See-Saw Modality Balance: See Gradient, and Sew Impaired Vision-Language Balance to Mitigate Dominant Modality Bias

June Hyoung Kwon, MiHyeon Kim, Eunju Lee, Juhwan Choi, YoungBin Kim

See-Saw Modality Balance: See Gradient, and Sew Impaired Vision-Language Balance to Mitigate Dominant Modality Bias: 5 upvotes on Hugging Face Daily Papers, #38 of 48 papers on 2025-03-21. Day-by-day upvote history.

Vision-language (VL) models have demonstrated strong performance across various tasks. However, these models often rely on a specific modality for predictions, leading to "dominant modality bias.'' This bias significantly hurts performance, especially when one modality is impaired. In this study, we analyze model behavior under dominant modality bias and theoretically show that unaligned gradients or differences in gradient magnitudes prevent balanced convergence of the loss. Based on these findings, we propose a novel framework, BalGrad to mitigate dominant modality bias. Our approach includes inter-modality gradient reweighting, adjusting the gradient of KL divergence based on each modality's contribution, and inter-task gradient projection to align task directions in a non-conflicting manner. Experiments on UPMC Food-101, Hateful Memes, and MM-IMDb datasets confirm that BalGrad effectively alleviates over-reliance on specific modalities when making predictions.

Paper page on Hugging Face · arXiv

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.