RewardBench 2 is a new benchmark that supplies challenging fresh human prompts for reward model evaluation, yielding lower average scores but higher correlation with downstream best-of-N sampling and RLHF training performance.
Qurating: Selecting high-quality data for training language models, 2024
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
UNVERDICTED 2representative citing papers
MindLoom synthesizes frontier-level reasoning data by decomposing solutions into thought mode chains, training a retrieval model for mode selection, composing new problems with distribution-aligned sampling, and applying rollout-based difficulty labeling for fine-tuning.
citing papers explorer
-
MindLoom: Composing Thought Modes for Frontier-Level Reasoning Data Synthesis
MindLoom synthesizes frontier-level reasoning data by decomposing solutions into thought mode chains, training a retrieval model for mode selection, composing new problems with distribution-aligned sampling, and applying rollout-based difficulty labeling for fine-tuning.