Papers with Continual Learning
Overcoming Catastrophic Forgetting by Exemplar Selection in Task-oriented Dialogue System (2024.findings-acl)
Copied to clipboard
| Challenge: | Experimental results show that HESIT effectively alleviates catastrophic forgetting by exemplar selection, and achieves state-of-the-art performance on the largest CL benchmark of ToDs in terms of all metrics. |
| Approach: | They propose a method to overcome catastrophic forgetting in task-oriented dialogue systems by tracing their hyper-gradients and a retraining strategy that uses influential exemplars for periodic retrains. |
| Outcome: | The proposed method achieves state-of-the-art on the largest CL benchmark of ToDs in terms of all metrics. |
Coordinated Replay Sample Selection for Continual Federated Learning (2023.emnlp-industry)
Copied to clipboard
Jack Good, Jimit Majmudar, Christophe Dupuy, Jixuan Wang, Charith Peris, Clement Chung, Richard Zemel, Rahul Gupta
| Challenge: | Continual Federated Learning (CFL) combines decentralized learning with continuous learning . ubiquity of personal devices with a network connection offers rich source of data for learning problems . |
| Approach: | They propose to combine decentralized learning with a continuous learning approach . they propose to coordinate gradient-based replay sample selection across clients . |
| Outcome: | The proposed method shows gains early in the low replay size regime, when the budget for storing past data is small. |
Exclusive Supermask Subnetwork Training for Continual Learning (2023.findings-acl)
Copied to clipboard
| Challenge: | Continual Learning (CL) methods focus on accumulating knowledge over time while preventing catastrophic forgetting. |
| Approach: | They propose a CL method that finds a supermask for each new task that keeps or removes each weight to produce a subnetwork. |
| Outcome: | The proposed method outperforms strong previous methods on NLP and Vision domains while preventing forgetting. |
EvoWiki: Evaluating LLMs on Evolving Knowledge (2025.acl-long)
Copied to clipboard
Wei Tang, Yixin Cao, Yang Deng, Jiahao Ying, Bo Wang, Yizhe Yang, Yuyue Zhao, Qi Zhang, Xuanjing Huang, Yu-Gang Jiang, Yong Liao
| Challenge: | Existing knowledge evolution benchmarks are static and fail to capture the evolving nature of LLMs and knowledge. |
| Approach: | They propose an evolving dataset that categorizes information into stable, evolved, and uncharted states. |
| Outcome: | The proposed dataset is auto-updatable and enables evaluation of continuously changing knowledge and newly released LLMs. |
Exploring Continual Learning for Code Generation Models (2023.acl-short)
Copied to clipboard
Prateek Yadav, Qing Sun, Hantian Ding, Xiaopeng Li, Dejiao Zhang, Ming Tan, Parminder Bhatia, Xiaofei Ma, Ramesh Nallapati, Murali Krishna Ramanathan, Mohit Bansal, Bing Xiang
| Challenge: | Large-scale code generation models such as Copilot and CodeT5 are expensive to train and re-train. |
| Approach: | They propose a benchmark for Continual Learning (CL) that covers a wide range of tasks with different input and output programming languages. |
| Outcome: | The proposed method improves on Prompt Pooling with Teacher Forcing, which suffers catastrophic forgetting due to stark distribution shifts in coding tasks. |
ConTinTin: Continual Learning from Task Instructions (2022.acl-long)
Copied to clipboard
| Challenge: | a new learning paradigm is proposed for NLP, which seeks supervision for solving a target task. |
| Approach: | They propose a new learning paradigm that uses textual instructions to learn new tasks . the main goal of machine learning algorithms lies in seeking supervision for solving a target task. |
| Outcome: | The proposed learning paradigm is based on a stream of more than 60 tasks . it makes full use of task instructions to improve forward-transfer and backward-transference . |
Distilling Causal Effect from Miscellaneous Other-Class for Continual Named Entity Recognition (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for Named Entity Recognition (NER) are not able to learn Other-Class in the same way as new entity types. |
| Approach: | They propose a unified causal framework to retrieve causality from new entity types and Other-Class. |
| Outcome: | The proposed method outperforms the state-of-the-art method on three benchmark datasets. |
COPR: Continual Human Preference Learning via Optimal Policy Regularization (2025.findings-acl)
Copied to clipboard
Han Zhang, Lin Gui, Yu Lei, Yuanzhao Zhai, Yehong Zhang, Zhuo Zhang, Yulan He, Hui Wang, Yue Yu, Kam-Fai Wong, Bin Liang, Ruifeng Xu
| Challenge: | Reinforcement Learning from Human Feedback (RLHF) is effective for aligning Large Language Models with human preferences, but its complex process limits its ability to continually learn human feedback. |
| Approach: | They propose a non-RL offline method to convert historical optimal policies into optimization constraints when continually learning new preferences. |
| Outcome: | The proposed method outperforms strong CL baselines in terms of reward-based evaluations and human assessment. |
Spectral Disentanglement: Rank-Aware Task Adaptation for Rehearsal-free Continual Learning in LLMs (2026.acl-long)
Copied to clipboard
| Challenge: | Continual Learning (CL) for Large Language Models faces a fundamental Stability-Plasticity Dilemma . Rank-Blindness enforces a single rank constraint across diverse tasks, leading to catastrophic forgetting of earlier tasks and underfitting on complex new ones. |
| Approach: | They propose a rank-spectrum-based rehearsal-free framework that explicitly disentangles knowledge into two orthogonal subspaces. |
| Outcome: | The proposed framework achieves a superior stability-plasticity balance compared to single-rank baselines. |
Federated Continual Learning for Text Classification via Selective Inter-client Transfer (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Continual Learning (CL) is a privacy-preserving machine learning technique that enables collaborative training of ML models by sharing model parameters across distributed clients. |
| Approach: | They propose a framework which selectively combines model parameters of foreign clients to maximize knowledge transfer while preserving privacy. |
| Outcome: | The proposed framework improves the performance of a text classification task using five datasets from diverse domains while preserving privacy. |
Robust Uncertainty Quantification for Self-Evolving Large Language Models via Continual Domain Pretraining (2026.findings-acl)
Copied to clipboard
| Challenge: | Conformal prediction (CP) has shown promise in offering correctness guarantees for LLMs, but it faces major challenges in continual domain pretraining (CDP). |
| Approach: | They propose an adaptive rejection and non-exchangeable CP framework that allows the LLM to selectively abstain from answering when its confidence or competence shifts significantly. |
| Outcome: | Experiments show that the proposed framework improves performance under continuous domain pretraining scenarios. |
F-MALLOC: Feed-forward Memory Allocation for Continual Learning in Neural Machine Translation (2024.naacl-long)
Copied to clipboard
| Challenge: | Existing approaches to address Catastrophic Forgetting (CF) have been developed to avoid forgetting and maintain system extensibility. |
| Approach: | They propose a method to reduce Catastrophic Forgetting (CF) by decomposing feed-forward layers into discrete memory cells and ensuring robust extendability. |
| Outcome: | The proposed method achieves higher BLEU scores and almost zero forgetting while maintaining robust extendability. |
Fine-tuned Language Models are Continual Learners (2022.emnlp-main)
Copied to clipboard
| Challenge: | Recent work on large language models relies on intuition that most tasks can be described via natural language instructions. |
| Approach: | They propose that a model should be able to keep extending its knowledge without forgetting previous skills. |
| Outcome: | The proposed model can learn 8 new diverse language generation tasks while maintaining good performance on previous tasks, spanning in total of 70 datasets. |
A Generative Adaptive Replay Continual Learning Model for Temporal Knowledge Graph Reasoning (2025.acl-long)
Copied to clipboard
| Challenge: | Existing Continual Learning (CL)-based Temporal Knowledge Graph Reasoning methods are incomplete and reorganize historical facts without preserving historical knowledge. |
| Approach: | They propose a method which generates and adaptively replays historical entity distributions from the whole historical context. |
| Outcome: | The proposed method outperforms baselines in reasoning and mitigating forgetting. |
Continual Gradient Low-Rank Projection Fine-Tuning for LLMs (2025.acl-long)
Copied to clipboard
| Challenge: | Low-Rank Adaptation (LoRA) offers efficiency but constrains the model’s ability to learn new tasks and transfer knowledge due to its low-rank nature and reliance on explicit parameter constraints. |
| Approach: | They propose a training strategy that synergistically combines full and low-rank parameters and jointly updating within a unified low-ranked gradient subspace. |
| Outcome: | Extensive experiments on continual learning benchmarks show that GORP improves performance compared to state-of-the-art approaches. |