Existing class-incremental learning methods struggle in multi-label scenarios (MLCIL) due to the inherent contradiction of learning objectives arising from co-occurring and incomplete labels. We argue that the core obstacle is the model's ambiguous boundary between known and unknown knowledge, which undermines historic...
Ao-Ting Zhang, Dong-Bao Yang, Chang Liu et al.· 0 citations
Graphical User Interface (GUI) agents enable the fulfillment of complex user instructions through multi-turn interactions with software environments, requiring step-wise reasoning and long-horizon memory to guide actions and retain task-relevant information, respectively. Recent on-policy self-distillation (OPSD) metho...
Yan Zhang, Dai-Qing Wu, Hua-Wen Shen et al.· 0 citations
The MaP is proposed, a unified framework that seamlessly harmonizes divergent GUI navigation tasks and effectively mitigates gradient conflicts and significantly outperforms the direct mixture training, establishing a robust paradigm for multi-task GUI navigation.
Yan Zhang, Pei Fu, Dai-Qing Wu et al.· 0 citations
This work introduces Emotion Statement Judgement (ESJ), a statement-verification formulation that preserves the expressiveness of the input space while constraining outputs to discriminative judgements, and builds EmObserver, an emotion-oriented MLLM optimized on ESJ through an elaborate multi-stage recipe.
Daiqing Wu, Dong-Bao Yang, Jiashu Yao et al.· arXiv.org· 1 citation
To minimize knowledge interference during fusion, this work presents a gradient-based orthogonal refreshing strategy that projects gradient updates of new domains onto the orthogonal complement of the fused historical subspace, supporting continual adaptation without forgetting.
Aoting Zhang, Dongbao Yang, Chang Liu et al.· arXiv.org· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
