AI Paper Atlas

Research map / Benchmarks / HR-Bench 8K

HR-Bench 8K leaderboard

12 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on HR-Bench 8K. Best accuracy: 79.2 (FOVEA, ICML 2026).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

HR-Bench 8K · accuracy

12 papers, each with its best reported number for its own method; higher is better.

#accuracy ↑MethodPaperTopic
179.2FOVEA
Qwen3-VL-30B-A3B-Instruct
The Perceptual Bandwidth Bottleneck in Vision-Language Models: Active Visual Reasoning via Sequential Experimental Design
ICML 2026 · Table 1
Multimodal LLMs
274.3PyVision-Image
Qwen2.5-VL-7B
PyVision-RL: Forging Open Agentic Vision Models via RL
ICML 2026 · Table 1
Multimodal LLMs
373.7ARM-Thinker-7B
Qwen2.5-VL-7B
ARM-Thinker: Reinforcing Multimodal Generative Reward Models with Agentic Tool Use and Visual Reasoning
CVPR 2026 · Table 2
Multimodal LLMs
473.3Mini-o3 (Ours)
Qwen2.5-VL-7B-Instruct
Mini-o3: Scaling Up Reasoning Patterns and Interaction Turns for Visual Search
ICLR 2026 · Table 1
Multimodal LLMs
572.3CodeDance-7B
Qwen2.5-VL-7B
CodeDance: A Dynamic Tool-integrated MLLM for Executable Visual Reasoning
CVPR 2026 · Table 1
Multimodal LLMs
672.1VisionLeaf
Qwen2.5-VL-7B
VisionLeaf: Entropy-Guided Leaf-First Reasoning for Efficient and Accurate Think-with-Image
CVPR 2026 · Table 1
Reasoning and chain of thought
771.9HART-7B
Qwen2.5-VL-7B
HART: High-Resolution Annotation-Free Reasoning Technique through a Closed-loop Framework
ECCV 2026 · Table 3
Multimodal LLMs
871.8SpecEyes (min)
DeepEyes + Qwen3-VL-2B
SpecEyes: Accelerating Agentic Multimodal LLM via Speculative Planning and Perception
ECCV 2026 · Table 1
Multimodal LLMs
970.1VisReflect
Qwen2.5-VL-7B
VisReflect: Latent Visual Reflection for Fine-Grained Perception in Long Visual Context
ECCV 2026 · Table 1
Multimodal LLMs
1069.9CoVT (4 visual tokens)
Qwen2.5-VL-7B
Chain-of-Visual-Thought: Teaching VLMs to See and Think Better with Continuous Visual Tokens
ECCV 2026 · Table 1
Multimodal LLMs
2 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.