AI Paper Atlas

Research map / Benchmarks / POPE Random

POPE Random leaderboard

34 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on POPE Random. Best accuracy: 99.3 (VBackChecker (Ours), AAAI 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.

POPE Random · accuracy

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

#accuracy ↑MethodPaperTopic
199.3VBackChecker (Ours)Seeing Is Believing: Rich-Context Hallucination Detection for MLLMs via Backward Visual Grounding
AAAI 2026 · Table 3
Multimodal LLMs
291.83MGAP (Ours)
Qwen3-VL 8B
Mitigating Manifold Departure: Uncertainty-Aware Subspace Rectification for Trustworthy MLLM Decoding
ICML 2026 · Table 1
Multimodal LLMs
391.45CRG
Qwen2.5-VL-7B-Instruct
Mitigating Hallucinations in Large Vision-Language Models via Causal Route Gating
ICML 2026 · Table 1
Multimodal LLMs
491.03GROUNDHOG-7B
Vicuna-7B
GROUNDHOG: Grounding Large Language Models to Holistic Segmentation
CVPR 2024 · POPE results table
Multimodal LLMs
590.4HulluEdit
LLaVA-1.5-7B
HulluEdit: Single-Pass Evidence-Consistent Subspace Editing for Mitigating Hallucinations in Large Vision-Language Models
CVPR 2026 · Table 1
Multimodal LLMs
690.24Ferret
Vicuna
Ferret: Refer and Ground Anything Anywhere at Any Granularity
ICLR 2024 · Table: Results on the object h
Multimodal LLMs
790.2Owl
LLaVA-1.5
Causally-Grounded Dual-Path Attention Intervention for Object Hallucination Mitigation in LVLMs
AAAI 2026 · Table 2
Multimodal LLMs
890.2LLaVA + linear probing n=1000
LLaVA
The First to Know: How Token Distributions Reveal Hidden Knowledge in Large Vision-Language Models?
ECCV 2024 · Table 3
Multimodal LLMs
990.1VAF
LLaVA-v1.5-13B
ClearSight: Visual Signal Enhancement for Object Hallucination Mitigation in Multimodal Large Language Models
CVPR 2025 · POPE results table
Multimodal LLMs
1089.8Fox
InstructBLIP
Dismantling Pathological Shortcuts: A Causal Framework for Faithful LVLM Decoding
ICML 2026 · Table 1
Multimodal LLMs
12 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

POPE Random · F1

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

#F1 ↑MethodPaperTopic
193.4+EnAR
InternVL3.5-8B
Envision, Attend, Then Respond: Counterfactual Hallucination Mitigation in Large Vision-Language Models
CVPR 2026 · Table 3
Multimodal LLMs
291.43REVIS
Qwen2.5-VL-7B-Instruct
REVIS: Sparse Latent Steering to Mitigate Object Hallucination in Large Vision-Language Models
ICML 2026 · Table 2
Multimodal LLMs
391.3LLaVA-1.5+TDSR
LLaVA-1.5-7B
Top-Down Semantic Refinement for Image Captioning
AAAI 2026 · text
Multimodal LLMs
491.18TAF
Qwen2.5-VL-7B
Taming the Phantom: Token-Asymmetric Filtering for Hallucination Mitigation in Large Vision-Language Models
AAAI 2026 · Table 1
Multimodal LLMs
590.84ObjecTok-LLaVA
Vicuna-7B
ObjecTok: Learning Holistic and Robust Object Tokens for MLLMs
AAAI 2026 · Table 1
Multimodal LLMs
689.68ED
LLaVA-1.5 7B
Do You Keep an Eye on What I Ask? Mitigating Multimodal Hallucination via Attention-Guided Ensemble Decoding
ICLR 2025 · Table: POPE results
Multimodal LLMs
789.55VisFlow
LLaVA-1.5
Not All Tokens and Heads Are Equally Important: Dual-Level Attention Intervention for Hallucination Mitigation
AAAI 2026 · Table 1
Multimodal LLMs
888.5SFT + VIGIL
Qwen2.5-VL-7B
Staying VIGILant: Mitigating Visual Laziness in MLLMs via Information-Theoretic Alignment
ECCV 2026 · Table 1
Multimodal LLMs
988.33LIONLION: Empowering Multimodal Large Language Model with Dual-Level Visual Knowledge
CVPR 2024 · POPE table
Multimodal LLMs
1087.2VCD+ACD
LLaVA-1.5-7B
Beyond Logits: Coherent Hallucination Mitigation via Attention Contrastive Decoding
ICML 2026 · Table 5
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.