List your company

Value Alignment on cross-demographic (test)

48.6AccuracyQwen3-8B-DVMap
26.03231.89137.7543.609May 14, 2026

Evaluation results

MethodLinks
Qwen3-8B-DVMapLLM=Qwen3-8B-DVMap2026.0548.683.90.1321
Gpt-4oLLM=Gpt-4o2026.0548.583.80.1418
Qwen3-next-80B-a3BLLM=Qwen3-next-80B-a3B2026.0547.682.50.1449
Llama-3.3-70B-InstructLLM=Llama-3.3-70B-Instruct2026.0546.483.30.1504
Gpt-4o-miniLLM=Gpt-4o-mini2026.0546.382.40.1476
Qwen3-14BLLM=Qwen3-14B2026.0546.283.50.146
Gemini-2.5-flash-liteLLM=Gemini-2.5-flash-lite2026.0545.379.70.1538
DeepSeek-v3.2-expLLM=DeepSeek-v3.2-exp2026.0545.182.30.1342
Claude-3.7-sonnetLLM=Claude-3.7-sonnet2026.0526.946.40.1503