| Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation | | 86.88 | | | Gemini-3.1-Pro | 2026-05-18 |
| Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation | | 86.62 | | | Gemini-3.5-Flash | 2026-05-18 |
| SenseNova-MARS: Empowering Multimodal Agentic Reasoning and Search via Reinforcement Learning | | 86.6 | | | SenseNova-MARS-32B | 2025-12-30 |
| Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation | | 85.50 | | | Qwen3.5-397B-A17B | 2026-05-18 |
| Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation | | 85.50 | | | Vision-OPD-9B | 2026-05-18 |
| Zooming without Zooming: Region-to-Image Distillation for Fine-Grained Multimodal Perception | | 82.00 | | | ZwZ-8B | 2026-02-12 |
| Learning Active Perception for Pixel-Space Reasoning via Visual-Intent Stratified GRPO | | 81.6 | | | VIS-GRPO | 2026-09-08 |
| Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation | | 80.63 | | | Qwen3.5-9B | 2026-05-18 |
| Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation | | 80.38 | | | Qwen3-VL-235B-A22B-Instruct | 2026-05-18 |
| Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation | | 80.38 | | | Vision-OPD-4B | 2026-05-18 |
| Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation | | 80.13 | | | Qwen3.5-4B | 2026-05-18 |
| Zooming without Zooming: Region-to-Image Distillation for Fine-Grained Multimodal Perception | | 79.50 | | | ZwZ-4B | 2026-02-12 |
| Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation | | 78.88 | | | GLM-4.6V | 2026-05-18 |
| SenseNova-MARS: Empowering Multimodal Agentic Reasoning and Search via Reinforcement Learning | | 78.4 | | | SenseNova-MARS-8B | 2025-12-30 |
| Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation | | 78.38 | | | GPT-5.2 | 2026-05-18 |
| Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation | | 78.00 | | | Kimi-K2.6 | 2026-05-18 |
| Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation | | 77.88 | | | GPT-5.4 | 2026-05-18 |
| DeepScan: A Training-Free Framework for Visually Grounded Reasoning in Large Vision-Language Models | | 76.3 | 84.3 | 68.3 | Qwen2.5-VL-72B | 2026-03-04 |
| Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation | | 75.25 | | | Qwen3-VL-8B-Instruct | 2026-05-18 |
| Region-Level Policy Optimization for Fine-grained MLLM Perception | ✓ Link | 75.0 | | | Vision-RL 2 | 2026-09-17 |
| DeepEyesV2: Toward Agentic Multimodal Model | | 73.8 | | | DeepEyesV2 | 2025-11-07 |
| Zooming without Zooming: Region-to-Image Distillation for Fine-Grained Multimodal Perception | | 73.25 | | | ZwZ-7B | 2026-02-12 |
| DeepEyes: Incentivizing "Thinking with Images" via Reinforcement Learning | ✓ Link | 72.6 | 86.8 | 58.5 | DeepEyes | 2025-05-20 |
| DeepScan: A Training-Free Framework for Visually Grounded Reasoning in Large Vision-Language Models | | 72.4 | 87.2 | 57.6 | DeepScan | 2026-03-04 |
| DeepEyesV2: Toward Agentic Multimodal Model | | 72.0 | | | Thyme | 2025-11-07 |
| DeepEyes: Incentivizing "Thinking with Images" via Reinforcement Learning | ✓ Link | 70.4 | 84.5 | 56.3 | Qwen2.5-VL-32B | 2025-05-20 |
| DeepScan: A Training-Free Framework for Visually Grounded Reasoning in Large Vision-Language Models | | 69.8 | 84.4 | 57.2 | TreeVGR | 2026-03-04 |
| DeepScan: A Training-Free Framework for Visually Grounded Reasoning in Large Vision-Language Models | | 69.8 | 86.5 | 53.2 | DyFo | 2026-03-04 |
| Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation | | 69.38 | | | MiMo-VL-7B-RL | 2026-05-18 |
| DeepScan: A Training-Free Framework for Visually Grounded Reasoning in Large Vision-Language Models | | 68.6 | 83.9 | 54.0 | ZoomRefine | 2026-03-04 |
| DeepScan: A Training-Free Framework for Visually Grounded Reasoning in Large Vision-Language Models | | 67.3 | 71.8 | 62.8 | InternVL3-78B | 2026-03-04 |
| DeepScan: A Training-Free Framework for Visually Grounded Reasoning in Large Vision-Language Models | | 67.0 | 71.3 | 62.8 | InternVL3-38B | 2026-03-04 |
| DeepEyes: Incentivizing "Thinking with Images" via Reinforcement Learning | ✓ Link | 66.9 | 80.0 | 54.3 | Pixel-Reasoner | 2025-05-20 |
| DeepEyes: Incentivizing "Thinking with Images" via Reinforcement Learning | ✓ Link | 65.3 | 78.8 | 51.8 | Qwen2.5-VL-7B | 2025-05-20 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 62.8 | 69.2 | 56.7 | Gemini 1.5 Flash | 2024-08-28 |
| DeepScan: A Training-Free Framework for Visually Grounded Reasoning in Large Vision-Language Models | | 62.0 | 64.3 | 59.8 | InternVL3-8B | 2026-03-04 |
| Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation | | 61.50 | | | MiniCPM-V-4.5 | 2026-05-18 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 61.4 | 69.0 | 53.8 | InternVL-2-llama3-76B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 61.3 | 75.0 | 47.5 | DC² (InternVL-1.5-26B) | 2024-08-28 |
| DeepScan: A Training-Free Framework for Visually Grounded Reasoning in Large Vision-Language Models | | 60.9 | 68.8 | 53.0 | LLaVA-OneVision-72B | 2026-03-04 |
| DeepEyes: Incentivizing "Thinking with Images" via Reinforcement Learning | ✓ Link | 59.8 | 67.3 | 52.3 | LLaVA-OneVision-7B | 2025-05-20 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 57.9 | 69.3 | 48.5 | InternVL-1.5-26B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 55.5 | 62.0 | 49.0 | GPT-4o | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 52.5 | 54.0 | 51.0 | Qwen-VL-Max | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 51.3 | 55.3 | 47.3 | Xcomposer2-4kHD-7B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 47.4 | 44.5 | 50.3 | LLaVA-v1.6-34B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 46.9 | 49.5 | 44.3 | LLaVA-HR-X-13B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 46.5 | 52.0 | 41.0 | Qwen-VL-Plus | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 42.3 | 40.5 | 45.0 | DC² (LLaVA-v1.6-7B) | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 41.6 | 42.0 | 41.3 | LLaVA-HR-X-7B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 41.1 | 42.5 | 39.8 | CogVLM-Chat-17B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 40.8 | 37.2 | 44.2 | LLaVA-v1.6-7B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 40.5 | 43.3 | 37.8 | Phi3-Vision-4.2B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 40.5 | 40.0 | 41.0 | DC² (LLaVA-v1.5-13B) | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 40.0 | 39.0 | 41.0 | DC² (Yi-VL-6B) | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 39.0 | 39.5 | 38.5 | Yi-VL-34B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 38.9 | 38.5 | 39.3 | Yi-VL-6B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 38.1 | 38.0 | 38.3 | LLaVA-v1.6-13B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 37.9 | 36.0 | 39.8 | InternLM-Xcomposer2-7B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 37.8 | 37.5 | 38.0 | LLaVA-v1.5-13B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 35.6 | 42.5 | 28.8 | SEAL-7B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 35.1 | 34.5 | 35.8 | InternLM-Xcomposer-7B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 33.8 | 33.8 | 33.8 | MPLUG2-7B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 33.4 | 33.8 | 33.0 | Deepseek-VL-7B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 32.1 | 33.0 | 31.3 | LLaVA-v1.5-7B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 28.8 | 34.8 | 22.8 | idefics-9B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 28.4 | 28.8 | 28.0 | Qwen-VL-7B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 26.1 | 24.8 | 27.5 | fuyu-8B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 26.1 | 26.0 | 26.3 | minigptv2-7B | 2024-08-28 |
| Divide, Conquer and Combine: A Training-Free Framework for High-Resolution Image Perception in Multimodal Large Language Models | | 19.0 | 19.5 | 18.5 | VisualGLM-6B | 2024-08-28 |