面向视觉-语言场景的迭代自过滤数据选择
Data Selection Through Iterative Self-Filtering for Vision-Language Settings
AI总结：
提出一种自举式方法，通过迭代训练CLIP模型并自选择改进的数据混合，无需额外数据或预训练模型即可提升下游性能。
摘要：
大量干净数据的可用性对于训练神经网络至关重要。然而，在大规模情况下，人工监督不切实际，导致数据集庞大且可能非常嘈杂。迄今为止，缓解这一障碍以生成高性能视觉-语言模型的尝试涉及启发式方法、精心策划的参考数据集以及使用预训练模型。在这里，我们提出了一种新颖的自举方法，其中CLIP模型在一个不断演变的、自选择的数据集上进行训练。这个不断演变的数据集构成了过滤后的、高概率干净样本以及来自整个分布的多样化样本的平衡。我们提出的自过滤方法在训练模型和选择随后改进的数据混合之间迭代。通过所提出方法过滤的视觉-语言数据集进行训练，无需额外数据或预训练模型即可提升下游性能。
英文摘要：
The availability of large amounts of clean data is paramount to training neural networks. However, at large scales, manual oversight is impractical, resulting in sizeable datasets that can be very noisy. Attempts to mitigate this obstacle to producing performant vision-language models have so far involved heuristics, curated reference datasets, and using pre-trained models. Here we propose a novel, bootstrapped method in which a CLIP model is trained on an evolving, self-selected dataset. This evolving dataset constitutes a balance of filtered, highly probable clean samples as well as diverse samples from the entire distribution. Our proposed Self-Filtering method iterates between training the model and selecting a subsequently improved data mixture. Training on vision-language datasets filtered by the proposed approach improves downstream performance without the need for additional data or pre-trained models.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
