Composed Image Retrieval (or, Image Retreival conditioned on Language Feedback) is a relatively new retrieval task, where an input query consists of an image and short textual description of how to modify the image.
For humans, the advantage of a bi-modal query is clear: some concepts and attributes are more succinctly described visually, others through language. By cross-referencing the two modalities, a reference image can capture the general gist of a scene, while the text can specify finer details.
We identify a major challenge of this task as the inherent ambiguity in knowing what information is important (typically one object of interest in the scene) and what can be ignored (e.g., the background and other irrelevant objects).
We release the first dataset of open-domain, real-life images with human-generated modification sentences, which support research on one-shot composed image retrieval, dialogue systems, fine-grained visiolinguistic reasoning, and more.
All 3 leaderboards whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.
Syntology 5 newer papers were judged to report on this dataset's benchmarks so far (Claude Sonnet 4.5, reading their tables: a model's judgement, not a result; 0 placed): Zero-Shot Composed Image Retrieval (ZS-CIR) · CIRR (3 judged, 0 placed), Composed Image Retrieval (CoIR) · CIRR (3 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).
papers with a benchmark row: 35 · with a code link: 32 · where Syntology ran a sample: 16 (12 with a run with no instrument failure, 4 where every run was a failure of Syntology's instrument) Syntology
Page 1 of 1: papers 1 to 35 of 35 with a leaderboard row on this dataset's benchmarks, newest first by the archive's date (ties by slug; undated papers last).
The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset, not that list; the archive's count for this dataset is 61. The Syntology column is from Syntology's graph, stated per sample; it is not part of any archive number. A 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” (C is a part of N, never taken away from it); the figure “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. “Ran” means the sample executed on a synthesized fixture, not that the paper's result was reproduced. When the archive marks a repository official for the paper, the cell starts with that repository's state (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. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record.