List your company

Counterfactual Prediction on CausalVLBench

87.7PrecisionBRIDGEVLM-7B
62.01268.68175.3582.019Jun 10, 2026

Evaluation results

MethodLinks
BRIDGEVLM-7BModel Size=7B2026.0687.79083.798.5
QWEN2.5-VL-32BModel Size=32B2026.0687.490.886.798.4
GEMINI-2.0-FLASH (CM)Category=Commercial2026.0686.590.788.397.4
LLAVA-ONEVISION-7BModel Size=7B2026.0683.588.58596.9
DEEPSEEK-VL2-S-16BModel Size=16B2026.0677.95751.441.6
PHI-4-MMI-7B (CAUSAL-TRACE)Model Size=7B, Setting=Causal-trace supervision2026.0676.584.883.194.9
PHI-4-MMI-7B (ANSWER)Model Size=7B, Setting=Answer-only fine-tuning2026.0676.284.280.995.6
PHI-4-MMI-7B (NO-FT, CAUSAL-PROMPT)Model Size=7B, Setting=No-FT, Causal-Prompt2026.066376.873.394