List your company

Hallucination Evaluation on FINER-DOCCI 3K MCQs per setting 1.0

65.9Multi-Object Paired AccuracyInternVL-3.5 + FINER-Tuning
1.52418.23734.9551.663Mar 18, 2026

Evaluation results

MethodLinks
InternVL-3.5 + FINER-TuningSize=14B, tuning=FINER-Tuning2026.0365.9655723
Gemini-2.5-FlashSize=-2026.0364.464.556.749.6
InternVL-3.5 + FINER-TuningSize=8B, tuning=FINER-Tuning2026.0362.660.152.723.7
InternVL-3.5Size=38B2026.0362.364.854.236.6
InternVL-3.5Size=14B2026.0358.655.941.415.6
InternVL-3.5Size=8B2026.0358.154.341.816.8
Qwen2.5-VL + FINER-TuningSize=7B, tuning=FINER-Tuning2026.0349.852.243.428
Qwen2.5-VLSize=7B2026.0348.747.536.723.4
RLAIF-VSize=12B2026.0346.531.732.419.4
LLaVA-1.6 + FINER-TuningSize=7B, tuning=FINER-Tuning2026.0326.429.424.718.5
RLHF-VSize=13B2026.0313.27.28.17
LLaVA-1.6Size=7B2026.0310.112.38.213.3
OPA-DPOSize=13B2026.038.15.58.38
LLaVA-RLHFSize=13B2026.037.335.15.3
LRV-V2Size=13B2026.036.35.46.15.2
Random GuessSize=-2026.034444