List your company

Review Feedback Generation on RMR-75K (val)

76.2Pairwise Win RateRBTACT
21.70435.8525064.148Mar 10, 2026

Evaluation results

MethodLinks
RBTACTOpponent=LimGen2026.0376.2
GPT-5-chatOpponent=LimGen2026.0371.4
RBTACTOpponent=MARG2026.0368.6
DeepSeek-V3.2Opponent=LimGen2026.0368.6
RBTACTOpponent=Qwen-32026.0366.7
Llama-3.1-70BOpponent=LimGen2026.0366.7
RBTACTOpponent=DeepReviewer2026.0365.7
RBTACTOpponent=RBTACT-SFT2026.0365.7
DeepReviewer-14bOpponent=LimGen2026.0364.8
RBTACTOpponent=DeepSeek2026.0363.8
GPT-5-chatOpponent=MARG2026.0362.9
RBTACTOpponent=Llama-3.12026.0361.9
MARGOpponent=LimGen2026.0361.9
GPT-5-chatOpponent=DeepReviewer2026.0360
GPT-5-chatOpponent=Qwen-32026.0360
DeepSeek-V3.2Opponent=MARG2026.0359
DeepSeek-V3.2Opponent=Qwen-32026.0358.1
Qwen-3-32BOpponent=LimGen2026.0358.1
RBTACTOpponent=GPT-52026.0357.1
GPT-5-chatOpponent=Llama-3.12026.0357.1
DeepSeek-V3.2Opponent=DeepReviewer2026.0357.1
Llama-3.1-70BOpponent=Qwen-32026.0356.2
RBTACT-SFTOpponent=LimGen2026.0356.2
GPT-5-chatOpponent=DeepSeek2026.0355.2
Llama-3.1-70BOpponent=MARG2026.0355.2
GPT-5-chatOpponent=RBTACT-SFT2026.0354.3
DeepSeek-V3.2Opponent=Llama-3.12026.0354.3
DeepSeek-V3.2Opponent=RBTACT-SFT2026.0353.3
DeepReviewer-14bOpponent=Qwen-32026.0353.3
Llama-3.1-70BOpponent=DeepReviewer2026.0352.4
Qwen-3-32BOpponent=RBTACT-SFT2026.0352.4
Llama-3.1-70BOpponent=RBTACT-SFT2026.0351.4
DeepReviewer-14bOpponent=MARG2026.0351.4
DeepReviewer-14bOpponent=RBTACT-SFT2026.0350.5
MARGOpponent=Qwen-32026.0350.5
RBTACT-SFTOpponent=MARG2026.0350.5
MARGOpponent=RBTACT-SFT2026.0349.5
Qwen-3-32BOpponent=MARG2026.0349.5
MARGOpponent=DeepReviewer2026.0348.6
RBTACT-SFTOpponent=Llama-3.12026.0348.6
DeepReviewer-14bOpponent=Llama-3.12026.0347.6
RBTACT-SFTOpponent=DeepReviewer2026.0347.6
RBTACT-SFTOpponent=Qwen-32026.0347.6
Qwen-3-32BOpponent=DeepReviewer2026.0346.7
RBTACT-SFTOpponent=DeepSeek2026.0346.7
Llama-3.1-70BOpponent=DeepSeek2026.0345.7
RBTACT-SFTOpponent=GPT-52026.0345.7
DeepSeek-V3.2Opponent=GPT-52026.0344.8
MARGOpponent=Llama-3.12026.0344.8
Qwen-3-32BOpponent=Llama-3.12026.0343.8
LimGenOpponent=RBTACT-SFT2026.0343.8
GPT-5-chatOpponent=RBTACT2026.0342.9
Llama-3.1-70BOpponent=GPT-52026.0342.9
DeepReviewer-14bOpponent=DeepSeek2026.0342.9
Qwen-3-32BOpponent=DeepSeek2026.0341.9
LimGenOpponent=Qwen-32026.0341.9
MARGOpponent=DeepSeek2026.0341
DeepReviewer-14bOpponent=GPT-52026.0340
Qwen-3-32BOpponent=GPT-52026.0340
Llama-3.1-70BOpponent=RBTACT2026.0338.1
LimGenOpponent=MARG2026.0338.1
MARGOpponent=GPT-52026.0337.1
DeepSeek-V3.2Opponent=RBTACT2026.0336.2
LimGenOpponent=DeepReviewer2026.0335.2
DeepReviewer-14bOpponent=RBTACT2026.0334.3
RBTACT-SFTOpponent=RBTACT2026.0334.3
Qwen-3-32BOpponent=RBTACT2026.0333.3
LimGenOpponent=Llama-3.12026.0333.3
MARGOpponent=RBTACT2026.0331.4
LimGenOpponent=DeepSeek2026.0331.4
LimGenOpponent=GPT-52026.0328.6
LimGenOpponent=RBTACT2026.0323.8