List your company

Hallucination Evaluation on FINER-CompreCap 1.0 (whole)

80Multi-obj Acc (Paired)InternVL-3.5 + FINER-Tuning
0.9621.484262.52Mar 18, 2026

Evaluation results

MethodLinks
InternVL-3.5 + FINER-TuningSize=14B, tuning=FINER-Tuning2026.038078.971.230.1
InternVL-3.5Size=38B2026.0377.878.166.850.9
InternVL-3.5 + FINER-TuningSize=8B, tuning=FINER-Tuning2026.0377.178.964.134.2
Gemini-2.5-FlashSize=-2026.0375.777.377.858.2
InternVL-3.5Size=8B2026.037572.549.823.5
InternVL-3.5Size=14B2026.0374.568.14721.8
Qwen2.5-VL + FINER-TuningSize=7B, tuning=FINER-Tuning2026.0371.46738.334.8
Qwen2.5-VLSize=7B2026.0369.262.530.128.9
RLAIF-VSize=12B2026.0362.239.619.220.5
LLaVA-1.6 + FINER-TuningSize=7B, tuning=FINER-Tuning2026.0348.438.424.222.1
LLaVA-1.6Size=7B2026.0325.3137.615.3
RLHF-VSize=13B2026.0313.46.11.610.8
LLaVA-RLHFSize=13B2026.0311.421.16.9
OPA-DPOSize=13B2026.0310.932.26.9
LRV-V2Size=13B2026.036.16.85.64
Random GuessSize=-2026.034444