List your company

Reward Modeling Evaluation on Reward Bench Factuality 2

56.6Pairwise AccuracyDistribution-Calibrated Aggregation
-1.01613.94228.943.858Dec 2, 2025Dec 14, 2025Dec 26, 2025Jan 8, 2026Jan 20, 2026Feb 1, 2026Feb 14, 2026

Evaluation results

MethodLinks
Distribution-Calibrated Aggregationn=12, Judge LLM=gemini-2.5-flash2025.1256.6
Distribution-Calibrated Aggregationn=4, Judge LLM=gemini-2.5-flash2025.1253.6
USCn=12, Judge LLM=gemini-2.5-flash2025.1247.5
USCn=4, Judge LLM=gemini-2.5-flash2025.1247.2
GSCn=12, Judge LLM=gemini-2.5-flash2025.1246.1
SCn=4, Judge LLM=gemini-2.5-flash2025.1245
GSCn=4, Judge LLM=gemini-2.5-flash2025.1244.5
SCn=12, Judge LLM=gemini-2.5-flash2025.1242.4
CI-SCn=12, Judge LLM=gemini-2.5-flash2025.1241.1
Soft-SCn=4, Judge LLM=gemini-2.5-flash2025.1241
CI-SCn=4, Judge LLM=gemini-2.5-flash2025.1240.9
Soft-SCn=12, Judge LLM=gemini-2.5-flash2025.1239.9
Pointwise RatingModel Family=QWEN3, Model Size=8B2026.0234.1
Listwise Ranking*Model Family=GEMMA3, Model Size=4B2026.0233.2
FLIPModel Family=QWEN3, Model Size=4B2026.0231.3
FLIPModel Family=QWEN3, Model Size=8B2026.0231.1
Pointwise RatingModel Family=QWEN3, Model Size=4B2026.0230.5
FLIPModel Family=GEMMA3, Model Size=270M2026.0229.8
FLIPModel Family=QWEN3, Model Size=1.7B2026.0229.6
Pairwise Ranking*Model Family=MISTRAL, Model Size=7B2026.0228.5
FLIPModel Family=QWEN3, Model Size=0.6B2026.0228.4
FLIPModel Family=OLMO2, Model Size=7B2026.0227.8
Pairwise Ranking*Model Family=OLMO2, Model Size=7B2026.0227.3
FLIPModel Family=LLAMA3, Model Size=8B2026.0227.3
Listwise Ranking*Model Family=MISTRAL, Model Size=7B2026.0227.2
FLIPModel Family=MISTRAL, Model Size=7B2026.0227.1
Pairwise Ranking*Model Family=QWEN3, Model Size=4B2026.0226.5
FLIPModel Family=GEMMA3, Model Size=4B2026.0226.5
FLIPModel Family=LLAMA3, Model Size=3B2026.0226.4
FLIPModel Family=LLAMA3, Model Size=1B2026.0226.1
Pairwise Ranking*Model Family=QWEN3, Model Size=1.7B2026.0226.1
FLIPModel Family=OLMO2, Model Size=1B2026.0225.8
Pairwise Ranking*Model Family=GEMMA3, Model Size=4B2026.0225.6
Listwise Ranking*Model Family=LLAMA3, Model Size=8B2026.0225.1
FLIPModel Family=GEMMA3, Model Size=1B2026.0224.9
Pairwise Ranking*Model Family=QWEN3, Model Size=8B2026.0224.5
Pairwise Ranking*Model Family=LLAMA3, Model Size=8B2026.0222.2
Pairwise Ranking*Model Family=GEMMA3, Model Size=1B2026.0221.7
Listwise Ranking*Model Family=OLMO2, Model Size=7B2026.0221.3
Listwise Ranking*Model Family=GEMMA3, Model Size=1B2026.0220.9
Listwise Ranking*Model Family=LLAMA3, Model Size=3B2026.0220.5
Pointwise RatingModel Family=QWEN3, Model Size=1.7B2026.0220.2
Listwise Ranking*Model Family=QWEN3, Model Size=1.7B2026.0219.8
Listwise Ranking*Model Family=OLMO2, Model Size=1B2026.0218.3
Pairwise Ranking*Model Family=LLAMA3, Model Size=3B2026.0216.9
Pointwise RatingModel Family=GEMMA3, Model Size=270M2026.0216.8
Listwise Ranking*Model Family=QWEN3, Model Size=0.6B2026.0216.6
Listwise Ranking*Model Family=LLAMA3, Model Size=1B2026.0216
Pairwise Ranking*Model Family=QWEN3, Model Size=0.6B2026.0215.8
Pointwise RatingModel Family=LLAMA3, Model Size=1B2026.0215.6
Pointwise RatingModel Family=LLAMA3, Model Size=8B2026.0215.6
Pointwise RatingModel Family=OLMO2, Model Size=1B2026.0215.5
Listwise Ranking*Model Family=GEMMA3, Model Size=270M2026.0215.3
Pointwise RatingModel Family=GEMMA3, Model Size=4B2026.0214.6
Pointwise RatingModel Family=QWEN3, Model Size=0.6B2026.0213.5
Pointwise RatingModel Family=LLAMA3, Model Size=3B2026.0213.3
Pairwise Ranking*Model Family=LLAMA3, Model Size=1B2026.0212
Pairwise Ranking*Model Family=GEMMA3, Model Size=270M2026.0211.3
Pointwise RatingModel Family=GEMMA3, Model Size=1B2026.0210.2
Listwise Ranking*Model Family=QWEN3, Model Size=8B2026.029.8
Listwise Ranking*Model Family=QWEN3, Model Size=4B2026.028.9
Pointwise RatingModel Family=OLMO2, Model Size=7B2026.027.8
Pointwise RatingModel Family=MISTRAL, Model Size=7B2026.025.5
Pairwise Ranking*Model Family=OLMO2, Model Size=1B2026.021.2