HybridAL treats the training strategy in active learning (retrain from scratch each round, or fine-tune from the previous checkpoint) as a decision variable rather than a fixed implementation detail.</p>\n<p>The structure we found: retraining is most useful in early rounds, when each new batch can substantially reshape the labeled distribution, while fine-tuning becomes safe once the model trajectory stabilizes. HybridAL monitors an online stabilization signal and switches once, after sustained stabilization. We study two signals, spectral exponent change (Δα, weight-based) and accuracy change (ΔAcc, validation-based), which sit at different points on the time–calibration trade-off.</p>\n<p>Across three encoder backbones and six text classification tasks, five seeds each: endpoint macro-F1 non-inferior to both retraining and fine-tuning at a 0.010 margin, up to 49% of retraining time saved, and a substantial fraction of retraining's calibration advantage recovered in NLL. Against schedules that switch at a pre-committed round, HybridAL reaches lower NLL at moderate additional cost.</p>\n<p>Accepted to EMNLP 2026 Main Conference.</p>\n","updatedAt":"2026-09-10T09:20:17.939Z","author":{"_id":"6995cfd70f9c4f0aebdebc51","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/a9OVt11nTWAXafft1XW0L.png","fullname":"Nagham Omar","name":"naghamo","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.9071957468986511},"editors":["naghamo"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/a9OVt11nTWAXafft1XW0L.png"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2609.06806","authors":[{"_id":"6aa1d6c5a2aeb74440b1dceb","user":{"_id":"6995cfd70f9c4f0aebdebc51","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/a9OVt11nTWAXafft1XW0L.png","isPro":false,"fullname":"Nagham Omar","user":"naghamo","type":"user","name":"naghamo"},"name":"Nagham Omar","status":"claimed_verified","statusLastChangedAt":"2026-09-10T00:45:05.071Z","hidden":false},{"_id":"6aa1d6c5a2aeb74440b1dcec","name":"Maya Rozenshtein","hidden":false},{"_id":"6aa1d6c5a2aeb74440b1dced","name":"Evgeny Mishlyakov","hidden":false},{"_id":"6aa1d6c5a2aeb74440b1dcee","name":"Avigdor Gal","hidden":false}],"mediaUrls":["https://cdn-uploads.huggingface.co/production/uploads/6995cfd70f9c4f0aebdebc51/ftlrKyqAn47zCawgO_-bL.png"],"publishedAt":"2026-09-06T00:00:00.000Z","submittedOnDailyAt":"2026-09-10T00:00:00.000Z","title":"Train Smarter, Not Harder: Switching Signal-Guided Training in Active Learning","submittedOnDailyBy":{"_id":"6995cfd70f9c4f0aebdebc51","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/a9OVt11nTWAXafft1XW0L.png","isPro":false,"fullname":"Nagham Omar","user":"naghamo","type":"user","name":"naghamo"},"summary":"Training strategy, namely whether to retrain from scratch or fine-tune from the previous checkpoint, is an overlooked decision variable in active learning. We show that this choice has exploitable structure: retraining is most useful in early rounds, when each batch can substantially reshape the labeled distribution, while fine-tuning becomes safer once the model trajectory stabilizes. We propose HybridAL, an adaptive training schedule that monitors an online stabilization signal and switches from retraining to fine-tuning after sustained stabilization. Two complementary signals, spectral exponent change Δα (weight-based) and accuracy change ΔAcc (validation-based), span different points on the time-calibration trade-off. Across three encoder backbones and six text-classification tasks (five seeds each), HybridAL keeps endpoint macro-F1 non-inferior to retraining and fine-tuning at a 0.010 margin, saves up to 49% of retraining time, and recovers a substantial fraction of retraining's calibration advantage as measured by negative log-likelihood (NLL). Compared with schedules that switch at a pre-committed round, HybridAL obtains lower NLL at moderate additional cost, showing that trajectory-dependent switching provides a stronger time-calibration trade-off than fixed early switching.","upvotes":7,"discussionId":"6aa1d6c6a2aeb74440b1dcef","githubRepo":"https://github.com/naghamo/hybridAL","githubRepoAddedBy":"user","ai_summary":"HybridAL adaptively switches from retraining to fine-tuning during active learning based on online stabilization signals, reducing training time while preserving accuracy and improving calibration.","ai_keywords":["active learning","retraining","fine-tuning","spectral exponent","calibration","negative log-likelihood","macro-F1","encoder backbones","text-classification"],"ai_summary_model":"thinkingmachines/Inkling-Small","githubStars":0,"organization":{"_id":"6393322be2364bc1eea56e45","name":"Technion","fullname":"Technion Israel institute of technology","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/1670591001944-63926124526c29d5b5011374.jpeg"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6995cfd70f9c4f0aebdebc51","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/a9OVt11nTWAXafft1XW0L.png","isPro":false,"fullname":"Nagham Omar","user":"naghamo","type":"user"},{"_id":"693de3cf54eeb72eef8aaf84","avatarUrl":"/avatars/3fcdb736903c8dbee16561ff9a2026bf.svg","isPro":false,"fullname":"Mahmoud Jabarin","user":"mahmoud48-2","type":"user"},{"_id":"699ee88d41369b8e677c72b2","avatarUrl":"/avatars/d38d9e5ee67aeb4b53d816d1fe1dc457.svg","isPro":false,"fullname":"Maya R","user":"MayaRoze","type":"user"},{"_id":"684d57f26e04c265777ead3f","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/cuOj-bQqukSZreXgUJlfm.png","isPro":false,"fullname":"Joakim Lee","user":"Reinforcement4All","type":"user"},{"_id":"676a8870430a9872e9b0ac7a","avatarUrl":"/avatars/481a0a52731ed16f80789572cdf02bf9.svg","isPro":false,"fullname":"Idan Horowitz","user":"idanh8","type":"user"},{"_id":"67850a03aee1dd1f38025ec4","avatarUrl":"/avatars/629b828d07ffce9eadea092cfeed8cf3.svg","isPro":false,"fullname":"Omri Lazover","user":"omri-lazover","type":"user"},{"_id":"6a2da6c8ca070ee12c6e396c","avatarUrl":"/avatars/0355287dcabaa67dbc7f0b10b87451f9.svg","isPro":false,"fullname":"Joe Mama","user":"JoeMama123123123","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"6393322be2364bc1eea56e45","name":"Technion","fullname":"Technion Israel institute of technology","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/1670591001944-63926124526c29d5b5011374.jpeg"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2609/2609.06806.md","query":{}}">
Train Smarter, Not Harder: Switching Signal-Guided Training in Active Learning
Abstract
HybridAL adaptively switches from retraining to fine-tuning during active learning based on online stabilization signals, reducing training time while preserving accuracy and improving calibration.
Training strategy, namely whether to retrain from scratch or fine-tune from the previous checkpoint, is an overlooked decision variable in active learning. We show that this choice has exploitable structure: retraining is most useful in early rounds, when each batch can substantially reshape the labeled distribution, while fine-tuning becomes safer once the model trajectory stabilizes. We propose HybridAL, an adaptive training schedule that monitors an online stabilization signal and switches from retraining to fine-tuning after sustained stabilization. Two complementary signals, spectral exponent change Δα (weight-based) and accuracy change ΔAcc (validation-based), span different points on the time-calibration trade-off. Across three encoder backbones and six text-classification tasks (five seeds each), HybridAL keeps endpoint macro-F1 non-inferior to retraining and fine-tuning at a 0.010 margin, saves up to 49% of retraining time, and recovers a substantial fraction of retraining's calibration advantage as measured by negative log-likelihood (NLL). Compared with schedules that switch at a pre-committed round, HybridAL obtains lower NLL at moderate additional cost, showing that trajectory-dependent switching provides a stronger time-calibration trade-off than fixed early switching.
Community
HybridAL treats the training strategy in active learning (retrain from scratch each round, or fine-tune from the previous checkpoint) as a decision variable rather than a fixed implementation detail.
The structure we found: retraining is most useful in early rounds, when each new batch can substantially reshape the labeled distribution, while fine-tuning becomes safe once the model trajectory stabilizes. HybridAL monitors an online stabilization signal and switches once, after sustained stabilization. We study two signals, spectral exponent change (Δα, weight-based) and accuracy change (ΔAcc, validation-based), which sit at different points on the time–calibration trade-off.
Across three encoder backbones and six text classification tasks, five seeds each: endpoint macro-F1 non-inferior to both retraining and fine-tuning at a 0.010 margin, up to 49% of retraining time saved, and a substantial fraction of retraining's calibration advantage recovered in NLL. Against schedules that switch at a pre-committed round, HybridAL reaches lower NLL at moderate additional cost.
Accepted to EMNLP 2026 Main Conference.
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/2609.06806 in a model README.md to link it from this page.
Cite arxiv.org/abs/2609.06806 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2609.06806 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.