v1
• Date: Sat, 16 May 2026 09:21:56 GMT
• Title: LASAR: Towards Spatio-temporal Reasoning with Latent Cognitive Map
• Abstract summary: A fundamental in embodied AI is verifying if agents build internal models of structure or merely learn to mimic task-specific expert trajectories.<n>This is critical as foundational approaches rooted in action-centric tasks (e.g., VLN) and reasoningcentric tasks (e.g.A) often share a common limitation: they lack a learning signal that forces them to encode fine-grained spatial relationships over long-range, fragmented experiences.<n>We propose LASAR, an architecture featuring a dual-memory system designed to maintain both episodic experiences and a semantic cognitive map.
• Score: 12.196103884300753
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: A fundamental challenge in embodied AI is verifying if agents build internal models of spatial structure or merely learn to mimic task-specific expert trajectories. This is critical as foundational approaches rooted in action-centric tasks (e.g., VLN) and reasoning-centric tasks (e.g., EQA) often share a common limitation: they lack a learning signal that forces them to encode fine-grained spatial relationships (like topology or distance) over long-range, fragmented experiences. To address this, we first propose LASAR, an architecture featuring a dual-memory system designed to maintain both episodic experiences and a semantic cognitive map. We then introduce Spatio-temporal Contextual Representation Learning (ST-CRL), a contrastive objective designed to train this architecture. ST-CRL leverages spatio-temporal cues from cognitive queries generated through annotated spatio-temporal context in simulation to build sample pairs, thereby forming the internal cognitive map from the agent's experiences. Experiments demonstrate that our method achieves 2\%-3.5\% gains in both zero-shot generalization on standard VLN-CE and VSI-Bench benchmarks. We also demonstrate that our proposed cognitive map has high self-consistency.

Related papers

• OmniView-Space: Reinforcing Spatial Reasoning via Multi-Perspective Spatial Mapping [69.48762031013514]
  We propose OmniView-Space, a framework designed to maintain spatial consistency through multimodal egocentric evidence.<n>Our approach consists of three core components: (1) Multi-Perspective Spatial Mapping (MPSM), which re-anchors reconstructed geometry into a query-aligned visual cognitive map and a textual spatial graph; (2) Tool-Guided Egocentric Reasoning, an interleaved policy trained to actively select the ego anchor required by the query and request the corresponding MPSM evidence; and (3) Cognitive-Map Distillation, which uses MPSM-generated trajectories and ego-frame rewards to train the model to reason with self-generated cognitive maps
  arXiv  Detail & Related papers  (2026-07-01T12:45:12Z)
• From Where Things Are to What They Are For: Benchmarking Spatial-Functional Intelligence in Multimodal LLMs [19.943841049221625]
  We introduce the Spatial-Functional Intelligence Benchmark (SFI-Bench), a video-based benchmark with over 1,500 expert-annotated questions.<n>SFI-Bench systematically evaluates two complementary dimensions of advanced reasoning.
  arXiv  Detail & Related papers  (2026-05-04T01:19:36Z)
• From Perception to Planning: Evolving Ego-Centric Task-Oriented Spatiotemporal Reasoning via Curriculum Learning [79.65573577667944]
  We present EgoTSR, a curriculum-based framework for learning task-basedtemporal reasoning.<n>EgoTSR is built on the premise embodied reasoning should evolve from explicit spatial understanding to task-state assessment.<n>Experiments demonstrate that EgoTSR effectively eliminates biases, achieving 92.4% accuracy on long-horizon logical reasoning tasks.
  arXiv  Detail & Related papers  (2026-04-12T08:14:49Z)
• Spatial-VLN: Zero-Shot Vision-and-Language Navigation With Explicit Spatial Perception and Exploration [16.651645602449577]
  Vision-and-Language Navigation (VLN) agents leveraging Large Language Models (LLMs) excel in generalization but suffer from insufficient spatial perception.<n>We present Spatial-VLN, a perception-guided exploration framework designed to overcome these challenges.
  arXiv  Detail & Related papers  (2026-01-19T06:53:02Z)
• Origin-Conditional Trajectory Encoding: Measuring Urban Configurational Asymmetries through Neural Decomposition [3.4560970433900366]
  We introduce a conditional trajectory encoder that learns spatial and movement representations while preserving origin-dependent asymmetries.<n>This framework decomposes urban navigation into shared cognitive patterns and origin-specific spatial narratives.<n>Results from six synthetic cities and real-world validation on Beijing's Xicheng District demonstrate that urban morphology creates systematic cognitive inequalities.
  arXiv  Detail & Related papers  (2025-12-03T12:54:16Z)
• How Far are VLMs from Visual Spatial Intelligence? A Benchmark-Driven Perspective [103.44502230776352]
  We present a systematic investigation of Visual Spatial Reasoning (VSR) in Vision-Language Models (VLMs)<n>We categorize spatial intelligence into three levels of capability, ie, basic perception, spatial understanding, spatial planning, and curate SIBench, a spatial intelligence benchmark encompassing nearly 20 open-source datasets across 23 task settings.
  arXiv  Detail & Related papers  (2025-09-23T12:00:14Z)
• Beyond Pixels: Introducing Geometric-Semantic World Priors for Video-based Embodied Models via Spatio-temporal Alignment [2.9493863710375674]
  VEME is a novel method for achieving human-like reasoning in deep learning models for complex tasks in unknown environments.<n>Our framework integrates three key components: (1) a cross-language alignment framework bridging objects, spatial representations, and visual semantics with-temporal cues; (2) a dynamic, implicit cognitive activated world embedding to enable task-relevants memory recall; and (3) instruction instruction-based navigation and reasonings for long-term planning and efficient exploration.
  arXiv  Detail & Related papers  (2025-08-29T19:47:25Z)
• CogDual: Enhancing Dual Cognition of LLMs via Reinforcement Learning with Implicit Rule-Based Rewards [53.36917093757101]
  Role-Playing Language Agents (RPLAs) have emerged as a significant application direction for Large Language Models (LLMs)<n>We introduce textbfCogDual, a novel RPLA adopting a textitcognize-then-respond reasoning paradigm.<n>By jointly modeling external situational awareness and internal self-awareness, CogDual generates responses with improved character consistency and contextual alignment.
  arXiv  Detail & Related papers  (2025-07-23T02:26:33Z)
• A theoretical framework for self-supervised contrastive learning for continuous dependent data [79.62732169706054]
  Self-supervised learning (SSL) has emerged as a powerful approach to learning representations, particularly in the field of computer vision.<n>We propose a novel theoretical framework for contrastive SSL tailored to emphsemantic independence between samples.<n>Specifically, we outperform TS2Vec on the standard UEA and UCR benchmarks, with accuracy improvements of $4.17$% and $2.08$%, respectively.
  arXiv  Detail & Related papers  (2025-06-11T14:23:47Z)
• SITE: towards Spatial Intelligence Thorough Evaluation [121.1493852562597]
  Spatial intelligence (SI) represents a cognitive ability encompassing the visualization, manipulation, and reasoning about spatial relationships.<n>We introduce SITE, a benchmark dataset towards SI Thorough Evaluation.<n>Our approach to curating the benchmark combines a bottom-up survey about 31 existing datasets and a top-down strategy drawing upon three classification systems in cognitive science.
  arXiv  Detail & Related papers  (2025-05-08T17:45:44Z)
• Learning Appearance-motion Normality for Video Anomaly Detection [11.658792932975652]
  We propose spatial-temporal memories augmented two-stream auto-encoder framework. It learns the appearance normality and motion normality independently and explores the correlations via adversarial learning. Our framework outperforms the state-of-the-art methods, achieving AUCs of 98.1% and 89.8% on UCSD Ped2 and CUHK Avenue datasets.
  arXiv  Detail & Related papers  (2022-07-27T08:30:19Z)

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.