MINER: Mining Multimodal Internal Representation for Efficient Retrieval

• URL: http://arxiv.org/abs/2605.06460v1
• Date: Thu, 07 May 2026 15:51:06 GMT
• Title: MINER: Mining Multimodal Internal Representation for Efficient Retrieval
• Abstract summary: MINER (Mining Multimodal Internal RepreseNtation for Efficient Retrieval) is a lightweight plug-in module that probes and fuses internal signals across transformer layers into a single compact embedding.<n>Across ViDoRe V1/V2/V3, MINER outperforms existing dense single-vector retrievers on the majority of benchmarks, with up to 4.5% nDCG@5 improvement over its corresponding backbone.
• Score: 30.225207646393542
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Visual document retrieval has become essential for accessing information in visually rich documents. Existing approaches fall into two camps. Late-interaction retrievers achieve strong quality through fine-grained token-level matching but store hundreds of vectors per page, incurring large index footprints and high serving costs. By contrast, dense single-vector retrievers retain storage and latency advantages but consistently lag in quality because they compress all information into a single final-layer embedding. In this work, we first conduct a layerwise diagnostic on single-vector retrievers, revealing that retrieval-relevant signal resides in internal representations. Motivated by these findings, we propose MINER (Mining Multimodal Internal RepreseNtation for Efficient Retrieval), a lightweight plug-in module that probes and fuses internal signals across transformer layers into a single compact embedding without modifying the backbone or sacrificing single-vector efficiency. The first Retrieval-Aligned Layer Probing stage attaches a lightweight probe at each layer, surfacing which dimensions carry retrieval-relevant information. The subsequent Adaptive Sparse Multi-Layer Fusion stage applies performance-adaptive neuron-level masking to the selected layers and fuses the surviving signals into the final dense vector. Across ViDoRe V1/V2/V3, MINER outperforms existing dense single-vector retrievers on the majority of benchmarks, with up to 4.5% nDCG@5 improvement over its corresponding backbone. Compared to strong late-interaction baselines, in some settings MINER substantially narrows the nDCG@$5$ gap to $0.2$ while preserving the storage and serving advantages of dense retrieval.

Related papers

• PUMA: Post-Hoc Sparsification of Universal Multimodal Embeddings for Efficient Retrieval [50.986807025432476]
  We introduce PUMA, a sparse autoencoder recipe that maps universal multimodal embeddings to compact sparse codes without retraining the backbone.<n>We evaluate on five benchmarks covering text-to-image and composed image retrieval.<n>Puma reduces vector storage by 8-16x (FP32) and is up to 25x faster than exact dense scoring on larger candidate pools.
  arXiv  Detail & Related papers  (2026-08-26T13:26:39Z)
• No More K-means: Single-Stage Sparse Coding for Efficient Multi-Vector Retrieval [51.43543998583709]
  Single-stage Sparse Retrieval (SSR) is a paradigm shift that replaces expensive clustering with efficient sparse coding.<n>It reduces indexing time by 15x compared to ColBERTv2, halves retrieval latency, and simultaneously improves retrieval performance.
  arXiv  Detail & Related papers  (2026-05-28T15:53:34Z)
• Multi-Vector Index Compression in Any Modality [73.7330345057813]
  Late interaction has emerged as a dominant paradigm for information retrieval in text, images, visual documents, and videos.<n>We introduce four approaches for index compression: sequence resizing, memory tokens, hierarchical pooling, and a novel attention-guided clustering (AGC)<n>AGC uses an attention-guided mechanism to identify the most semantically salient regions of a document as cluster centroids and to weight token aggregation.
  arXiv  Detail & Related papers  (2026-02-24T18:57:33Z)
• Sculpting the Vector Space: Towards Efficient Multi-Vector Visual Document Retrieval via Prune-then-Merge Framework [39.59931739606983]
  Visual Document Retrieval (VDR) aims to retrieve relevant pages within vast corpora of visually-rich documents.<n>Current efficiency methods like pruning and merging address imperfectly, creating a difficult trade-off between compression rate and feature fidelity.<n>We introduce Prune-then-Merge, a novel two-stage framework that synergizes these complementary approaches.
  arXiv  Detail & Related papers  (2026-02-23T06:45:19Z)
• Multivector Reranking in the Era of Strong First-Stage Retrievers [11.098422338598454]
  We reproduce several state-of-the-art multivector retrieval methods on two publicly available datasets.<n>We show that replacing the token-level gather phase with a single-vector document retriever produces a smaller and more semantically coherent candidate set.<n>Our two-stage approach achieves over $24times$ speedup over the state-of-the-art multivector retrieval systems, while maintaining comparable or superior retrieval quality.
  arXiv  Detail & Related papers  (2026-01-08T18:22:18Z)
• URaG: Unified Retrieval and Generation in Multimodal LLMs for Efficient Long Document Understanding [55.45331924836242]
  We present URaG, a framework that Unifies Retrieval and Generation within a single MLLM.<n>We show that URaG achieves state-of-the-art performance while reducing computational overhead by 44-56%.
  arXiv  Detail & Related papers  (2025-11-13T17:54:09Z)
• MemSearcher: Training LLMs to Reason, Search and Manage Memory via End-to-End Reinforcement Learning [73.27233666920618]
  We propose MemSearcher, an agent workflow that iteratively maintains a compact memory and combines the current turn with it.<n>At each turn, MemSearcher fuses the user's question with the memory to generate reasoning traces, perform search actions, and update memory to retain only information essential for solving the task.<n>We introduce multi-context GRPO, an end-to-end RL framework that jointly optimize reasoning, search strategies, and memory management of MemSearcher Agents.
  arXiv  Detail & Related papers  (2025-11-04T18:27:39Z)
• Efficient Constant-Space Multi-Vector Retrieval [25.834026445124874]
  We propose encoding documents to a fixed number of vectors, which are no longer necessarily tied to the input tokens.<n>We find that passages can be effectively encoded into a fixed number of vectors while retaining most of the original effectiveness.
  arXiv  Detail & Related papers  (2025-04-02T15:22:23Z)
• MUVERA: Multi-Vector Retrieval via Fixed Dimensional Encodings [15.275864151890511]
  We introduce MUVERA (MUlti-VEctor Retrieval Algorithm), a retrieval mechanism which reduces multi-vector search to single-vector similarity search. MUVERA achieves consistently good end-to-end recall and latency across a diverse set of the BEIR retrieval datasets.
  arXiv  Detail & Related papers  (2024-05-29T20:40:20Z)
• Improving Out-of-Distribution Generalization of Neural Rerankers with Contextualized Late Interaction [52.63663547523033]
  Late interaction, the simplest form of multi-vector, is also helpful to neural rerankers that only use the [] vector to compute the similarity score. We show that the finding is consistent across different model sizes and first-stage retrievers of diverse natures.
  arXiv  Detail & Related papers  (2023-02-13T18:42:17Z)
• CITADEL: Conditional Token Interaction via Dynamic Lexical Routing for Efficient and Effective Multi-Vector Retrieval [72.90850213615427]
  Multi-vector retrieval methods combine the merits of sparse (e.g. BM25) and dense (e.g. DPR) retrievers. These methods are orders of magnitude slower and need much more space to store their indices compared to their single-vector counterparts. We propose conditional token interaction via dynamic lexical routing, namely CITADEL, for efficient and effective multi-vector retrieval.
  arXiv  Detail & Related papers  (2022-11-18T18:27:35Z)
• Multi-scale Interactive Network for Salient Object Detection [91.43066633305662]
  We propose the aggregate interaction modules to integrate the features from adjacent levels. To obtain more efficient multi-scale features, the self-interaction modules are embedded in each decoder unit. Experimental results on five benchmark datasets demonstrate that the proposed method without any post-processing performs favorably against 23 state-of-the-art approaches.
  arXiv  Detail & Related papers  (2020-07-17T15:41:37Z)

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.