用合成数据训练大模型，不会加剧其推理盲区
Does Training on Synthetic Data Make Models Less Robust?
- 用Llama-2模型生成合成数据进行微调
- 合成数据未加重对启发式策略的依赖
- 适合关注模型训练数据安全性的研究者
越来越多的大语言模型（LLMs）采用合成数据进行训练，这些数据常由同类或相似的LLM生成。这引发了担忧：合成数据是否会强化模型已有的推理偏差？本文以Llama-2-7B-hf模型为实验对象，在自然语言推理（NLI）任务上开展模拟实验。使用MultiNLI作为通用任务，HANS作为专门检测特定启发式策略的盲点评估集。结果表明，尽管合成数据微调未减少启发式策略的使用，但也没有像预期那样加剧其影响，说明合成数据并未显著放大模型的推理盲区。
原文摘要 · Abstract (English)
An increasingly common practice is to train large language models (LLMs) using synthetic data. Often this synthetic data is produced by the same or similar LLMs as those it is being used to train. This raises the question of whether the synthetic data might in fact exacerbate certain "blindspots" by reinforcing heuristics that the LLM already encodes. In this paper, we conduct simulated experiments on the natural language inference (NLI) task with Llama-2-7B-hf models. We use MultiNLI as the general task and HANS, a targeted evaluation set designed to measure the presence of specific heuristic strategies for NLI, as our "blindspot" task. Our goal is to determine whether performance disparities between the general and blind spot tasks emerge. Our results indicate that synthetic data does not reinforce blindspots in the way we expected. Specifically, we see that, while fine-tuning with synthetic data doesn't necessarily reduce the use of the heuristic, it also does not make it worse as we hypothesized.
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。