— INFLogic-Qwen2.5-32B-RL-Preview enhances the reasoning capabilities of DeepSeek-R1-Distill-Qwen-32B through fine-tuning on our proprietary logical reasoning dataset using reinforcement learning with verifiable rewards (RLVR). — As of March 27th, 2025, this model achieves state-of-the-art performance among open-source LLMs on ZebraLogicBench, demonstrating enhanced logical reasoning abilities. We report pass@1 scores using vLLM 0.5.3 (temperature=0.6, topp=0.95). For MATH-500 and GPQA, we used Open R1’s evaluation scripts. Other models’ results come from their original reports.* We thank Chao Qu, Haozhe Wang, Jiaran Hao, and Liuyihan Song for their valuable discussions and support.
Модальности:
Генерация текста
Области применения:
Диалог / чат
Задача: Генерация текста
Автор: infly
Теги: qwen2, conversational, en, text-generation-inference, endpoints_compatible
Лайков: 4 | Загрузок: 10
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
