AI Paper Atlas

Research map / Benchmarks / HallusionBench

HallusionBench leaderboard

53 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on HallusionBench. Best accuracy: 92.32 (VisPlay (Iter 3), CVPR 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.

HallusionBench · accuracy

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

#accuracy ↑MethodPaperTopic
192.32VisPlay (Iter 3)
Qwen2.5-VL-7B-Instruct
VisPlay: Self-Evolving Vision-Language Models
CVPR 2026 · Table 1
Multimodal LLMs
272.9Durian on GRPO
Qwen2.5-VL-7B-Instruct; Geometry3K (2.1K) RL training
Enhancing Multi-Modal LLMs Reasoning via Difficulty-Aware Group Normalization
ICML 2026 · Table 1
Multimodal LLMs
371.8Video-FLAIR
Qwen2.5-VL-7B
Video-FLAIR: Not Whether to Reason, But How
NeurIPS 2026 · Table 1
Video-language understanding
471Visual Para-Thinker
Qwen2.5-VL-7B
Visual Para-Thinker: Divide-and-Conquer Reasoning for Visual Comprehension
ICML 2026 · Table 1
Multimodal LLMs
570.7Semantic-back-7B
Qwen2.5-VL-7B
Look-Back: Implicit Visual Re-focusing in MLLM Reasoning
AAAI 2026 · Table 3
Multimodal LLMs
670.6MSSR
Qwen2.5-VL-7B
Stable and Efficient Single-Rollout RL for Multimodal Reasoning
CVPR 2026 · Table 1
Reasoning and chain of thought
770.2OpenVLThinker (Iterative)
Qwen2.5-VL-7B
OpenVLThinker: Complex Vision-Language Reasoning via Iterative SFT-RL Cycles
NeurIPS 2025 · Single-stage baselines table
Multimodal LLMs
869.4Qwen3-VL-30B + VRGA
Qwen3-VL-30B
Deeper Thought, Weaker Aim: Understanding and Mitigating Perceptual Impairment during Reasoning in Multimodal Large Language Models
CVPR 2026 · Table 2
Visual question answering
968.98DPE (Iter 3)
Qwen2.5-VL-7B-Instruct
From Blind Spots to Gains: Diagnostic-Driven Iterative Training for Large Multimodal Models
ICML 2026 · Table 1
RL from feedback
1068.63Ours
Qwen2.5-VL-7B
Visual Enhanced Depth Scaling for Multimodal Latent Reasoning
NeurIPS 2026 · Table 2
Multimodal LLMs
22 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

HallusionBench · score

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

#score ↑MethodPaperTopic
166.6UVU
Qwen2.5-3B-Instruct
UVU: Improving Multimodal Understanding via Vision-Language Unified Autoregressive Paradigm
CVPR 2026 · Table 4
Multimodal LLMs
266.4SEEA-R1
Qwen2.5-VL-7B
SEEA-R1: Tree-Structured Reinforcement Fine-Tuning for Self-Evolving Embodied Agents
NeurIPS 2025 · Table: Performance comparison
LLM agents
362VLM-R³
Qwen2.5-VL 7B
VLM-R³: Region Recognition, Reasoning, and Refinement for Enhanced Multimodal Chain-of-Thought
NeurIPS 2025 · Main Result table
Multimodal LLMs
460.4Meteor-7B
InternLM2-7B
Meteor: Mamba-based Traversal of Rationale for Large Language and Vision Models
NeurIPS 2024 · Table 1
Multimodal LLMs
560.4InternLM-XC2 + MMDU-45k
InternLM-XC2
MMDU: A Multi-Turn Multi-Image Dialog Understanding Benchmark and Instruction-Tuning Dataset for LVLMs
NeurIPS 2024 · Fine-tuning Results, Table
Multimodal LLMs
655.7Qwen2.5VL-7B + IPA
Qwen2.5VL-7B
Instruction-Oriented Preference Alignment for Enhancing Multi-Modal Comprehension Capability of MLLMs
ICCV 2025 · Table 1
Multimodal LLMs
754.7R1-VL-7B
Qwen2-VL-7B
R1-VL: Learning to Reason with Multimodal Large Language Models via Step-wise Group Relative Policy Optimization
ICCV 2025 · Table 1
Reasoning and chain of thought
854.2SAIL
Mistral-7B-v0.1
The Scalability of Simplicity: Empirical Analysis of Vision-Language Learning with a Single Transformer
ICCV 2025 · Table 2
Cross-modal learning
953.6Qwen2.5-VL +Cold Start +Multimodal RL
Qwen2.5-VL
Open Vision Reasoner: Transferring Linguistic Cognitive Behavior for Visual Reasoning
NeurIPS 2025 · Perception benchmark table
Multimodal LLMs
1051.31MAA
LLaVA-1.6-Mistral-7B
Robust Vision-Language Models via Manifold-Adversarial Adapters
ICML 2026 · Table 1
Multimodal LLMs
11 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.