List your company

Hallucination Robustness on HallusionBench

58.4ScorePerceptionDLM-Base
39.78444.61749.4554.283Jan 31, 2026Feb 22, 2026Mar 17, 2026Apr 9, 2026May 2, 2026May 25, 2026Jun 17, 2026

Evaluation results

MethodLinks
PerceptionDLM-BaseModel Size=8B2026.0658.4
PolarMemBackbone=QWEN2.5-VL-7B, Augmentation=PolarMem2026.0157.8
PolarMemBackbone=InternVL-3.5-8B, Augmentation=PolarMem2026.0156.3
ViLoMemBackbone=QWEN2.5-VL-32B, Augmentation=ViLoMem2026.0155.1
ViLoMemBackbone=QWEN2.5-VL-7B, Augmentation=ViLoMem2026.0154.9
Vanilla RAGBackbone=InternVL-3.5-8B, Augmentation=Vanilla RAG2026.0154.9
InternVL-3.5-8BBackbone=InternVL-3.5-8B, Augmentation=None2026.0154.7
ViLoMemBackbone=InternVL-3.5-8B, Augmentation=ViLoMem2026.0153.8
QWEN2.5-VL-32BBackbone=QWEN2.5-VL-32B, Augmentation=None2026.0153.5
PolarMemBackbone=QWEN2.5-VL-32B, Augmentation=PolarMem2026.0152.9
QWEN2.5-VL-7BBackbone=QWEN2.5-VL-7B, Augmentation=None2026.0152.7
Vanilla RAGBackbone=QWEN2.5-VL-32B, Augmentation=Vanilla RAG2026.0152.7
PolarMemBackbone=LLaVA-Next-Mistral-7B, Augmentation=PolarMem2026.0152
Qwen2.5-VLModel Size=7B2026.0651.9
ViLoMemBackbone=LLaVA-Next-Mistral-7B, Augmentation=ViLoMem2026.0151.5
LLaDA-VModel Size=8B2026.0650.9
Vanilla RAGBackbone=QWEN2.5-VL-7B, Augmentation=Vanilla RAG2026.0150.3
InternVL3Model Size=8B2026.0649.9
ViLoMemBackbone=DeepSeek-VL2, Augmentation=ViLoMem2026.0149.6
PolarMemBackbone=LLaVA-Next-Llama-3-8B, Augmentation=PolarMem2026.0149
PolarMemBackbone=InternVL-3.5-4B, Augmentation=PolarMem2026.0148
PolarMemBackbone=DeepSeek-VL2, Augmentation=PolarMem2026.0148
ViLoMemBackbone=LLaVA-Next-Llama-3-8B, Augmentation=ViLoMem2026.0146.9
DeepSeek-VL2Backbone=DeepSeek-VL2, Augmentation=None2026.0146.2
LLaVA-Next-Mistral-7BBackbone=LLaVA-Next-Mistral-7B, Augmentation=None2026.0145.4
ViLoMemBackbone=InternVL-3.5-4B, Augmentation=ViLoMem2026.0145.2
ViLoMemBackbone=DeepSeek-VL2-Small, Augmentation=ViLoMem2026.0145.1
InternVL-3.5-4BBackbone=InternVL-3.5-4B, Augmentation=None2026.0144.8
Vanilla RAGBackbone=LLaVA-Next-Mistral-7B, Augmentation=Vanilla RAG2026.0144.8
LLaVA-Next-Llama-3-8BBackbone=LLaVA-Next-Llama-3-8B, Augmentation=None2026.0144.7
SDAR-VLModel Size=8B2026.0644.4
Vanilla RAGBackbone=DeepSeek-VL2, Augmentation=Vanilla RAG2026.0144
Vanilla RAGBackbone=InternVL-3.5-4B, Augmentation=Vanilla RAG2026.0143.4
DeepSeek-VL2-SmallBackbone=DeepSeek-VL2-Small, Augmentation=None2026.0143.4
PolarMemBackbone=DeepSeek-VL2-Small, Augmentation=PolarMem2026.0142.8
Vanilla RAGBackbone=LLaVA-Next-Llama-3-8B, Augmentation=Vanilla RAG2026.0142.8
Vanilla RAGBackbone=DeepSeek-VL2-Small, Augmentation=Vanilla RAG2026.0140.5