AI Paper Atlas

Research map / Benchmarks / POPE MSCOCO Adversarial

POPE MSCOCO Adversarial leaderboard

10 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on POPE MSCOCO Adversarial. Best accuracy: 84.32 (HIO, NeurIPS 2024).

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 MSCOCO Adversarial · accuracy

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

#accuracy ↑MethodPaperTopic
184.32HIO
LLaVA1.5
Alleviating Hallucinations in Large Vision-Language Models through Hallucination-Induced Optimization
NeurIPS 2024 · POPE table
Multimodal LLMs
283.93FuzzyCD
LLaVA-1.5 (Vicuna-7B)
Fuzzy Contrastive Decoding to Alleviate Object Hallucination in Large Vision-Language Models
ICCV 2025 · Table 2
Multimodal LLMs
383.9AIR
LLaVA-1.5-7B
Look Carefully: Adaptive Visual Reinforcements in Multimodal Large Language Models for Hallucination Mitigation
ICLR 2026 · Table 2
Multimodal LLMs
483.7CausalMM Multimodal
LLaVA-1.5
Mitigating Modality Prior-Induced Hallucinations in Multimodal Large Language Models via Deciphering Attention Causality
ICLR 2025 · Table: Main results on POPE ta
Multimodal LLMs
582.22Octopus
LLaVA-1.5-7B
Octopus: Alleviating Hallucination via Dynamic Contrastive Decoding
CVPR 2025 · Table 2
Multimodal LLMs
681.63DeGF
LLaVA-1.5
Self-Correcting Decoding with Generative Feedback for Mitigating Hallucinations in Large Vision-Language Models
ICLR 2025 · POPE results table
Multimodal LLMs
780.53FLASH (Ours)
LLaVA-1.5 7B
Beyond Attention Imbalance: Mitigating Hallucinations via Spectral Surgery
ICML 2026 · Table 1
Multimodal LLMs
880.16ALEAHallu
LLaVA-1.5
Look Closer! An Adversarial Parametric Editing Framework for Hallucination Mitigation in VLMs
AAAI 2026 · Table 2
Multimodal LLMs
979.92DEGAP
LLaVA-1.5 7B
DEGAP: Dynamic Entropy-Guided Attention Perturbation for Contrastive Decoding in Large Vision-Language Models
ICML 2026 · Table 3
Multimodal LLMs
1079.4ONLY (Ours)
LLaVA-1.5
ONLY: One-Layer Intervention Sufficiently Mitigates Hallucinations in Large Vision-Language Models
ICCV 2025 · Table 1
Multimodal LLMs

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