Reward Bench 2 - a allenai Collection

allenai 's Collections

MolmoAct Data Mixture

IFBench

OLMo 2

olmOCR

OLMoE (January 2025)

PixMo

Tulu 3 Datasets

Molmo

OLMoE (November 2024)

Tulu V2.5 Suite

Paloma

SciRIFF

AI2 Safety Toolkit

Zebra Logic Bench

OLMo 2 Preview Post-trained Models

ACE

Reward Bench 2

updated Jun 3

Datasets, spaces, and models for Reward Bench 2 benchmark and paper!

allenai/reward-bench-2

Viewer • Updated Jun 4 • 1.87k • 5.7k • 26
Running

406

406

Reward Bench Leaderboard

📐

Display and analyze reward model evaluation results
allenai/reward-bench-2-results

Preview • Updated Aug 12 • 84 • 3
allenai/Llama-3.1-70B-Instruct-RM-RB2

Text Classification • Updated Jun 4 • 28 • 1
allenai/Llama-3.1-8B-Instruct-RM-RB2

Text Classification • Updated Jun 4 • 1.21k • 1
allenai/Llama-3.1-Tulu-3-8B-RL-RM-RB2

Text Classification • Updated Jun 4 • 44
allenai/Llama-3.1-Tulu-3-8B-DPO-RM-RB2

Text Classification • Updated Jun 4 • 37
allenai/Llama-3.1-Tulu-3-8B-SFT-RM-RB2

Text Classification • Updated Jun 4 • 34
allenai/Llama-3.1-8B-Base-RM-RB2

Text Classification • Updated Jun 4 • 24
allenai/Llama-3.1-Tulu-3-70B-SFT-RM-RB2

Text Classification • Updated Jun 4 • 28
RewardBench 2: Advancing Reward Model Evaluation

Paper • 2506.01937 • Published Jun 2 • 7