Continual learning (CL) seeks to adapt pretrained models while preserving prior capabilities. Yet the tasks supplied for preservation cover only part of what a model knows, while forgetting can extend beyond them. Recent work shows that language models can supplement this limited evidence by generating useful replay without access to their original training data. This raises a question about self-preservation: can a model help choose preservation targets that protect its own capabilities more broadly? To study this question, we organize task descriptions and their examples into Task Banks, whose entries can be selected and combined to guide replay generation by a frozen model. Continual instruction tuning with Qwen models of varying sizes yields three findings. First, the new task and learning rate shape which prior capabilities degrade, while tracking previously correct answers reveals losses that average accuracy can conceal. Second, generated replay reduces forgetting on tasks outside the bank. Combining partial banks strengthens this protection, improving mean retention of previously correct answers over either bank alone across the tested sizes. Third, the model can also help choose replay targets: selection informed by the new task improves mean held-out retention over random selection at 4B, with similar new-task gains, although smaller models show no corresponding advantage. Under a fixed example budget, however, replacing new-task examples with replay improves preservation at the cost of smaller new-task gains. Together, these findings show that models can support broader self-preservation by generating replay data and, in some settings, choosing which tasks to replay.
LyapLock: Bounded Knowledge Preservation in Sequential Large Language Model Editing
Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所);School of Cyber Security, University of Chinese Academy of Sciences(中国科学院大学网络空间安全学院)
Target-Aware Calibration Data Selection for Preserving Uncertainty in Quantized Language Models
The Hong Kong University of Science and Technology(香港科技大学);The Hong Kong University of Science and Technology (Guangzhou)(香港科技大学(广州));City University of Hong Kong(香港城市大学);Shanghai Institute of Optics and Fine Mechanics(上海光学精密机械研究所);University of Oxford(牛津大学);Yale University(耶鲁大学)
WorldBench: Culturally Grounded Benchmark for Multilingual Agents
University of Edinburgh(爱丁堡大学);School of Informatics, University of Edinburgh(爱丁堡大学信息学院);School of Artificial Intelligence, Shanghai Jiao Tong University(上海交通大学人工智能学院)
Mind the Style Gap: Meta-Evaluation of Style and Attribute Transfer Metrics
Department of Computer Science, Aarhus University, Denmark(计算机科学系,奥胡斯大学);Department of Computer Science, University of Copenhagen, Denmark(计算机科学系,哥本哈根大学)
LIMBA: An Open-Source Framework for the Preservation and Valorization of Low-Resource Languages using Generative Models
University of Cagliari(卡利亚里大学);VisioScientiae S.r.l.(VisioScientiae有限公司);The Cloud Alchemist S.r.l.(Cloud Alchemist有限公司);University for Foreigners of Siena(锡耶纳外国人大学)
Prompt-OT: An Optimal Transport Regularization Paradigm for Knowledge Preservation in Vision-Language Model Adaptation
Morgan Stanley(摩根士丹利);Clemson University(克莱姆森大学);Arizona State University(亚利桑那州立大学);Washington University in St. Louis(圣路易斯华盛顿大学);University of Arizona(亚利桑那大学);University of Notre Dame(圣母大学)
Style Transfer in Text: Exploration and Evaluation
Institute of Computer Science and Technology, Peking University(北京大学计算机科学技术研究所);Information Science Institute, University of Southern California(南加州大学信息科学研究所);Beijing Institute of Big Data Research(北京大数据研究院)
Seamless Language Expansion: Enhancing Multilingual Mastery in Self-Supervised Models
Department of Systems Engineering and Engineering Management(系统工程与工程管理系);The Chinese University of Hong Kong(香港中文大学);Centre for Perceptual and Interactive Intelligence (CPII) Limited(感知与交互智能中心(CPII)有限公司)
本文提出将 LoRA 集成到已有自监督模型并结合数据组合与重新聚类保留策略,使 mHuBERT 高效扩展至普通话,在语音重合成任务中显著提升 MOS 并降低 WER,同时保持原有语言能力不退化。