# Relation-R1: Progressively Cognitive Chain-of-Thought Guided Reinforcement Learning for Unified Relation Comprehension

arXiv: https://arxiv.org/abs/2504.14642 · HTML: https://mlsanity.com/paper/2504.14642/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Placed in Multimodal reasoning, one of 17 directions in LLM reasoning.
- Builds on 12 of its 54 references that are in the map.
- First posted on arXiv 20 Apr 2025.

## Placed in

- LLM reasoning / [Multimodal reasoning](https://mlsanity.com/lr/multimodal-reasoning/)
- Multimodal LLMs / vision-language models / [VLM benchmarks and visual reasoning](https://mlsanity.com/mllm/vlm-benchmarks-and-visual-reasoning/)

## Builds on (most cited first, 12 of 12)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [OpenAI o1 System Card](https://mlsanity.com/paper/2412.16720/) (2024)
- [SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training](https://mlsanity.com/paper/2501.17161/) (2025)
- [Vision-R1: Incentivizing Reasoning Capability in Multimodal Large Language Models](https://mlsanity.com/paper/2503.06749/) (2025)
- [Visual-RFT: Visual Reinforcement Fine-Tuning](https://mlsanity.com/paper/2503.01785/) (2025)
- [R1-Onevision: Advancing Generalized Multimodal Reasoning through Cross-Modal Formalization](https://mlsanity.com/paper/2503.10615/) (2025)
- [MM-Eureka: Exploring the Frontiers of Multimodal Reasoning with Rule-based Reinforcement Learning](https://mlsanity.com/paper/2503.07365/) (2025)
- [Seg-Zero: Reasoning-Chain Guided Segmentation via Cognitive Reinforcement](https://mlsanity.com/paper/2503.06520/) (2025)
- [Virgo: A Preliminary Exploration on Reproducing o1-like MLLM](https://mlsanity.com/paper/2501.01904/) (2025)
- [Compile Scene Graphs with Reinforcement Learning](https://mlsanity.com/paper/2504.13617/) (2025)

## Built on by (most cited first, 9 of 9)

- [WeThink: Toward General-purpose Vision-Language Reasoning via Reinforcement Learning](https://mlsanity.com/paper/2506.07905/) (2025)
- [SpatialThinker: Reinforcing Scene Graph-Grounded Spatial Reasoning via Dense Rewards](https://mlsanity.com/paper/2511.07403/) (2025)
- [Reinforcement Fine-Tuning Powers Reasoning Capability of Multimodal Large Language Models](https://mlsanity.com/paper/2505.18536/) (2025)
- [Rethinking Token-Level Policy Optimization for Multimodal Chain-of-Thought](https://mlsanity.com/paper/2603.22847/) (2026)
- [HOID-R1: Reinforcement Learning for Open-World Human-Object Interaction Detection Reasoning with Multimodal Large Language Model](https://mlsanity.com/paper/2508.11350/) (2025)
- [SinkTrack: Attention Sink based Context Anchoring for Large Language Models](https://mlsanity.com/paper/2604.10027/) (2026)
- [Anatomy-R1: Enhancing Anatomy Reasoning in Multimodal Large Language Models via Anatomical Similarity Curriculum and Group Diversity Augmentation](https://mlsanity.com/paper/2512.19512/) (2025)
- [Interaction-Centric Knowledge Infusion and Transfer for Open-Vocabulary Scene Graph Generation](https://mlsanity.com/paper/2511.05935/) (2025)
- [SGG-Rrm 3: From Next-Token Prediction to End-to-End Unbiased Scene Graph Generation](https://mlsanity.com/paper/2603.07961/) (2026)

## Abstract (arXiv)

Recent advances in multi-modal large language models (MLLMs) have significantly improved object-level grounding and region captioning. However, they remain limited in visual relation understanding, struggling even with binary relation detection, let alone textitN-ary relations involving multiple semantic roles. The core reason is the lack of modeling for textitstructural semantic dependencies among multi-entities, leading to unreliable outputs, hallucinations, and over-reliance on language priors (eg, defaulting to ``person drinks a milk'' if a person is merely holding it). To this end, we propose Relation-R1, the textitfirst unified relation comprehension framework that explicitly integrates cognitive chain-of-thought (CoT)-guided supervised fine-tuning (SFT) and group relative policy optimization (GRPO) within a reinforcement learning (RL) paradigm. Specifically, we first establish foundational reasoning capabilities via SFT, enforcing structured outputs with thinking processes. Then, GRPO is utilized to refine these outputs via multi-rewards optimization, prioritizing visual-semantic grounding over language-induced biases, thereby improving generalization capability. Furthermore, we investigate the impact of various CoT strategies within this framework, demonstrating that a specific-to-general progressive approach in CoT guidance further improves generalization, especially in capturing synonymous textitN-ary relations. Extensive experiments on widely-used PSG and SWiG datasets demonstrate that Relation-R1 achieves state-of-the-art performance in both binary and textitN-ary relation understanding.