DCR-TSF: Dual-Level Compositional Reasoning with Triadic Spherical Fusion for Training-Free Zero-Shot Composed Image Retrieval

Gwang-Ho Na; Ho-Joong Kim; Seong-Whan Lee

cs.CVOct 19, 2026

DCR-TSF: Dual-Level Compositional Reasoning with Triadic Spherical Fusion for Training-Free Zero-Shot Composed Image Retrieval

Authors: Gwang-Ho Na, Ho-Joong Kim, Seong-Whan Lee

Organizations: Dept. of Artificial Intelligence, Korea University, Seoul, Korea

Abstract

Zero-shot composed image retrieval (ZS-CIR) aims to retrieve a target image that reflects an intended textual modification while maintaining semantic consistency with a reference image. Existing training-free ZS-CIR methods typically rely on fine-grained generated target descriptions, which often overemphasize reference-specific visual details and lead to appearance-biased retrieval. Moreover, since these descriptions are generated through multimodal large language models (MLLMs), the direct influence of the original image-text query can be weakened in the final retrieval representation. In this paper, we propose DCR-TSF, a training-free ZS-CIR framework that combines Dual-level Compositional Reasoning (DCR) and Triadic Spherical Fusion (TSF). DCR generates fine- and coarse-grained descriptions from the reference image and modification text, capturing complementary target-side semantics across different granularities. To preserve the original image-text query information, TSF constructs a composed-query anchor by spherically interpolating the reference image feature and modification text feature, and fuses it with the fine- and coarse-grained representations using a weighted Karcher mean on the unit hypersphere. The resulting final query representation balances detailed visual evidence, high-level semantic abstraction, and direct preservation of the original composed query. Extensive experiments on CIRCO, CIRR, and Fashion-IQ demonstrate that our method achieves state-of-the-art performance among training-free methods.

Figures & tables

Panel of Figure 1 — Figure 1: Biases in Fine- and Coarse-grained Descriptions for ZS-CIR. Colored text refers to observed elements in the reference image. Background color indicates phrases appearing in the modification text. The bar colors in the text-to-image similarity score plots correspond to the candidate images. (a): Fine-grained description is biased towards the visual details of the reference image. (b): Coarse-grained description lacks details of the reference image.

Figure 2: Overview of DCR-TSF. Given a reference image $I_{r}$ and a modification text $D_{m}$ , DCR generates fine- and coarse-grained descriptions, which are encoded into $T_{\text{fine}}$ and $T_{\text{coarse}}$ . TSF constructs a composed-query anchor $Q_{\text{comp}}$ from $V_{r}$ and $T_{m}$ , and fuses it with $T_{\text{fine}}$ and $T_{\text{coarse}}$ using a weighted Karcher mean [ 11 ] . The resulting final query representation $T_{\text{final}}$ is used to compute the final retrieval score $S^{(i)}$ .

Figure 3: Comparison between OSrCIR and our coarse-grained description prompt. OSrCIR [ 28 ] generates a detailed target description through a CoT prompt, preserving rich reference-specific visual evidence that may be irrelevant to the modification condition. In contrast, our coarse-grained description prompt identifies stable visual anchors and focuses on modification-related semantics, omitting excessive reference-specific details and providing a complementary abstraction for DCR.

Backbone	Method	Training -free	CIRCO				CIRR
Backbone	Method	Training -free	mAP@5	mAP@10	mAP@25	mAP@50	R@1	R@5	R@10	R sub @1	R sub @2	R sub @3	Avg.
ViT-B/32	PALAVRA [ 6 ]	✗	4.61	5.32	6.33	6.80	16.62	43.49	58.51	41.61	65.30	80.94	42.55
	SEARLE [ 4 ]	✗	9.35	9.94	11.13	11.84	24.00	53.42	66.82	54.89	76.60	88.19	54.16
	Slerp [ 10 ]	✓	6.35	7.11	8.12	8.75	24.22	50.94	63.49	57.86	78.27	89.25	54.40
	CIReVL [ 12 ]	✓	14.94	15.42	17.00	17.82	23.94	52.51	66.00	60.17	80.05	90.19	56.34
	LDRE [ 34 ]	✓	17.96	18.32	20.21	21.11	25.69	55.13	69.04	60.53	80.65	90.70	57.83

Table 1: Comparison with the state-of-the-art methods on the CIRCO and CIRR test sets. Best results are highlighted in bold, and the second best are underlined.

Backbone	Method	Training -free	Shirt		Dress		Tops&Tee		Avg.
Backbone	Method	Training -free	R@10	R@50	R@10	R@50	R@10	R@50	R@10	R@50	R mean
ViT-B/32	PALAVRA [ 6 ]	✗	21.49	37.05	17.25	35.94	20.55	38.76	19.76	37.25	28.50
	SEARLE [ 4 ]	✗	24.44	41.61	18.54	39.51	25.70	46.46	22.89	42.53	32.71
	Slerp [ 10 ]	✓	20.53	41.00	23.75	40.92	26.98	46.77	23.75	42.90	33.33
	CIReVL [ 12 ]	✓	28.36	47.84	25.29	46.36	31.21	53.85	28.29	49.35	38.82
	LDRE [ 34 ]	✓	27.38	46.27	19.97	41.84	27.07	48.78	24.81	45.63	35.22

Table 2: Comparison with the state-of-the-art methods on Fashion-IQ. R

{}_{\text{mean}}

represents the average results across all the metrics. Best results are highlighted in bold, and the second best are underlined.

FD	CD	Comp	CIRCO		CIRR		Fashion-IQ
FD	CD	Comp	mAP@5	mAP@10	R@5	R sub @1	R@10	R@50
✓			25.55	26.66	60.00	61.54	35.57	56.03
	✓		32.03	33.34	64.29	69.06	32.83	52.84
		✓	17.32	17.88	62.75	67.42	40.58	60.39
✓	✓		34.77	35.95	67.04	68.15	38.43	59.35
✓		✓	32.01	32.88	69.33	67.66	43.86	65.23

Table 3: Ablation study with CLIP ViT-G/14. FD, CD, and Comp denote fine-grained descriptions, coarse-grained descriptions, and the composed-query anchor, respectively. The last row corresponds to the full DCR-TSF.

Panel of Figure 4 — Figure 4: Hyperparameter sensitivity of TSF fusion weights. Ternary plots show validation results for different TSF fusion-weight configurations on each dataset using CLIP ViT-G/14. The star marker indicates the best-performing configuration selected for the main experiments.

Method	mAP@5	mAP@10	mAP@25	mAP@50
Qwen2.5-VL-3B [ 3 ]	23.72	24.89	27.09	28.09
Qwen2.5-VL-7B [ 3 ]	30.62	31.86	34.29	35.42
Qwen2.5-VL-32B [ 3 ]	34.77	35.95	38.56	39.79
Qwen2.5-VL-72B [ 3 ]	35.26	36.69	39.46	40.62
Qwen2-VL-72B [ 32 ]	34.38	36.11	38.81	39.89
Qwen3-VL-4B [ 31 ]	33.21	34.48	37.23	38.37

Table 5: Analysis on scalability and model variation in MLLMs on the CIRCO test set with CLIP ViT-G/14.

BLIP-2 & GPT-3.5-turbo	Multi-caption generator	LLM-based editing reasoner	Total
SEIZE [ 33 ]	1.560 $\pm$ 0.056	3.912 $\pm$ 0.812	5.472 $\pm$ 0.791
Qwen2.5-VL-32B	Image-scale reasoning	Object-scale reasoning	Total
CoTMR [ 27 ]	1.892 $\pm$ 0.011	1.661 $\pm$ 0.014	3.553 $\pm$ 0.023
Qwen2.5-VL-32B	Fine-grained description	Coarse-grained description	Total
DCR-TSF (Ours)	1.883 $\pm$ 0.019	1.584 $\pm$ 0.018	3.467 $\pm$ 0.030

Table 6: Inference time comparison with existing training-free ZS-CIR methods. We report the average latency per composed query in seconds. Each column decomposes the total inference time into the main reasoning stages of each method.

Panel of Figure 6 — Figure 6: Qualitative comparison on CIRCO with ViT-G/14. Each example shows a reference image, its modification text, and the retrieved images with their ranks under four settings: Fine-Grained Description , Coarse-Grained Description , DCR , and DCR-TSF . The ground-truth image is highlighted in green.

Panel of Figure 12 — Figure A : Prompt of fine descriptions $P_{\text{fine}}$

Panel of Figure 13 — Figure B : Prompt of coarse descriptions $P_{\text{coarse}}$

Backbone	FD	CD	Comp	CIRCO		CIRR		Fashion-IQ
Backbone	FD	CD	Comp	mAP@5	mAP@10	R@5	R sub @1	R@10	R@50
ViT-B/32	✓			18.21	18.83	55.88	58.96	32.08	52.84
		✓		21.28	21.89	59.98	65.88	29.91	50.87
			✓	11.00	11.41	59.21	63.90	33.97	54.49
	✓	✓		22.95	23.76	62.29	65.16	35.57	57.02
	✓		✓	22.15	22.86	65.74	64.89	39.50	60.82

Table A : Extended ablation study across CLIP backbones. FD, CD, and Comp denote the fine-grained representation, coarse-grained representation, and composed-query anchor, respectively. We report results on CIRCO, CIRR, and Fashion-IQ using ViT-B/32, ViT-L/14, and ViT-G/14.

Backbone	Fusion Strategy	CIRCO	CIRR	Fashion-IQ
Backbone	Fusion Strategy	mAP@5	R@5	R mean
ViT-B/32	Weighted Avg.	25.11	67.18	51.32
	Sequential Slerp	25.05	67.22	51.31
	Karcher Mean	25.07	67.28	51.33
ViT-L/14	Weighted Avg.	31.92	69.95	51.46
	Sequential Slerp	31.56	69.88	51.41

Table B : Comparison of fusion strategies across CLIP backbones. Results are reported on CIRCO, CIRR, and Fashion-IQ with ViT-B/32 and ViT-L/14.

Method	Reasoning Backbone	mAP@5	mAP@10
CIReVL [ 12 ]	BLIP-2 [ 16 ] + GPT-3.5-Turbo [ 5 ]	26.77	27.59
OSrCIR [ 28 ]	GPT-4o [ 1 ]	30.47	31.14
SEIZE [ 33 ]	BLIP-2 [ 16 ] + GPT-3.5-Turbo [ 5 ]	32.46	33.77
CoTMR [ 27 ]	Qwen2-VL-72B [ 32 ]	32.23	32.72
DCR (Ours)	Qwen2-VL-72B [ 32 ]	34.38	36.11
DCR (Ours)	Qwen2.5-VL-32B [ 3 ]	34.77	35.95

Table C : Comparison of reasoning backbones on the CIRCO test set with CLIP ViT-G/14.

Backbone	Fusion Weights	CIRCO	CIRR	Fashion-IQ
Backbone	Fusion Weights	mAP@5	R@5	R mean
ViT-B/32	Equal Weights	25.02	66.77	50.81
ViT-B/32	Dataset-specific	25.07	67.28	51.33
ViT-L/14	Equal Weights	31.60	69.25	50.46
ViT-L/14	Dataset-specific	31.80	69.95	51.52

Table D : Sensitivity analysis of fusion weights. Comparison of equal and dataset-specific fusion weights on CIRCO, CIRR, and Fashion-IQ using CLIP ViT-B/32 and ViT-L/14.

	CIRCO (mAP@5)	CIRR (R@5)	Fashion-IQ (R mean )
Performance $\uparrow$	35.35 / 34.63	72.61 / 70.80	55.71 / 58.30
Modality Gap $\downarrow$	0.5524 / 0.0084	0.5366 / 0.0133	0.5990 / 0.0195

Table E : Analysis of modality gap reduction. Comparison of retrieval performance and modality gap with and without I0T post on the validation splits of three benchmarks. Values are reported as DCR-TSF / I0T post .

Panel of Figure 19 — Figure C : Comparison of DCR with caption ensemble baselines on CIRCO and Fashion-IQ.

Panel of Figure 20 — Figure D : Qualitative comparison on Fashion-IQ with ViT-G/14. Each example shows a reference image, its modification text, and the retrieved images with their ranks under four settings: Fine-Grained Description , Coarse-Grained Description , Composed-query Anchor , and DCR-TSF . The ground-truth image is highlighted in green.

Panel of Figure 21 — Figure E : Qualitative comparison on CIRR with ViT-G/14. Each example shows a reference image, its modification text, and the retrieved images with their ranks under four settings: Fine-Grained Description , Coarse-Grained Description , Composed-query Anchor , and DCR-TSF . The ground-truth image is highlighted in green.

Panel of Figure 22 — Figure G : Failure cases caused by negation on CIRR with ViT-G/14. CLIP assigns high similarity to images containing objects or interactions explicitly excluded by negated modification texts (e.g., “no books” or “no human interacting”). Since TSF directly incorporates the modification text, this limitation propagates to retrieval and can rank such images above the ground-truth target.

Composed Image Retrieval

Composed image retrieval finds images that match a reference image together with a text instruction describing the desired change.

See all 65 papers

DiCE-CIR: Direct Composition Learning for Efficient Zero-Shot Composed Image Retrieval
Jul 6, 2026Gwang-Ho Na, Ho-Joong Kim, Seong-Whan LeeCross-Modal Alignment Compositional Representation Learning
Department of Artificial Intelligence, Korea University, Anam-dong, Seongbuk-ku, Seoul 02841, Korea
WISER: Wider Search, Deeper Thinking, and Adaptive Fusion for Training-Free Zero-Shot Composed Image Retrieval
Feb 26, 2026Tianyue Wang, Leigang Qu, Tianyu Yang +4Composed Image Retrieval Multimodal Fusion
Foundation Model Research Center, Institute of Automation, Chinese Academy of Sciences · School of Advanced Interdisciplinary Sciences, University of Chinese Academy of Sciences · National University of Singapore +3
G-MIXER: Geodesic Mixup-based Implicit Semantic Expansion and Explicit Semantic Re-ranking for Zero-Shot Composed Image Retrieval
Apr 16, 2026Jiyoung Lim, Heejae Yang, Jee-Hyong LeeZero-Shot Learning Composed Image Retrieval
Sungkyunkwan University

Jul 6, 2026J/K move · Enter open · S save

DiCE-CIR: Direct Composition Learning for Efficient Zero-Shot Composed Image Retrieval

Gwang-Ho Na, Ho-Joong Kim, Seong-Whan Lee

Department of Artificial Intelligence, Korea University, Anam-dong, Seongbuk-ku, Seoul 02841, Korea

Cross-Modal Alignment Compositional Representation Learning Composed Image Retrieval

Open paper AskPDF arXiv

Zero-shot composed image retrieval (ZS-CIR) aims to retrieve a target image from a multimodal query consisting of a reference image and an edit text describing the desired modification. Recent ZS-CIR studies have relied on projection-based methods that map a reference image into pseudo-word tokens in the text embedding space. However, such methods require additional projection and re-encoding steps, increasing training complexity, reducing efficiency, and introducing a discrepancy between training and inference. In this paper, we propose DiCE-CIR, a direct composition learning method that predicts composed query representations by directly composing a reference image and an edit text. To enable scalable training without manually annotated triplets, we automatically construct compositional training samples from large-scale image-caption pairs using a large language model. Based on these samples, we train a lightweight composition module with objectives that promote alignment with the target, edit-consistent semantic transformation, and retrieval discriminability. We conduct extensive experiments on ZS-CIR benchmarks and show that DiCE-CIR achieves state-of-the-art performance on CIRCO and competitive performance on CIRR while maintaining high computational efficiency.

Abstract

Composed Image Retrieval

Explore similar work

DiCE-CIR: Direct Composition Learning for Efficient Zero-Shot Composed Image Retrieval

WISER: Wider Search, Deeper Thinking, and Adaptive Fusion for Training-Free Zero-Shot Composed Image Retrieval

G-MIXER: Geodesic Mixup-based Implicit Semantic Expansion and Explicit Semantic Re-ranking for Zero-Shot Composed Image Retrieval