Browse State-of-the-Art › Compositional Zero-Shot Learning
Compositional Zero-Shot Learning
31 papers with code · 4 benchmarks · 6 datasets archive 2025-07-28
Compositional Zero-Shot Learning (CZSL) is a computer vision task in which the goal is to recognize unseen compositions fromed from seen state and object during training. The key challenge in CZSL is the inherent entanglement between the state and object within the context of an image. Some example benchmarks for this task are MIT-states, UT-Zappos, and C-GQA. Models are usually evaluated with the Accuracy for both seen and unseen compositions, as well as their Harmonic Mean(HM).
( Image credit: Heosuab )
Description from the archive archive 2025-07-28.
Benchmarks archive 2025-07-28
4 leaderboard tables shown for this task, 4 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.
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): MIT-States (2 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).
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
6 datasets whose archive record lists this task, ordered by the archive's paper count.
Subtasks archive 2025-07-28
No subtask under this task in the archive's task tree.
Parent tasks archive 2025-07-28
Papers archive 2025-07-28
archive papers tagged: 65 · with a code link: 31 · where Syntology ran a sample: 18 (17 with a run with no instrument failure, 1 where every run was a failure of Syntology's instrument) Syntology
Show: all tagged papersonly where code ran (18 of 65 tagged: 17 with a run with no instrument failure, 1 where every run was a failure of Syntology's instrument)
Page 1 of 1: papers 1 to 65 of 65, 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.
-
26 May 2023 2 repositories listedIn this paper, we study the problem of Compositional Zero-Shot Learning (CZSL), which is to recognize novel attribute-object combinations with pre-existing concepts.
-
3 May 2021 2 repositories listedIn this work, we overcome this assumption operating on the open world setting, where no limit is imposed on the compositional space at test time, and the search space contains a large number of unseen compositions.
-
29 Jan 2021 2 repositories listedAfter estimating the feasibility score of each composition, we use these scores to either directly mask the output space or as a margin for the cosine similarity between visual features and compositional embeddings…
-
15 May 2025 1 repository listed Syntology official (archive's flag): 22 ran · 23 ran (of which 13 constructed an object rather than computing a result; 19 with no instrument failure: 0 honoured, 0 violated, 19 with no contract checked; 4 where Syntology's instrument failed) · 5 unverified (of 28 harvested samples) · 28 pointer-only (licence)To address this issue, we propose a Multi-Stage Cross-modal Interaction (MSCI) model that effectively explores and utilizes intermediate-layer information from CLIP's visual encoder.
-
10 Feb 2025 1 repository listed Syntology official (archive's flag): 2 ran · 2 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified (of 3 harvested samples) · 3 pointer-only (licence)Existing CZSL solutions typically rely on oversimplified data assumptions, e. g., modeling each primitive with a single centroid primitive representation, ignoring the natural diversities of the attribute (resp.
-
5 Dec 2024 1 repository listedOpen-World Compositional Zero-Shot Learning (OW-CZSL) addresses the challenge of recognizing novel compositions of known primitives and entities.
-
18 Nov 2024 1 repository listedMoreover, we propose attribute smoothing with auxiliary attributes generated by Large Language Model (LLM) for seen compositions, addressing the issue of overconfidence by encouraging the model to learn more attributes…
-
18 Jul 2024 1 repository listedWe are exploring Open World Compositional Zero-Shot Learning (OW-CZSL) in this study, where our test space encompasses all potential combinations of attributes and objects.
-
Contextual Interaction via Primitive-based Adversarial Training For Compositional Zero-shot Learning21 Jun 2024 1 repository listedInspired by the success of vanilla adversarial learning in Cross-Domain Few-Shot Learning, we take a step further and devise a model-agnostic and Primitive-Based Adversarial training (PBadv) method to deal with this…
-
30 Nov 2023 1 repository listed Syntology official (archive's flag): 4 ran · 4 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; 3 where Syntology's instrument failed) · 7 unverified (of 11 harvested samples) · 11 pointer-only (licence)With this in mind, we propose a simple yet effective approach to optimize VLMs in fine-grained understanding, achieving significant improvements on SPEC without compromising the zero-shot performance.
-
9 Nov 2023 1 repository listed Syntology official (archive's flag): 7 ran · 7 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 0 where Syntology's instrument failed) · 5 unverified (of 12 harvested samples) · 12 pointer-only (licence)In this work, we propose GIP-COL (Graph-Injected Soft Prompting for COmpositional Learning) to better explore the compositional zero-shot learning (CZSL) ability of VLMs within the prompt-based learning framework.
-
8 Aug 2023 1 repository listed Syntology official (archive's flag): 11 ran · 11 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 1 honoured, 0 violated, 7 with no contract checked; 3 where Syntology's instrument failed) · 1 unverified (of 12 harvested samples) · 12 pointer-only (licence)Previous works for CZSL often suffer from grasping the contextuality between attribute and object, as well as the discriminability of visual features, and the long-tailed distribution of real-world compositional data.
-
29 May 2023 1 repository listed Syntology official (archive's flag): 6 ran · 6 ran (of which 4 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 2 where Syntology's instrument failed) · 3 unverified (of 9 harvested samples) · 9 pointer-only (licence)Compositional Zero-Shot Learning (CZSL) aims to train models to recognize novel compositional concepts based on learned concepts such as attribute-object combinations.
-
23 May 2023 1 repository listed Syntology official (archive's flag): 7 ran · 7 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 3 where Syntology's instrument failed) · 1 unverified (of 8 harvested samples) · 3 pointer-only (licence)Compositional zero-shot learning (CZSL) task aims to recognize unseen compositional visual concepts, e. g., sliced tomatoes, where the model is learned only from the seen compositions, e. g., sliced potatoes and red…
-
27 Mar 2023 1 repository listed Syntology official (archive's flag): 8 ran · 8 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 3 where Syntology's instrument failed) · 3 unverified (of 11 harvested samples) · 2 pointer-only (licence)The key to CZSL is learning the disentanglement of the attribute-object composition.
-
27 Mar 2023 1 repository listedRecent compositional zero-shot learning (CZSL) methods adapt pre-trained vision-language models (VLMs) by constructing trainable prompts only for composed state-object pairs.
-
19 Nov 2022 1 repository listed Syntology official (archive's flag): 3 ran · 3 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified (of 4 harvested samples) · 2 pointer-only (licence)Existing methods either learn the combined state-object representation, challenging the generalization of unseen compositions, or design two classifiers to identify state and object separately from image features…
-
22 Aug 2022 1 repository listedCompositional zero-shot learning (CZSL) refers to recognizing unseen compositions of known visual primitives, which is an essential ability for artificial intelligence systems to learn and understand the world.
-
29 Jun 2022 1 repository listed Syntology official (archive's flag): 3 ran · 3 ran (of which 1 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) · 4 unverified (of 7 harvested samples) · 7 pointer-only (licence)Some methods recognize state and object with two trained classifiers, ignoring the impact of the interaction between object and state; the other methods try to learn the joint representation of the state-object…
-
1 Jun 2022 1 repository listed Syntology official (archive's flag): 2 ran · 2 ran (of which 2 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; every one of the 2 samples that ran constructed an object rather than computing a result (of 2 harvested samples)Compositional Zero-Shot Learning (CZSL) aims to recognize novel compositions using knowledge learned from seen attribute-object compositions in the training set.
-
17 May 2022 1 repository listedWe use visual decomposed features to hallucinate embeddings that are representative for the seen and novel compositions to better regularize the learning of our model.
-
13 May 2022 1 repository listed Syntology official: harvested, nothing ran · 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified; the one sample that ran constructed an object rather than computing a result (of 2 harvested samples) · 2 pointer-only (licence)The goal of open-world compositional zero-shot learning (OW-CZSL) is to recognize compositions of state and objects in images, given only a subset of them during training and no prior on the unseen compositions.
-
7 Apr 2022 1 repository listed 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: 1 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified (of 2 harvested samples)We perform additional experiments to show that CSP improves generalization to higher-order attribute-attribute-object compositions (e. g., old white cat) and combinations of pretrained attributes and fine-tuned objects.
-
3 Mar 2022 1 repository listed 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 (of 2 harvested samples) · 2 pointer-only (licence)We perform extensive experiments on over ten datasets and the proposed method achieves significant improvements on different data scarcity applications without any bells and whistles, including the tasks of long-tailed…
-
9 Oct 2021 1 repository listed Syntology official (archive's flag): 3 ran · 3 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified (of 3 harvested samples)To model the compositional nature of these concepts, it is a good choice to learn them as transformations, e. g., coupling and decoupling.
-
10 Aug 2021 1 repository listedThe concept module generates semantically meaningful features for primitive concepts, whereas the visual module extracts visual features for attributes and objects from input images.
-
1 Jun 2021 1 repository listedNext we propagate the independent prototypes through a compositional graph, to learn compositional prototypes of novel attribute-object combinations that reflect the dependencies of the target distribution.
-
3 Feb 2021 1 repository listed Syntology official (archive's flag): 6 ran · 6 ran (of which 3 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 3 where Syntology's instrument failed) · 5 unverified (of 11 harvested samples) · 11 pointer-only (licence)In compositional zero-shot learning, the goal is to recognize unseen compositions (e. g. old dog) of observed visual primitives states (e. g. old, cute) and objects (e. g. car, dog) in the training set.
-
25 Jun 2020 1 repository listedThis leads to consistent misclassification of samples from a new distribution, like new combinations of known components.
-
1 Apr 2020 1 repository listedTo model the compositional nature of these general concepts, it is a good choice to learn them through transformations, such as coupling and decoupling.
-
27 Mar 2018 1 repository listed 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) · 1 unverified (of 3 harvested samples)
-
EVA: Mixture-of-Experts Semantic Variant Alignment for Compositional Zero-Shot Learning26 Jun 2025 0 repositories listed
-
Feasibility with Language Models for Open-World Compositional Zero-Shot Learning16 May 2025 0 repositories listed
-
Visual Adaptive Prompting for Compositional Zero-Shot Learning27 Feb 2025 0 repositories listed
-
Learning Primitive Relations for Compositional Zero-Shot Learning24 Jan 2025 0 repositories listed
-
23 Jan 2025 0 repositories listed Syntology 4 ran (of which 3 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified (of 8 harvested samples) · 8 pointer-only (licence)
-
Separated Inter/Intra-Modal Fusion Prompts for Compositional Zero-Shot Learning22 Jan 2025 0 repositories listed
-
Exploring Transferable Homogeneous Groups for Compositional Zero-Shot Learning18 Jan 2025 0 repositories listed
-
Duplex: Dual Prototype Learning for Compositional Zero-Shot Learning13 Jan 2025 0 repositories listed
-
Beyond Image Classification: A Video Benchmark and Dual-Branch Hybrid Discrimination Framework for Compositional Zero-Shot Learning1 Jan 2025 0 repositories listed
-
LOGICZSL: Exploring Logic-induced Representation for Compositional Zero-shot Learning1 Jan 2025 0 repositories listed
-
Compositional Zero-Shot Learning with Contextualized Cues and Adaptive Contrastive Training10 Dec 2024 0 repositories listed
-
Focus-Consistent Multi-Level Aggregation for Compositional Zero-Shot Learning30 Aug 2024 0 repositories listed
-
Cross-composition Feature Disentanglement for Compositional Zero-shot Learning19 Aug 2024 0 repositories listed
-
TsCA: On the Semantic Consistency Alignment via Conditional Transport for Compositional Zero-Shot Learning16 Aug 2024 0 repositories listed
-
Anticipating Future Object Compositions without Forgetting15 Jul 2024 0 repositories listed
-
MAC: A Benchmark for Multiple Attributes Compositional Zero-Shot Learning18 Jun 2024 0 repositories listed
-
CSCNET: Class-Specified Cascaded Network for Compositional Zero-Shot Learning9 Mar 2024 0 repositories listed
-
Context-based and Diversity-driven Specificity in Compositional Zero-Shot Learning27 Feb 2024 0 repositories listed
-
Beyond Seen Primitive Concepts and Attribute-Object Compositional Learning1 Jan 2024 0 repositories listed
-
Revealing the Proximate Long-Tail Distribution in Compositional Zero-Shot Learning26 Dec 2023 0 repositories listed
-
Compositional Zero-Shot Learning for Attribute-Based Object Reference in Human-Robot Interaction21 Dec 2023 0 repositories listed
-
Prompt Tuning for Zero-shot Compositional Learning2 Dec 2023 0 repositories listed
-
Compositional Zero-shot Learning via Progressive Language-based Observations23 Nov 2023 0 repositories listed
-
HOMOE: A Memory-Based and Composition-Aware Framework for Zero-Shot Learning with Hopfield Network and Soft Mixture of Experts23 Nov 2023 0 repositories listed
-
DRPT: Disentangled and Recurrent Prompt Tuning for Compositional Zero-Shot Learning2 May 2023 0 repositories listed
-
Distilled Reverse Attention Network for Open-world Compositional Zero-Shot Learning1 Mar 2023 0 repositories listed
-
Mutual Balancing in State-Object Components for Compositional Zero-Shot Learning19 Nov 2022 0 repositories listed
-
ProCC: Progressive Cross-primitive Compatibility for Open-World Compositional Zero-Shot Learning19 Nov 2022 0 repositories listed
-
Simple Primitives with Feasibility- and Contextuality-Dependence for Open-World Compositional Zero-shot Learning5 Nov 2022 0 repositories listed
-
Learning Attention Propagation for Compositional Zero-Shot Learning20 Oct 2022 0 repositories listed
-
On Leveraging Variational Graph Embeddings for Open World Compositional Zero-Shot Learning23 Apr 2022 0 repositories listed
-
3D Compositional Zero-shot Learning with DeCompositional Consensus29 Nov 2021 0 repositories listed
-
22 Oct 2021 0 repositories listed
-
Compositional Zero-Shot Learning via Fine-Grained Dense Feature Composition1 Dec 2020 0 repositories listed
Syntology lines on 18 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