AI 中文总结
研究在特定规则下从单张无约束面部进行多任务情感识别问题，基于共享情感潜变量提出强度-奇偶规则，用参数隔离解决成员相关性及多样性问题，提升了系统整体验证分数。
AI 中文摘要
第11届ABAW挑战赛多任务赛道的领先成果依赖大量集成，但很少明确添加哪些成员。本文针对从单张无约束面部进行联合效价-唤醒估计、8类表情识别和12类动作单元检测展开研究，在部分长尾标签及禁止在Aff-Wild2上预训练的规则下，基于共享情感潜变量提出强度-奇偶规则，通过参数隔离解决问题，最终系统提升了整体验证分数。构建在共享情感潜变量之上，该潜变量将两个情感监督主干中的缺失标签边缘化，我们提出了一种强度-奇偶规则：只有当一个添加的成员与当前成员不相关且在个体准确性上与它们相近时，它才会降低集成误差。该规则揭示了一个具体障碍，因为在单个主干上重新播种，甚至不同的微调课程都会重新收敛到0.98的预测相关性，且不会增加多样性。参数隔离消除了这一障碍：将每个适应限制在共享主干的不相交低秩子空间中，产生的专家在保持0.91的不相关性的同时仍保持相近水平，其中最强的是一个适应AffectNet的专家。由此产生的系统将整体验证分数提高到1.6949，而组织者的ConvNeXt-with-MixAugment基线为0.45；通过每个AU校准并将共享潜变量头部的效价-唤醒副产品作为另一个相近水平进行合并，最强配置达到1.7259。
英文摘要
Leading entries on the multi-task track of the 11th ABAW challenge rely on heavy ensembling, yet which member is worth adding to an already strong ensemble is rarely made explicit. We study this question for joint valence-arousal estimation, 8-way expression recognition, and 12-way action-unit detection from a single unconstrained face, under partial, long-tailed labels and a rule that forbids pretraining on Aff-Wild2. Building on a shared affect-latent that marginalizes the missing labels across two affect-supervised backbones, we propose a strength-parity rule: an added member lowers the ensemble error only when it is both decorrelated from the current members and a near-peer of them in individual accuracy. The rule exposes a concrete obstacle, as on a single backbone re-seeding and even distinct fine-tuning curricula re-converge to a prediction correlation of 0.98 and add no diversity. Parameter-isolation removes it: confining each adaptation to a disjoint low-rank subspace of a shared backbone yields experts that stay decorrelated at 0.91 while remaining near-peers, the strongest of them an AffectNet-adapted expert. The resulting system raises the overall validation score to 1.6949, against the organizers ConvNeXt-with-MixAugment baseline of 0.45; with per-AU calibration and by pooling the shared-latent heads valence-arousal byproduct as a further near-peer, the strongest configuration reaches 1.7259. Source code are available at https://github.com/cprl-team/MTL-ABAW-11th.
