When Generated Images Look Right and Retrieve Wrong: Coverage-Guided Cross-Scale Re-Indexing for Knowledge-Faithful Generative Perception

• URL: http://arxiv.org/abs/2608.20810v1
• Date: Fri, 21 Aug 2026 07:28:01 GMT
• Title: When Generated Images Look Right and Retrieve Wrong: Coverage-Guided Cross-Scale Re-Indexing for Knowledge-Faithful Generative Perception
• Abstract summary: We propose CERES, a closed-loop multimodal indexing framework that builds a three-level semantic pyramid.<n>It mines implicit concepts via a co-language-aware router, performs scale-routed cross-attention into a lightweight U-Net generator, and verifies coverage by re-indexing the generated image with the same frozen VLM.<n>On four pansharpening benchmarks across seven settings, CERES delivers the new state of the art with the largest gains where scale variation is most extreme.
• Score: 12.968836079025344
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Multimodal information systems increasingly route generated visual content back through the same vision-language index that informed its production, so the output must remain retrievable by the queries it was meant to serve. When the scene contains entities at vastly different scales, existing language-guided generators condition on a single, globally pooled text embedding and quietly drop scale-specific concepts, breaking concept-query retrieval even when pixel fidelity is high. We formalise this failure as semantic collapse and propose CERES, a closed-loop multimodal indexing framework that builds a three-level semantic pyramid, mines implicit concepts via a co-occurrence-aware router, performs scale-routed cross-attention into a lightweight U-Net generator, and verifies coverage by re-indexing the generated image with the same frozen VLM. A continuously differentiable soft-Jaccard coverage objective returns dense gradients to the 0.39M-parameter generator under explicit non-degeneracy conditions, and coverage is verified by an independent DINOv2 linear probe trained only on external scene and object labels. On four pansharpening benchmarks across seven settings, CERES delivers the new state of the art with the largest gains where scale variation is most extreme. It also improves concept-query retrieval Recall@5 by +14.0 points and image-text mean reciprocal rank by 0.19 over the strongest baseline, showing that the closed loop preserves queryable content rather than self-referential feature consistency.

Related papers

• Retrieved Images as Visual Thought: Training-Free Multimodal In-Context Learning for the Open-vs-Closed Gap [6.097306160389619]
  ReVisIT is a train-free framework that treats each retrieved image-label pair as a unit of visual thought.<n>On VL-ICL Bench Fast Open MiniImageNet, Qwen3-VL-30B-A3B with ReVisIT reaches 98.5% at 4-shot, statistically indistinguishable from the 72B LLaVA-VisionOne SOTA (98.7%) on this near-saturated task at about 1/2.4 the parameters.
  arXiv  Detail & Related papers  (2026-07-01T08:31:09Z)
• OneRetrieval: Unifying Multi-Branch E-commerce Retrieval with an Editable Generative Model [34.47411112516513]
  Industrial e-commerce search serves hundreds of millions of items through a multi-branch retrieval stage fused by hand-tuned merging.<n>OneRetrieval matches the deep recall of the strongest generative baseline, with an intervention hit rate over an order of magnitude above closed-codebook encodings.<n>The system is deployed at Kuaishou, serving hundreds of millions of PVs daily.
  arXiv  Detail & Related papers  (2026-06-11T16:21:13Z)
• Advancing Narrative Long Video Generation via Training-Free Identity-Aware Memory [79.01059178883817]
  IAMFlow is a training-free identity-aware memory framework that explicitly models and tracks persistent entity identities.<n>A VLM asynchronously verifies and refines attributes from rendered frames, enabling explicit entity tracking in place of implicit similarity-based matching.<n>NarraStream-Bench is a benchmark for narrative streaming video generation that features 324 multi-prompt scripts spanning six dimensions.
  arXiv  Detail & Related papers  (2026-05-18T17:54:34Z)
• CMAG: Concept-Scaffolded Retrieval for Marketplace Avatar Generation [0.5131152350448099]
  Metaverse platforms rely on creator-driven marketplaces where avatars are assembled from discrete, taxonomy-labeled 3D assets.<n>We propose textbfCMAG, a concept-scaffolded retrieval and verified composition framework for marketplace generation.
  arXiv  Detail & Related papers  (2026-05-18T17:21:43Z)
• TIGER-FG: Text-Guided Implicit Fine-Grained Grounding for E-commerce Retrieval [10.927254533551563]
  E-commerce image search often takes a cropped image as the query, while each candidate is represented by full item images and structured text.<n>We propose TIGER-FG, a text-guided implicit fine-grained grounding framework for e-commerce retrieval.
  arXiv  Detail & Related papers  (2026-05-18T14:07:20Z)
• UniCustom: Unified Visual Conditioning for Multi-Reference Image Generation [65.53694602893042]
  We propose a unified visual conditioning framework that fuses ViT and VAE features before VLM encoding.<n>Experiments on two multi-reference generation benchmarks demonstrate that UniCustom consistently improves subject consistency, instruction following, and compositional fidelity.
  arXiv  Detail & Related papers  (2026-05-12T13:10:05Z)
• LFRAG: Layout-oriented Fine-grained Retrieval-Augmented Generation on Multimodal Document Understanding [23.227000200423458]
  We propose layout-oriented Fine-grained Retrieval-Augmented Generation (LFRAG), a novel framework that advances multimodal RAG from page-level to block-level retrieval.<n>With block-level late retrieval, LFRAG enables precise query-content alignment and reduces irrelevant content for downstream generation.<n>Experiments on LFDocQA demonstrate that LFRAG achieves state-of-the-art performance on retrieval tasks, outperforms the best baseline by 7.20% in answer accuracy, and reduces token consumption by 73.07% in generation tasks.
  arXiv  Detail & Related papers  (2026-04-18T05:04:49Z)
• Cheers: Decoupling Patch Details from Semantic Representations Enables Unified Multimodal Comprehension and Generation [66.53544128707817]
  Cheers is a unified multimodal model that decouples patch-level details from semantic representations.<n> Cheers matches or surpasses advanced UMMs in both visual understanding and generation.
  arXiv  Detail & Related papers  (2026-03-13T08:55:27Z)
• From Feature Interaction to Feature Generation: A Generative Paradigm of CTR Prediction Models [81.43473418572567]
  Click-Through Rate (CTR) prediction is a core task in recommendation systems.<n>We propose a novel generative framework to address embedding dimensional collapse and information redundancy.<n>We show that SFG consistently mitigates embedding collapse and reduces information redundancy, while yielding substantial performance gains.
  arXiv  Detail & Related papers  (2025-12-16T03:17:18Z)
• Vision-Free Retrieval: Rethinking Multimodal Search with Textual Scene Descriptions [81.33113485830711]
  We introduce a vision-free, single-encoder retrieval pipeline for vision-language models.<n>We migrate to a text-to-text paradigm with the assistance of VLLM-generated structured image descriptions.<n>Our approach achieves state-of-the-art zero-shot performance on multiple retrieval and compositionality benchmarks.
  arXiv  Detail & Related papers  (2025-09-23T16:22:27Z)
• Collaborative Group: Composed Image Retrieval via Consensus Learning from Noisy Annotations [67.92679668612858]
  We propose the Consensus Network (Css-Net), inspired by the psychological concept that groups outperform individuals. Css-Net comprises two core components: (1) a consensus module with four diverse compositors, each generating distinct image-text embeddings; and (2) a Kullback-Leibler divergence loss that encourages learning of inter-compositor interactions. On benchmark datasets, particularly FashionIQ, Css-Net demonstrates marked improvements. Notably, it achieves significant recall gains, with a 2.77% increase in R@10 and 6.67% boost in R@50, underscoring its
  arXiv  Detail & Related papers  (2023-06-03T11:50:44Z)

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.