The paper proposes Astra, an agentic framework that equips Vision-Language Models (VLMs) with the ability to perform spatial reasoning by actively generating and utilizing imagined visual evidence from a world simulator.
While Vision-Language Models (VLMs) have shown strong visual reasoning capabilities, their spatial reasoning abilities remain largely constrained to the observed images and text-oriented chain-of-thought. They often struggle to infer unobserved layouts, maintain cross-view consistency, and reason from alternative viewpoints when only limited egocentric observations are available. In this work, we study this problem as thinking with imagination, where a VLM actively acquires imagined visual evidence by interacting with a world simulator during reasoning. We propose Astra, an agentic spatial reasoning framework that empowers VLMs with action-conditioned visual imagination. Specifically, Astra couples Astra-VL, an RL-trained VLM policy, with Astra-WM, a Bagel-based world simulator that generates novel-view observations from context images and natural-language camera motions. To provide reliable imagined evidence, Astra-WM is trained with view consistency tuning to improve pose and content consistency across views. In the RL stage, we propose a world-simulator-in-the-loop two-phase RL curriculum to stabilize tool-use exploration and advance the model's ability to invoke the simulator only when imagined observations improve over direct answering. Experiments demonstrate that both the world simulator and the agentic policy are necessary: Astra-WM improves simulator-augmented Gemini-3-Flash on MMSI-Bench from 45.1 to 49.5, while Astra-VL improves the Qwen3-VL backbone from 29.8 to 38.8 on MMSI-Bench and from 36.8 to 42.7 on MindCube. These results show that imagined observations can provide useful spatial evidence, but effective world-model-augmented reasoning requires learning when, where, and how to imagine.
Beyond Content Safety: Real-Time Monitoring for Reasoning Vulnerabilities in Large Language Models
This paper introduces a novel framework, the Reasoning Safety Monitor, to detect…
Strengthening Human-Centric Chain-of-Thought Reasoning Integrity in LLMs via a Structured Prompt Fra…
The paper proposes a structured prompt engineering framework to enhance the inte…
Critical-CoT: A Robust Defense Framework against Reasoning-Level Backdoor Attacks in Large Language…
The paper introduces Critical-CoT, a novel two-stage fine-tuning defense framewo…
WebAgentGuard: A Reasoning-Driven Guard Model for Detecting Prompt Injection Attacks in Web Agents
The paper introduces WebAgentGuard, a novel reasoning-driven, multimodal guard m…
Safety, Security, and Cognitive Risks in World Models
This paper surveys the risks associated with world models, proposing a unified t…
TRAP: Hijacking VLA CoT-Reasoning via Adversarial Patches
This paper introduces TRAP, an adversarial attack that demonstrates how physical…
KidsNanny: A Two-Stage Multimodal Content Moderation Pipeline Integrating Visual Classification, Obj…
KidsNanny is a two-stage multimodal content moderation pipeline that achieves hi…
SecPI: Secure Code Generation with Reasoning Models via Security Reasoning Internalization
The paper introduces SecPI, a fine-tuning pipeline that teaches reasoning langua…