DeliCIR: Deliberative Test-Time Evolutionary Hierarchical Multi-Agents for Composed Image Retrieval

• URL: http://arxiv.org/abs/2605.22478v2
• Date: Sat, 23 May 2026 05:02:33 GMT
• Title: DeliCIR: Deliberative Test-Time Evolutionary Hierarchical Multi-Agents for Composed Image Retrieval
• Abstract summary: Composed Image Retrieval (CIR) requires both preserving the visual continuity of the reference image and faithfully executing the semantic variables specified in the modification text.<n>We propose a one-stop hierarchical Perception-to-Deliberation Framework (PDF), which is the first to introduce experience self-evolution and Test-Time Scaling Laws (TTS) into CIR.<n>This study indicates that experience-driven self-evolution and TTS represent a highly promising and scalable path for achieving zero-shot fine-grained multimedia retrieval.
• Score: 30.843984141042828
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Composed Image Retrieval (CIR) requires both preserving the visual continuity of the reference image and faithfully executing the semantic variables specified in the modification text, which constitute the core challenge of the task. Existing methods often suffer from Perception Myopia in a single space, or fall into Logic Drift in iterative collaboration due to the perception ceiling of the underlying retriever. To address this issue, we propose a one-stop hierarchical Perception-to-Deliberation Framework (PDF), which, to the best of our knowledge, is the first to introduce experience self-evolution and Test-Time Scaling Laws (TTS) into CIR. Relying on a hierarchical multi-agent architecture, PDF first utilizes an Intent Routing Manager to dynamically dispatch multi-view Worker perception signals based on modification intents to construct a high-recall candidate pool. Subsequently, the Decision Manager combines a Training-free Reasoning Policy Distillation mechanism with a Tournament-style TTS (T-TTS) strategy to achieve self-evolving fine-grained reasoning, yielding the final retrieval results. Experimental results demonstrate that PDF achieves SOTA performance on three benchmark datasets: CIRR, CIRCO, and FashionIQ. This study indicates that experience-driven self-evolution and TTS represent a highly promising and scalable path for achieving zero-shot fine-grained multimedia retrieval. The code will be made publicly available upon acceptance.

Related papers

• Towards Vision-Free CIR: Attribute-Augmented Scoring and LLM-Based Reranking for Zero-Shot Composed Image Retrieval [17.703391886251378]
  "Vision-Free'' approaches can be effective for standard image retrieval tasks.<n>But it remains unclear whether this paradigm can effectively handle a more complex, multimodal task, Composed Image Retrieval.<n>We introduce a Vision-Free CIR framework that addresses this challenge through two key techniques.
  arXiv  Detail & Related papers  (2026-07-14T10:58:12Z)
• DiCE-CIR: Direct Composition Learning for Efficient Zero-Shot Composed Image Retrieval [52.15696141238419]
  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.<n>We propose DiCE-CIR, a direct composition learning method that predicts composed query representations by directly composing a reference image and an edit text.
  arXiv  Detail & Related papers  (2026-07-06T04:37:30Z)
• PDCR: Perception-Decomposed Confidence Reward for Vision-Language Reasoning [80.94559742826083]
  Reinforcement Learning with Verifiable Rewards (RLVR) traditionally relies on a sparse, outcome-based signal.<n>Recent work shows that providing a fine-grained, model-intrinsic signal effectively improves language reasoning training by providing step-level guidance without costly external models.<n>While effective for unimodal text, we find that naively applying this global reward to vision-language (V-L) reasoning is a suboptimal strategy.<n>We propose Perception-Decomposed Confidence Reward (PDCR), a framework that solves this by aligning the reward structure with the task's heterogeneous nature.
  arXiv  Detail & Related papers  (2026-05-13T12:55:18Z)
• ReCALL: Recalibrating Capability Degradation for MLLM-based Composed Image Retrieval [64.14282916266998]
  Composed Image Retrieval aims to retrieve target images based on a hybrid query comprising a reference image and a modification text.<n>We propose ReCALL, a model-agnostic framework that follows a diagnose-generate-refine pipeline.<n>Experiments on CIRR and FashionIQ show that ReCALL consistently recalibrates degraded capabilities and achieves state-of-the-art performance.
  arXiv  Detail & Related papers  (2026-02-02T04:52:54Z)
• Integrating Diverse Assignment Strategies into DETRs [61.61489761918158]
  Label assignment is a critical component in object detectors, particularly within DETR-style frameworks.<n>We propose LoRA-DETR, a flexible and lightweight framework that seamlessly integrates diverse assignment strategies into any DETR-style detector.
  arXiv  Detail & Related papers  (2026-01-14T07:28:54Z)
• CIR-CoT: Towards Interpretable Composed Image Retrieval via End-to-End Chain-of-Thought Reasoning [93.05917922306196]
  Composed Image Retrieval (CIR) aims to find a target image from a reference image and a modification text.<n>CIR-CoT is the first end-to-end retrieval-oriented MLLM designed to integrate explicit Chain-of-Thought (CoT) reasoning.
  arXiv  Detail & Related papers  (2025-10-09T09:41:45Z)
• SQUARE: Semantic Query-Augmented Fusion and Efficient Batch Reranking for Training-free Zero-Shot Composed Image Retrieval [2.624097337766623]
  Composed Image Retrieval (CIR) aims to retrieve target images that preserve the visual content of a reference image while incorporating user-specified textual modifications.<n>We present a novel two-stage training-free framework that leverages Multimodal Large Language Models (MLLMs) to enhance ZS-CIR.
  arXiv  Detail & Related papers  (2025-09-30T14:41:24Z)
• Enhancing Supervised Composed Image Retrieval via Reasoning-Augmented Representation Engineering [15.191542305783702]
  We propose a framework that includes the Pyramid Matching Model with Training-Free Refinement.<n>Inspired by representation engineering, we extracted representations from COT data and injected them into the LVLMs.<n>Experiments on CIR benchmarks demonstrate that PMTFR surpasses state-of-the-art methods in supervised CIR tasks.
  arXiv  Detail & Related papers  (2025-08-15T07:10:10Z)
• ImageScope: Unifying Language-Guided Image Retrieval via Large Multimodal Model Collective Reasoning [62.61187785810336]
  ImageScope is a training-free, three-stage framework that unifies language-guided image retrieval tasks.<n>In the first stage, we improve the robustness of the framework by synthesizing search intents across varying levels of semantic granularity.<n>In the second and third stages, we reflect on retrieval results by verifying predicate propositions locally, and performing pairwise evaluations globally.
  arXiv  Detail & Related papers  (2025-03-13T08:43:24Z)
• CoTMR: Chain-of-Thought Multi-Scale Reasoning for Training-Free Zero-Shot Composed Image Retrieval [13.59418209417664]
  Zero-Shot Composed Image Retrieval (ZS-CIR) aims to retrieve target images by integrating information from a composed query without training samples.<n>We propose CoTMR, a training-free framework crafted for ZS-CIR with novel Chain-of-thought (CoT) and Multi-scale Reasoning.
  arXiv  Detail & Related papers  (2025-02-28T08:12:23Z)
• Pseudo-triplet Guided Few-shot Composed Image Retrieval [20.040511832864503]
  Composed Image Retrieval (CIR) is a challenging task that aims to retrieve the target image with a multimodal query. We propose a novel two-stage pseudo triplet guided few-shot CIR scheme, dubbed PTG-FSCIR. In the first stage, we propose an attentive masking and captioning-based pseudo triplet generation method, to construct pseudo triplets from pure image data. In the second stage, we propose a challenging triplet-based CIR fine-tuning method, where we design a pseudo modification text-based sample challenging score estimation strategy.
  arXiv  Detail & Related papers  (2024-07-08T14:53:07Z)
• ACTRESS: Active Retraining for Semi-supervised Visual Grounding [52.08834188447851]
  A previous study, RefTeacher, makes the first attempt to tackle this task by adopting the teacher-student framework to provide pseudo confidence supervision and attention-based supervision. This approach is incompatible with current state-of-the-art visual grounding models, which follow the Transformer-based pipeline. Our paper proposes the ACTive REtraining approach for Semi-Supervised Visual Grounding, abbreviated as ACTRESS.
  arXiv  Detail & Related papers  (2024-07-03T16:33:31Z)
• Vision-by-Language for Training-Free Compositional Image Retrieval [78.60509831598745]
  Compositional Image Retrieval (CIR) aims to retrieve the relevant target image in a database. Recent research sidesteps this need by using large-scale vision-language models (VLMs) We propose to tackle CIR in a training-free manner via Vision-by-Language (CIReVL)
  arXiv  Detail & Related papers  (2023-10-13T17:59:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.