GPQA-Diamond

The 'GPQA-Diamond' dataset/benchmark contains evaluation criteria for assessing the performance of language models in various tasks, specifically focusing on the quality of generated outputs in areas like science, medicine, instruction following, and creative writing.

Papers using GPQA-Diamond (15)

Nemotron-CrossThink: Scaling Self-Learning beyond Math Reasoning2025 · 24 citesRing-lite: Scalable Reasoning via C3PO-Stabilized Reinforcement Learning for LLMs2025DianJin-R1: Evaluating and Enhancing Financial Reasoning in Large Language Models2025 · 1 citesProcess Reward Models That Think2025 · 1 citesOpen-Reasoner-Zero: An Open Source Approach to Scaling Up Reinforcement Learning on the Base Model2025 · 1 citesRubric-Grounded RL: Structured Judge Rewards for Generalizable Reasoning2026TRACE: Distilling Where It Matters via Token-Routed Self On-Policy Alignment2026AMARIS: A Memory-Augmented Rubric Improvement System for Rubric-Based Reinforcement Learning2026QuantLRM: Quantization of Large Reasoning Models via Fine-Tuning Signals2026Transformation-Augmented GRPO for Enhancing Exploration in Reasoning of Large Language Models2026TraPO: A Semi-Supervised Reinforcement Learning Framework for Boosting LLM Reasoning2025MARSHAL: Incentivizing Multi-Agent Reasoning via Self-Play with Strategic LLMs2025Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domains2025Kimi K2: Open Agentic Intelligence2025Prior Prompt Engineering for Reinforcement Fine-Tuning2025