GFlowNets生成多样性解的能力在跨任务迁移中表现不佳。
Do GFlowNets Transfer? Case Study on the Game of 24/42
- 用流网络优化解的生成过程，提升多样性
- 在24点和42点游戏上测试，准确率与多样性均下降
- 适合研究生成模型迁移能力的学者参考
生成多样化解是类人推理的关键，但自回归语言模型通常只聚焦单一正确答案，限制了创造力。GFlowNets将解生成建模为流网络，有望提升多样性。本案例研究通过微调小型和中型大语言模型在24点游戏数据集上，测试其在42点数据集上的表现。结果表明，GFlowNets难以保持解的多样性和准确性，暴露出其跨任务泛化能力的显著局限，凸显未来需加强迁移学习能力的研究。
原文摘要 · Abstract (English)
Generating diverse solutions is key to human-like reasoning, yet autoregressive language models focus on single accurate responses, limiting creativity. GFlowNets optimize solution generation as a flow network, promising greater diversity. Our case study shows their limited zero-shot transferability by fine-tuning small and medium-sized large language models on the Game of 24 and testing them on the Game of 42 datasets. Results revealed that GFlowNets struggle to maintain solution diversity and accuracy, highlighting key limitations in their cross-task generalization and the need for future research in improved transfer learning capabilities.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。