Artemis: Structured Visual Reasoning for Perception Policy Learning

TL;DR

Artemis introduces a structured, proposal-based visual reasoning framework for perception policy learning, improving performance and generalization by aligning reasoning with spatial and object-centric representations.

Contribution

The paper presents Artemis, a novel perception-policy learning framework that employs structured, proposal-based reasoning in spatial space, addressing limitations of linguistic reasoning.

Findings

Achieves strong performance on grounding and detection tasks.

Generalizes well to counting and geometric perception tasks.

Improves perception-policy learning through spatially grounded reasoning.

Abstract

Recent reinforcement-learning frameworks for visual perception policy have begun to incorporate intermediate reasoning chains expressed in natural language. Empirical observations indicate that such purely linguistic intermediate reasoning often reduces performance on perception tasks. We argue that the core issue lies not in reasoning per se but in the form of reasoning: while these chains perform semantic reasoning in an unstructured linguistic space, visual perception requires reasoning in a spatial and object-centric space. In response, we introduce Artemis, a perception-policy learning framework that performs structured proposal-based reasoning, where each intermediate step is represented as a (label, bounding-box) pair capturing a verifiable visual state. This design enables explicit tracking of intermediate states, direct supervision for proposal quality, and avoids ambiguity introduced by language-based reasoning. Artemis is built on Qwen2.5-VL-3B, achieves strong performance on grounding and detection task and exhibits substantial generalization to counting and geometric-perception tasks. The consistent improvements across these diverse settings confirm that aligning reasoning with spatial representations enhances perception-policy learning. Owing to its strengthened visual reasoning, Artemis also achieves competitive performance on general MLLM benchmarks, illustrating that spatially grounded reasoning provides a principled route toward scalable and general perception policies.