Tasl: Continual dialog state tracking via task skill localization and consolidation

Y Feng, X Chu, Y Xu, G Shi, B Liu, XM Wu - arxiv preprint arxiv …, 2024 - arxiv.org
A practical dialogue system requires the capacity for ongoing skill acquisition and
adaptability to new tasks while preserving prior knowledge. However, current methods for …

Recurrent Knowledge Identification and Fusion for Language Model Continual Learning

Y Feng, X Wang, Z Lu, S Fu, G Shi, Y Xu… - arxiv preprint arxiv …, 2025 - arxiv.org
Continual learning (CL) is crucial for deploying large language models (LLMs) in dynamic
real-world environments without costly retraining. While recent model ensemble and model …

Zero-shot Cross-domain Dialogue State Tracking via Context-aware Auto-prompting and Instruction-following Contrastive Decoding

X Dong, Y Feng, Z Lu, G Shi, XM Wu - Proceedings of the 2024 …, 2024 - aclanthology.org
Zero-shot cross-domain dialogue state tracking (DST) enables us to manage task-oriented
dialogues in new, unseen domains without the cost of collecting in-domain data. Previous …

GeoEdit: Geometric Knowledge Editing for Large Language Models

Y Feng, L Zhan, Z Lu, Y Xu, X Chu, Y Wang… - arxiv preprint arxiv …, 2025 - arxiv.org
Regular updates are essential for maintaining up-to-date knowledge in large language
models (LLMs). Consequently, various model editing methods have been developed to …

Diversity-grounded Channel Prototypical Learning for Out-of-Distribution Intent Detection

B Liu, L Zhan, Y Feng, Z Lu, C **e, L Xue… - arxiv preprint arxiv …, 2024 - arxiv.org
In the realm of task-oriented dialogue systems, a robust intent detection mechanism must
effectively handle malformed utterances encountered in real-world scenarios. This study …