Classical continual learning (CL) has primarily focused on enabling models to update and retain knowledge through parameter-centric mechanisms, e.g., training strategies, architectural designs, and weight adaptation. However, emerging paradigms are reshaping the scope of CL beyond this traditional model adaptation view. For instance, on-policy learning broadens the space of update mechanisms; test-time training extends CL from the training phase to inference; and external harness components such as memory, skill libraries, and interaction protocols extend the evolutionary boundaries of model capabilities far beyond the static parameter space. Collectively, these developments indicate a transition from parameter-centric learning toward system-level adaptation. To characterize this transition, we examine the evolution of continual learning through three dimensions: When, How, and Where learning occurs. The How dimension encompasses off-policy, on-policy, and beyond-gradient optimization mechanics. The When dimension captures evolution across pre-training, post-training, and inference-time stages. The Where dimension delineates updates occurring within internal parameters versus external structural constraints. Anchored by this tri-axial framework, we systematically survey representative methods, trace the ongoing transition of continual learning, and discuss the key challenges, broader implications, and future directions arising from this paradigm shift.</p>\n","updatedAt":"2026-08-07T09:11:36.460Z","author":{"_id":"640c352eb355c14a7b4049a4","avatarUrl":"/avatars/2afd35962d9f3804dac4ec1f6008ef9a.svg","fullname":"Dan Zhang","name":"zd21","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":10,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8401758074760437},"editors":["zd21"],"editorAvatarUrls":["/avatars/2afd35962d9f3804dac4ec1f6008ef9a.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2608.06216","authors":[{"_id":"6a75a0eee1228e04b32383c0","name":"Zhiyan Hou","hidden":false},{"_id":"6a75a0eee1228e04b32383c1","name":"Dan Zhang","hidden":false},{"_id":"6a75a0eee1228e04b32383c2","name":"Tao Feng","hidden":false},{"_id":"6a75a0eee1228e04b32383c3","name":"Liyuan Wang","hidden":false},{"_id":"6a75a0eee1228e04b32383c4","name":"Wei Li","hidden":false},{"_id":"6a75a0eee1228e04b32383c5","name":"Xiangzhao Hao","hidden":false},{"_id":"6a75a0eee1228e04b32383c6","name":"Hongyan An","hidden":false},{"_id":"6a75a0eee1228e04b32383c7","name":"Junfeng Fang","hidden":false},{"_id":"6a75a0eee1228e04b32383c8","name":"Haokai Ma","hidden":false},{"_id":"6a75a0eee1228e04b32383c9","name":"Zhaohui Xu","hidden":false},{"_id":"6a75a0eee1228e04b32383ca","name":"Haiyun Guo","hidden":false},{"_id":"6a75a0eee1228e04b32383cb","name":"Jinqiao Wang","hidden":false},{"_id":"6a75a0eee1228e04b32383cc","name":"Tat-Seng Chua","hidden":false}],"publishedAt":"2026-08-06T00:00:00.000Z","submittedOnDailyAt":"2026-08-07T00:00:00.000Z","title":"Continual Learning in Transition","submittedOnDailyBy":{"_id":"640c352eb355c14a7b4049a4","avatarUrl":"/avatars/2afd35962d9f3804dac4ec1f6008ef9a.svg","isPro":false,"fullname":"Dan Zhang","user":"zd21","type":"user","name":"zd21"},"summary":"Classical continual learning (CL) has primarily focused on enabling models to update and retain knowledge through parameter-centric mechanisms, e.g., training strategies, architectural designs, and weight adaptation. However, emerging paradigms are reshaping the scope of CL beyond this traditional model adaptation view. For instance, on-policy learning broadens the space of update mechanisms; test-time training extends CL from the training phase to inference; and external harness components such as memory, skill libraries, and interaction protocols extend the evolutionary boundaries of model capabilities far beyond the static parameter space. Collectively, these developments indicate a transition from parameter-centric learning toward system-level adaptation. To characterize this transition, we examine the evolution of continual learning through three dimensions: When, How, and Where learning occurs. The How dimension encompasses off-policy, on-policy, and beyond-gradient optimization mechanics. The When dimension captures evolution across pre-training, post-training, and inference-time stages. The Where dimension delineates updates occurring within internal parameters versus external structural constraints. Anchored by this tri-axial framework, we systematically survey representative methods, trace the ongoing transition of continual learning, and discuss the key challenges, broader implications, and future directions arising from this paradigm shift.","upvotes":2,"discussionId":"6a75a0efe1228e04b32383cd","organization":{"_id":"640a887796aae649741a586f","name":"CASIA","fullname":"Chinese Academic of Science Institute of Automation","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/1678411888885-6388984e8a5dbe2f3dc5afee.jpeg"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"640c352eb355c14a7b4049a4","avatarUrl":"/avatars/2afd35962d9f3804dac4ec1f6008ef9a.svg","isPro":false,"fullname":"Dan Zhang","user":"zd21","type":"user"},{"_id":"63ac5701c21e60a3e9b58aa7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63ac5701c21e60a3e9b58aa7/g6EX7diOpuA94R2ab-rZC.png","isPro":true,"fullname":"Dipankar Sarkar","user":"dipankarsarkar","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"640a887796aae649741a586f","name":"CASIA","fullname":"Chinese Academic of Science Institute of Automation","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/1678411888885-6388984e8a5dbe2f3dc5afee.jpeg"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2608/2608.06216.md","query":{}}">
Continual Learning in Transition
Abstract
Classical continual learning (CL) has primarily focused on enabling models to update and retain knowledge through parameter-centric mechanisms, e.g., training strategies, architectural designs, and weight adaptation. However, emerging paradigms are reshaping the scope of CL beyond this traditional model adaptation view. For instance, on-policy learning broadens the space of update mechanisms; test-time training extends CL from the training phase to inference; and external harness components such as memory, skill libraries, and interaction protocols extend the evolutionary boundaries of model capabilities far beyond the static parameter space. Collectively, these developments indicate a transition from parameter-centric learning toward system-level adaptation. To characterize this transition, we examine the evolution of continual learning through three dimensions: When, How, and Where learning occurs. The How dimension encompasses off-policy, on-policy, and beyond-gradient optimization mechanics. The When dimension captures evolution across pre-training, post-training, and inference-time stages. The Where dimension delineates updates occurring within internal parameters versus external structural constraints. Anchored by this tri-axial framework, we systematically survey representative methods, trace the ongoing transition of continual learning, and discuss the key challenges, broader implications, and future directions arising from this paradigm shift.
Community
Classical continual learning (CL) has primarily focused on enabling models to update and retain knowledge through parameter-centric mechanisms, e.g., training strategies, architectural designs, and weight adaptation. However, emerging paradigms are reshaping the scope of CL beyond this traditional model adaptation view. For instance, on-policy learning broadens the space of update mechanisms; test-time training extends CL from the training phase to inference; and external harness components such as memory, skill libraries, and interaction protocols extend the evolutionary boundaries of model capabilities far beyond the static parameter space. Collectively, these developments indicate a transition from parameter-centric learning toward system-level adaptation. To characterize this transition, we examine the evolution of continual learning through three dimensions: When, How, and Where learning occurs. The How dimension encompasses off-policy, on-policy, and beyond-gradient optimization mechanics. The When dimension captures evolution across pre-training, post-training, and inference-time stages. The Where dimension delineates updates occurring within internal parameters versus external structural constraints. Anchored by this tri-axial framework, we systematically survey representative methods, trace the ongoing transition of continual learning, and discuss the key challenges, broader implications, and future directions arising from this paradigm shift.
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2608.06216 in a model README.md to link it from this page.
Cite arxiv.org/abs/2608.06216 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2608.06216 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.