Thinker: A vision-language foundation model for embodied intelligence
On the map: LLM reasoning → Spatial and embodied reasoning
Abstract (arXiv)
When large vision-language models are applied to the field of robotics, they encounter problems that are simple for humans yet error-prone for models. Such issues include confusion between third-person and first-person perspectives and a tendency to overlook information in video endings during temporal reasoning. To address these challenges, we propose Thinker, a large vision-language foundation model designed for embodied intelligence. We tackle the aforementioned issues from two perspectives. Firstly, we construct a large-scale dataset tailored for robotic perception and reasoning, encompassing ego-view videos, visual grounding, spatial understanding, and chain-of-thought data. Secondly, we introduce a simple yet effective approach that substantially enhances the model's capacity for video comprehension by jointly incorporating key frames and full video sequences as inputs. Our model achieves state-of-the-art results on two of the most commonly used benchmark datasets in the field of task planning.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning. Builds on 16 of its 18 references that are in the map. First posted on arXiv 29 Jan 2026.
Builds on
16 of 18 references.
Closest ancestors · 5
Same direction, newest first
- 2025ThinkAct: Vision-Language-Action Reasoning via Reinforced Visual Latent Planning22 Jul
- 2025Cosmos-Reason1: From Physical Common Sense To Embodied Reasoning18 Mar
- 2024EgoPlan-Bench2: A Benchmark for Multimodal Large Language Model Planning in Real-World Scenarios5 Dec
- 2023EgoPlan-Bench: Benchmarking Multimodal Large Language Models for Human-Level Planning11 Dec
- 2023RoboVQA: Multimodal Long-Horizon Reasoning for Robotics1 Nov
Other references · 11
- 2023GPT-4 Technical Reportcited by 18,396
- 2025Qwen2.5-VL Technical Reportcited by 5,587
- 2024GPT-4o System Cardcited by 4,588
- 2021LAION-400M: Open Dataset of CLIP-Filtered 400 Million Image-Text Pairscited by 1,555
- 2023Open X-Embodiment: Robotic Learning Datasets and RT-X Modelscited by 862
- 2024LLaVA-CoT: Let Vision Language Models Reason Step-by-Stepcited by 454
- 2024Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Modelscited by 329
- 2024RoboPoint: A Vision-Language Model for Spatial Affordance Prediction for Roboticscited by 251
- 2023PACO: Parts and Attributes of Common Objectscited by 146
- 2025RoboBrain: A Unified Brain Model for Robotic Manipulation from Abstract to Concretecited by 135
- 2025RoboBrain 2.0 Technical Reportcited by 89
Built on by
Citing papers per year: 2026: 1.
1 paper.
Similar, not cited
Close in wording, no citation either way
- 2026VLA-Thinker: Boosting Vision-Language-Action Models through Thinking-with-Image Reasoning
- 2025UrbanVideo-Bench: Benchmarking Vision-Language Models on Embodied Intelligence with Video Data in Urban Spaces
- 2023EmbodiedGPT: Vision-Language Pre-Training via Embodied Chain of Thought
- 2026Vision-Language-Action in Robotics: A Survey of Datasets, Benchmarks, and Data Engines
- 2025ROVER: Recursive Reasoning Over Videos with Vision-Language Models for Embodied Tasks
- 2025Embodied-R: Collaborative Framework for Activating Embodied Spatial Reasoning in Foundation Models via Reinforcement Learning
- 2026Capek 0.5: An Execution-Centric Vision-Language Model for Embodied Intelligence
- 2025ENACT: Evaluating Embodied Cognition with World Modeling of Egocentric Interaction