Research map / Vision-language and multimodal / Robot action and navigation
LASAR: Towards Spatio-temporal Reasoning with Latent Cognitive Map
Jinzhou Tang, Sidi Liu, Waikit Xiu, Weixing Chen, Keze Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add ST-CRL loss (ablation) | LASAR (IL+QA) · MindCraft-Test | QA-Acc: 60.6 -> 63.5 | Improved |
| add semantic atlas and episodic discriminability losses (ablation) | LASAR with ST-CRL · MindCraft-Test | QA-Acc: 63.5 -> 65.3 | Improved |
| add geometric feature fusion (VGGT) (ablation) | LASAR · MindCraft-Test | GCA: 62.1 -> 70.4 | Improved |
| add geometric feature fusion (VGGT) (ablation) | LASAR · MindCraft-Test | CMC: 66.3 -> 75.8 | Improved |
| add Spatial Semantic Memory (latent cognitive map) (ablation) | LASAR · MindCraft-Test | CMC: 58.2 -> 75.8 | Improved |
| add Spatial Semantic Memory (latent cognitive map) (ablation) | LASAR · MindCraft-Test | SR@WA: 45.7 -> 35.2 | Improved |
| add Spatial Semantic Memory (latent cognitive map) (ablation) | LASAR · MindCraft-Test | QA-Acc: 62.1 -> 65.3 | Improved |
| add ST-CRL and auxiliary losses to IL+QA training | LASAR (IL+QA) · VSI-Bench (zero-shot) | Avg: 44.8 -> 48.9 | Improved |
| add MindCraft QA training to imitation-only | LASAR (IL) · VSI-Bench (zero-shot) | Avg: 37.8 -> 44.8 | Improved |
| add QA loss to imitation-only training | LASAR (IL) · R2R val-unseen | SPL: 52.7 -> 47.4 | Hurt |
| add ST-CRL and auxiliary losses to IL+QA training | LASAR (IL+QA) · R2R val-unseen | SPL: 47.4 -> 53.9 | Improved |
| use LASAR instead of NaVILA | navigation on R2R · R2R val-unseen | SR: 54.8 -> 57.0 | Improved |
Closest papers by meaning
- Active Exploring like a Pigeon: Reinforcing Spatial Reasoning via Agentic Vision-Language Models (ICML 2026 · similarity 0.88)
- SpatialClaw: Rethinking Action Interface for Agentic Spatial Reasoning (NeurIPS 2026 · similarity 0.87)
- EvoEmpirBench: Dynamic Spatial Reasoning with Agent-ExpVer (AAAI 2026 · similarity 0.87)
- From Where Things Are to What They Are For: Benchmarking Spatial-Functional Intelligence in Multimodal LLMs (CVPR 2026 · similarity 0.87)
- Linear Mechanisms for Spatiotemporal Reasoning in Vision Language Models (ICLR 2026 · similarity 0.87)
- OST-Bench: Evaluating the Capabilities of MLLMs in Online Spatio-temporal Scene Understanding (NeurIPS 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
