Browse State-of-the-Art › Zero-Shot Composed Image Retrieval (ZS-CIR)
Zero-Shot Composed Image Retrieval (ZS-CIR)
25 papers with code · 12 benchmarks · 11 datasets archive 2025-07-28
Given a query composed of a reference image and a relative caption, Composed Image Retrieval (CIR) aims to retrieve target images that are visually similar to the reference one but incorporate the changes specified in the relative caption. The bi-modality of the query provides users with more precise control over the characteristics of the desired image, as some features are more easily described with language, while others can be better expressed visually.
Zero-Shot Composed Image Retrieval (ZS-CIR) is a subtask of CIR that aims to design an approach that manages to combine the reference image and the relative caption without the need for supervised learning.
Description from the archive archive 2025-07-28.
Benchmarks archive 2025-07-28
12 leaderboard tables shown for this task, 12 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 12 until expanded.
Syntology 3 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; 1 placed): CIRR (3 judged, 0 placed), CIRCO (1 judged, 1 placed), Fashion IQ (2 judged, 0 placed), FashionIQ (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
11 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: 36 · with a code link: 25 · where Syntology ran a sample: 13 (9 with a run with no instrument failure, 4 where every run was a failure of Syntology's instrument) Syntology
Show: all tagged papersonly where code ran (13 of 36 tagged: 9 with a run with no instrument failure, 4 where every run was a failure of Syntology's instrument)
Page 1 of 1: papers 1 to 36 of 36, 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.
-
28 Oct 2024 2 repositories listedTo address this limitation, this paper proposes a training-free method called Semantic Editing Increment for ZS-CIR (SEIZE) to retrieve the target image based on the query image and text without training.
-
11 Jul 2024 2 repositories listedTo conduct ZS-CIR, the prevailing methods employ pre-trained image-to-text models to transform the query image and text into a single text, which is then projected into the common feature space by CLIP to retrieve the…
-
5 May 2024 2 repositories listed Syntology official: no sample here; runs from other or unrecorded repositories · 3 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified (of 4 harvested samples) · 4 pointer-only (licence)Given a query consisting of a reference image and a relative caption, Composed Image Retrieval (CIR) aims to retrieve target images visually similar to the reference one while incorporating the changes specified in the…
-
27 Mar 2023 2 repositories listedComposed Image Retrieval (CIR) aims to retrieve a target image based on a query composed of a reference image and a relative caption that describes the difference between the two images.
-
4 Apr 2022 2 repositories listed Syntology official (archive's flag): 1 ran · 1 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; 0 where Syntology's instrument failed) · 0 unverified (of 1 harvested sample) · 1 pointer-only (licence)We propose an architecture for solving PerVL that operates by extending the input vocabulary of a pretrained model with new word embeddings for the new personalized concepts.
-
25 Mar 2025 1 repository listedExperimental results demonstrate that CoLLM achieves state-of-the-art performance across multiple CIR benchmarks and settings.
-
21 Mar 2025 1 repository listedThe key challenge for ZS-CIR tasks is to modify a reference image according to manipulation text to accurately retrieve a target image, especially when the reference image is missing essential target content.
-
ImageScope: Unifying Language-Guided Image Retrieval via Large Multimodal Model Collective Reasoning13 Mar 2025 1 repository listedTo this end, we propose ImageScope, a training-free, three-stage framework that leverages collective reasoning to unify LGIR tasks.
-
19 Dec 2024 1 repository listedDespite the rapidly growing demand for multimodal retrieval, progress in this field remains severely constrained by a lack of training data.
-
15 Dec 2024 1 repository listedExisting training-free zero-shot CIR (ZS-CIR) methods often employ a two-stage process: they first generate a caption for the reference image and then use Large Language Models for reasoning to obtain a target…
-
4 Dec 2024 1 repository listedThis work addresses composed image retrieval in the context of domain conversion, where the content of a query image is retrieved in the domain specified by the query text.
-
7 Sep 2024 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) · 0 unverified (of 2 harvested samples)Our approach is simple, easy to implement, and its effectiveness is validated through experiments on the FashionIQ and CIRR datasets.
-
13 Jun 2024 1 repository listedHowever, we highlight an inherent limitation in these projection-based CIR: a task discrepancy of text encoders between the original pre-training task of the encoders (text ↔ image) and the target CIR task (image + text…
-
24 May 2024 1 repository listedVarious attributes can be modified by the textual part, such as shape, color, or context.
-
17 Apr 2024 1 repository listed Syntology official (archive's flag): 3 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 (of 4 harvested samples)In addition, our method also performs well in zero-shot composed image retrieval, providing a new CIR solution for the low-resources scenario.
-
28 Mar 2024 1 repository listed Syntology official (archive's flag): 1 ran · 1 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; 1 where Syntology's instrument failed) · 0 unverified (of 1 harvested sample)MagicLens is built on a key novel insight: image pairs that naturally occur on the same web pages contain a wide range of implicit relations (e. g., inside view of), and we can bring those implicit relations explicit by…
-
24 Mar 2024 1 repository listed Syntology official (archive's flag): 10 ran · 10 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 1 honoured, 2 violated, 4 with no contract checked; 3 where Syntology's instrument failed) · 7 unverified (of 17 harvested samples) · 2 pointer-only (licence)The pseudo-word tokens generated in this stream are explicitly aligned with fine-grained semantics in the text embedding space.
-
4 Dec 2023 1 repository listed Syntology official (archive's flag): 3 ran · 3 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; 1 where Syntology's instrument failed) · 5 unverified (of 8 harvested samples) · 8 pointer-only (licence)Our LinCIR (Language-only training for CIR) can be trained only with text datasets by a novel self-supervision named self-masking projection (SMP).
-
13 Nov 2023 1 repository listed Syntology official (archive's flag): 3 ran · 3 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified (of 4 harvested samples) · 4 pointer-only (licence)Then, we propose a simple but novel pre-trained masked tuning method, which uses the text and the masked image to learn the modifications of the original image.
-
13 Oct 2023 1 repository listed Syntology official (archive's flag): 4 ran · 4 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 1 honoured, 0 violated, 2 with no contract checked; 1 where Syntology's instrument failed) · 4 unverified (of 8 harvested samples) · 2 pointer-only (licence)Finally, we show that CIReVL makes CIR human-understandable by composing image and text in a modular fashion in the language domain, thereby making it intervenable, allowing to post-hoc re-align failure cases.
-
28 Sep 2023 1 repository listed Syntology official (archive's flag): 1 ran · 1 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; 1 where Syntology's instrument failed) · 0 unverified (of 1 harvested sample)Different from Composed Image Retrieval task that requires expensive labels for training task-specific models, Zero-Shot Composed Image Retrieval (ZS-CIR) involves diverse tasks with a broad range of visual content…
-
28 Aug 2023 1 repository listed Syntology official (archive's flag): 1 ran · 1 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; 1 where Syntology's instrument failed) · 0 unverified (of 1 harvested sample)Composed Image Retrieval (CoIR) has recently gained popularity as a task that considers both text and image queries together, to search for relevant images in a database.
-
12 Jun 2023 1 repository listedIn this paper, we consider the problem of composed image retrieval (CIR), it aims to train a model that can fuse multi-modal information, e. g., text and images, to accurately retrieve images that match the query…
-
21 Mar 2023 1 repository listed Syntology official (archive's flag): 4 ran · 4 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 2 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified (of 6 harvested samples) · 1 pointer-only (licence)This paper proposes a novel diffusion-based model, CompoDiff, for solving zero-shot Composed Image Retrieval (ZS-CIR) with latent diffusion.
-
6 Feb 2023 1 repository listed Syntology official (archive's flag): 1 ran · 1 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; 0 where Syntology's instrument failed) · 0 unverified (of 1 harvested sample)Existing methods rely on supervised learning of CIR models using labeled triplets consisting of the query image, text specification, and the target image.
-
MLLM-Guided VLM Fine-Tuning with Joint Inference for Zero-Shot Composed Image Retrieval26 May 2025 0 repositories listedExisting Zero-Shot Composed Image Retrieval (ZS-CIR) methods typically train adapters that convert reference images into pseudo-text tokens, which are concatenated with the modifying text and processed by frozen text…
-
Multimodal Reasoning Agent for Zero-Shot Composed Image Retrieval26 May 2025 0 repositories listedZero-Shot Composed Image Retrieval (ZS-CIR) aims to retrieve target images given a compositional query, consisting of a reference image and a modifying text-without relying on annotated training data.
-
Data-Efficient Generalization for Zero-shot Composed Image Retrieval7 Mar 2025 0 repositories listedZero-shot Composed Image Retrieval (ZS-CIR) aims to retrieve the target image based on a reference image and a text description without requiring in-distribution triplets for training.
-
CoTMR: Chain-of-Thought Multi-Scale Reasoning for Training-Free Zero-Shot Composed Image Retrieval28 Feb 2025 0 repositories listedZero-Shot Composed Image Retrieval (ZS-CIR) aims to retrieve target images by integrating information from a composed query (reference image and modification text) without training samples.
-
PDV: Prompt Directional Vectors for Zero-shot Composed Image Retrieval11 Feb 2025 0 repositories listedZero-shot composed image retrieval (ZS-CIR) enables image search using a reference image and text prompt without requiring specialized text-image composition networks trained on large-scale paired data.
-
12 Jan 2025 0 repositories listed
-
24 Nov 2024 0 repositories listed
-
MoTaDual: Modality-Task Dual Alignment for Enhanced Zero-shot Composed Image Retrieval31 Oct 2024 0 repositories listed
-
Denoise-I2W: Mapping Images to Denoising Words for Accurate Zero-Shot Composed Image Retrieval22 Oct 2024 0 repositories listed
-
Training-free Zero-shot Composed Image Retrieval with Local Concept Reranking14 Dec 2023 0 repositories listed
-
13 Jun 2023 0 repositories listed
Syntology lines on 13 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