AI Paper Atlas

Research map / Continual learning / Continual fine-tuning of LLMs

Continual Action Quality Assessment via Adaptive Manifold-Aligned Graph Regularization

Kanglei Zhou, Qingyi Pan, Xingxing Zhang, Hubert P. H. Shum, Frederick W. B. Li, Xiaohui Liang, Liyuan Wang

TPAMI 2026 · Continual learning › Continual fine-tuning of LLMs · approach: Synthetic data and merging for CLIP · 9 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
full-parameter fine-tuning instead of a fixed backboneAQA regression heads (CoRe, HGCN, CoFInAl) · MTL-AQA, AQA-7, FineDivingSRCC / rMSE: +3.33% SRCC and -31.23% rMSEImproved
full-parameter fine-tuning instead of I3D-Adapter PEFTAQA regression heads · MTL-AQA, AQA-7, FineDivingSRCC / rMSE: +2.11% SRCC and -14.25% rMSEImproved
full fine-tuning instead of adaptersAQA regression heads · AQA-7SRCC / rMSE: +5.89% SRCC, -44.21% rMSEImproved
full fine-tuning instead of adaptersAQA regression heads · FineDivingSRCC / rMSE: +0.22% SRCC, -2.85% rMSENo clear effect
full fine-tuning instead of adaptersAQA regression heads · MTL-AQASRCC / rMSE: +0.23% SRCC, +4.31% rMSEMixed
upgrade MAGR to MAGR++ (layer-adaptive FPFT, manifold projector, IIJ-GR)MAGR · MTL-AQA (default, w/o difficulty label), offlineSRCC rho_avg: 0.8979 -> 0.9205Improved
upgrade MAGR to MAGR++MAGR · FineDiving (w/ dive number), offlineSRCC rho_avg: 0.8580 -> 0.8902Improved
upgrade MAGR to MAGR++MAGR · UNLV-Dive, offlineSRCC rho_avg: 0.7668 -> 0.8165Improved
upgrade MAGR to MAGR++MAGR · MTL-AQA (w/ difficulty), offlineSRCC rho_avg: 0.9237 -> 0.9383Improved
upgrade MAGR to MAGR++MAGR · UNLV-Dive, onlineSRCC rho_avg: 0.4202 -> 0.5117Improved
MAGR++ vs strongest baseline across offline and online CAQAMAGR++ · four CAQA benchmarksSRCC rho_avg: 1.6%-6.5% offline and 4.0%-21.8% online; average gains of 3.6% and 12.2%Improved
feature replay methods vs raw-data replay methodsfeature replay baselines · MTL-AQA, offlineSRCC rho_avgHurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy