Browse State-of-the-Art › Continual Learning

Continual Learning

1,142 papers with code · 33 benchmarks · 35 datasets archive 2025-07-28

Computer VisionMethodologyNatural Language Processing

Continual Learning (also known as Incremental Learning, Life-long Learning) is a concept to learn a model for a large number of tasks sequentially without forgetting knowledge obtained from the preceding tasks, where the data in the old tasks are not available anymore during training new ones.
If not mentioned, the benchmarks here are Task-CL, where task-id is provided on validation.

Source:
Continual Learning by Asymmetric Loss Approximation with Single-Side Overestimation
Three scenarios for continual learning
Lifelong Machine Learning
Continual lifelong learning with neural networks: A review

Description from the archive archive 2025-07-28.

Benchmarks archive 2025-07-28

33 leaderboard tables shown for this task, 33 with rows (a “benchmark” on this site is a table with at least one row, as on /sota), ordered by row count. “Best model” is the first row in the archive's own order at snapshot; nothing is re-ranked here and metric direction is not recorded in the archive. PwC's Trend sparklines are not in the archive, so that column is omitted. 10 shown of 33 until expanded.

Syntology 2 newer papers were judged to report on this task's benchmarks so far (Claude Sonnet 4.5, reading their tables: a model's judgement, not a result; 0 placed): Cifar100 (20 tasks) (1 judged, 0 placed), split CIFAR-100 (1 judged, 0 placed), 5-Datasets (1 judged, 0 placed), miniImagenet (1 judged, 0 placed). Drawn from 8,886 of 18,637 newer papers checked so far (733 more had nothing to place; 41 could not be read).

DatasetBest model (first row in archive order)PaperCodeSyntologyCompare
ASC (19 tasks) (15 rows) Multi-task Learning (MTL; Upper Bound) Achieving Forgetting Prevention and Knowledge Transfer in... code — Compare
visual domain decathlon (10 tasks) (14 rows) NetTailor NetTailor: Tuning the Architecture, Not Just the Weights code — Compare
Cifar100 (20 tasks) (9 rows) Model Zoo-Continual Model Zoo: A Growing "Brain" That Learns Continually code Syntology official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified Compare
Tiny-ImageNet (10tasks) (9 rows) ALTA-ViTB/16 Cross-Modal Alternating Learning with Task-Aware Representations... code — Compare
F-CelebA (10 tasks) (7 rows) CAT (CNN backbone) Continual Learning of a Mixed Sequence of Similar and Dissimilar Tasks code — Compare
20Newsgroup (10 tasks) (6 rows) CTR Achieving Forgetting Prevention and Knowledge Transfer in... code — Compare
CUBS (Fine-grained 6 Tasks) (6 rows) CondConvContinual EXTENDING CONDITIONAL CONVOLUTION STRUCTURES FOR ENHANCING... code — Compare
DSC (10 tasks) (6 rows) CTR Achieving Forgetting Prevention and Knowledge Transfer in... code — Compare
Flowers (Fine-grained 6 Tasks) (6 rows) CondConvContinual EXTENDING CONDITIONAL CONVOLUTION STRUCTURES FOR ENHANCING... code — Compare
ImageNet (Fine-grained 6 Tasks) (6 rows) ProgressiveNet Progressive Neural Networks code Syntology 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified Compare
Sketch (Fine-grained 6 Tasks) (6 rows) CondConvContinual EXTENDING CONDITIONAL CONVOLUTION STRUCTURES FOR ENHANCING... code — Compare
Stanford Cars (Fine-grained 6 Tasks) (6 rows) CPG Compacting, Picking and Growing for Unforgetting Continual Learning code Syntology official (archive's flag): 2 ran · 3 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified Compare
Wikiart (Fine-grained 6 Tasks) (6 rows) CondConvContinual EXTENDING CONDITIONAL CONVOLUTION STRUCTURES FOR ENHANCING... code — Compare
Cifar100 (10 tasks) (5 rows) ALTA-ViTB/16 Cross-Modal Alternating Learning with Task-Aware Representations... code — Compare
ImageNet-50 (5 tasks) (5 rows) RMN Understanding Catastrophic Forgetting and Remembering in Continual... code — Compare
Permuted MNIST (3 rows) RMN Understanding Catastrophic Forgetting and Remembering in Continual... code — Compare
split CIFAR-100 (2 rows) CODE-CL CODE-CL: Conceptor-Based Gradient Projection for Deep Continual Learning code — Compare
5-dataset - 1 epoch (1 row) TAG-RMSProp TAG: Task-based Accumulated Gradients for Lifelong learning code — Compare
5-Datasets (1 row) CODE-CL CODE-CL: Conceptor-Based Gradient Projection for Deep Continual Learning code — Compare
AIDS (1 row) TEST Advances and Challenges in Foundation Agents: From Brain-Inspired... code — Compare
CIFAR-100 AlexNet - 300 Epoch (1 row) IBM Towards Redundancy-Free Sub-networks in Continual Learning code — Compare
CIFAR-100 ResNet-18 - 300 Epochs (1 row) IBM Towards Redundancy-Free Sub-networks in Continual Learning code — Compare
Cifar100 (20 tasks) - 1 epoch (1 row) TAG-RMSProp TAG: Task-based Accumulated Gradients for Lifelong learning code — Compare
Coarse-CIFAR100 (1 row) Model Zoo-Continual Model Zoo: A Growing "Brain" That Learns Continually code Syntology official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified Compare
CUB-200-2011 (20 tasks) - 1 epoch (1 row) TAG-RMSProp TAG: Task-based Accumulated Gradients for Lifelong learning code — Compare
mini-Imagenet (20 tasks) - 1 epoch (1 row) TAG-RMSProp TAG: Task-based Accumulated Gradients for Lifelong learning code — Compare
miniImagenet (1 row) CODE-CL CODE-CL: Conceptor-Based Gradient Projection for Deep Continual Learning code — Compare
MiniImageNet ResNet-18 - 300 Epochs (1 row) IBM Towards Redundancy-Free Sub-networks in Continual Learning code — Compare
MLT17 (1 row) MRM MRN: Multiplexed Routing Network for Incremental Multilingual Text... code — Compare
Rotated MNIST (1 row) Model Zoo-Continual Model Zoo: A Growing "Brain" That Learns Continually code Syntology official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified Compare
Split CIFAR-10 (5 tasks) (1 row) H² Helpful or Harmful: Inter-Task Association in Continual Learning code — Compare
Split MNIST (5 tasks) (1 row) H² Helpful or Harmful: Inter-Task Association in Continual Learning code — Compare
TinyImageNet ResNet-18 - 300 Epochs (1 row) IBM Towards Redundancy-Free Sub-networks in Continual Learning code — Compare

Syntology column: samples harvested from the paper's repositories and executed on synthesized fixtures; “ran” is not a correctness claim and does not order the table. Each line reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. When the archive marks a repository official for the row's paper, the line starts with that repository's state (“official” is the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from. A dash means no Syntology record for that paper, not a recorded non-run. Each “ran” line links to that paper's sample list. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

Libraries

Not in the archive: the export carries no per-task library table, so there is nothing to show at snapshot 2025-07-28.

Datasets archive 2025-07-28

35 datasets whose archive record lists this task, ordered by the archive's paper count. 30 shown of 35 until expanded.

Subtasks archive 2025-07-28

5 subtasks in the archive's task tree.

Papers archive 2025-07-28

archive papers tagged: 2,644 · with a code link: 1,142 · where Syntology ran a sample: 337 (280 with a run with no instrument failure, 57 where every run was a failure of Syntology's instrument) Syntology

Show: all tagged papersonly where code ran (337 of 2,644 tagged: 280 with a run with no instrument failure, 57 where every run was a failure of Syntology's instrument)

Page 1 of 27: papers 1 to 100 of 2,644, in archive order: by repositories listed in the archive (most first), then newest first, not by stars (the archive holds no stars, so PwC's “Social” and “Latest” sorts cannot be reproduced). Papers that list no repository come after every paper that lists one.

Papers without a page here are shown as plain text. A Syntology line reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the figure “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code); hover it for the repositories the samples that ran came from. The first sentence of the abstract is shown for the first 30 papers; every abstract is on its paper's page.

Syntology lines on 51 of the papers shown; no Syntology record for the others (a paper without an arXiv id cannot be joined to the graph, and absence from the graph layer is not a recorded non-run). “Ran” means the sample executed on a synthesized fixture, not that the paper's result was reproduced; each line links to that paper's sample list. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections