# RVSD: Retrieval Vision Sparse Decoding for Mitigating Visual Hallucinations in Large Vision-Language Models

arXiv: https://arxiv.org/abs/2609.02731 · HTML: https://mlsanity.com/paper/2609.02731/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Hallucination mitigation in LVLMs, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 10 of its 23 references that are in the map.
- First posted on arXiv 2 Sep 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Hallucination mitigation in LVLMs](https://mlsanity.com/mllm/hallucination-mitigation-in-lvlms/)

## Builds on (most cited first, 10 of 10)

- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [MM-Vet: Evaluating Large Multimodal Models for Integrated Capabilities](https://mlsanity.com/paper/2308.02490/) (2023)
- [Mitigating Object Hallucinations in Large Vision-Language Models through Visual Contrastive Decoding](https://mlsanity.com/paper/2311.16922/) (2023)
- [Multi-Modal Hallucination Control by Visual Information Grounding](https://mlsanity.com/paper/2403.14003/) (2024)
- [VASparse: Towards Efficient Visual Hallucination Mitigation via Visual-Aware Token Sparsification](https://mlsanity.com/paper/2501.06553/) (2025)
- [Hallucinatory Image Tokens: A Training-free EAZY Approach on Detecting and Mitigating Object Hallucinations in LVLMs](https://mlsanity.com/paper/2503.07772/) (2025)
- [MINT: Mitigating Hallucinations in Large Vision-Language Models via Token Reduction](https://mlsanity.com/paper/2502.00717/) (2025)
- [PruneHal: Reducing Hallucinations in Multi-modal Large Language Models through Adaptive KV Cache Pruning](https://mlsanity.com/paper/2510.19183/) (2025)
- [Locate-then-Sparsify: Attribution Guided Sparse Strategy for Visual Hallucination Mitigation](https://mlsanity.com/paper/2603.16284/) (2026)

## Abstract (arXiv)

Large vision-language models have achieved remarkable success in vision-language tasks. However, they remain prone to Visual Hallucinations (VHs), undermining their reliability in real-world applications. Existing solutions typically require curated datasets, additional training, or multi-round decoding, resulting in considerable computational overhead. In this paper, we propose RVSD (underlineRetrieval underlineVision underlineSparse underlineDecoding), a training-free and plug-and-play decoding framework that, for the first time, unifies token sparsification and Semantic-Space Visual Retrieval (SSVR) within a single decoding pass. Within RVSD, we introduce a semantics-directed token selection strategy that selectively sparsifies redundant tokens while preserving critical visual information. We further propose the SSVR mechanism, which reformulates visual compensation as an on-demand cross-modal retrieval process within a shared semantic space. Extensive experiments demonstrate that RVSD achieves state-of-the-art performance in mitigating VHs while maintaining robust suppression capabilities under long-context generation settings. Our code is available here.footnotehttps://github.com/canjie-liu/RVSD