# Robust Grounding with MLLMs Against Occlusion and Small Objects via Language-Guided Semantic Cues

arXiv: https://arxiv.org/abs/2604.24036 · HTML: https://mlsanity.com/paper/2604.24036/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in VLM benchmarks and visual reasoning, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 26 of its 34 references that are in the map.
- First posted on arXiv 27 Apr 2026.

## Placed in

- Multimodal LLMs / vision-language models / [VLM benchmarks and visual reasoning](https://mlsanity.com/mllm/vlm-benchmarks-and-visual-reasoning/)

## Builds on (most cited first, 26 of 26)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Mask R-CNN](https://mlsanity.com/paper/1703.06870/) (2017)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [Gaussian Error Linear Units (GELUs)](https://mlsanity.com/paper/1606.08415/) (2016)
- [CIDEr: Consensus-based Image Description Evaluation](https://mlsanity.com/paper/1411.5726/) (2014)
- [InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models](https://mlsanity.com/paper/2504.10479/) (2025)
- [Kosmos-2: Grounding Multimodal Large Language Models to the World](https://mlsanity.com/paper/2306.14824/) (2023)
- [Modeling Context in Referring Expressions](https://mlsanity.com/paper/1608.00272/) (2016)
- [Generation and Comprehension of Unambiguous Object Descriptions](https://mlsanity.com/paper/1511.02283/) (2015)
- [DETRs Beat YOLOs on Real-time Object Detection](https://mlsanity.com/paper/2304.08069/) (2023)
- [Ferret: Refer and Ground Anything Anywhere at Any Granularity](https://mlsanity.com/paper/2310.07704/) (2023)
- [VisionLLM: Large Language Model is also an Open-Ended Decoder for Vision-Centric Tasks](https://mlsanity.com/paper/2305.11175/) (2023)
- [CrowdHuman: A Benchmark for Detecting Human in a Crowd](https://mlsanity.com/paper/1805.00123/) (2018)
- [The Unmanned Aerial Vehicle Benchmark: Object Detection and Tracking](https://mlsanity.com/paper/1804.00518/) (2018)
- [CityPersons: A Diverse Dataset for Pedestrian Detection](https://mlsanity.com/paper/1702.05693/) (2017)
- [Groma: Localized Visual Tokenization for Grounding Multimodal Large Language Models](https://mlsanity.com/paper/2404.13013/) (2024)
- [T-Rex2: Towards Generic Object Detection via Text-Visual Prompt Synergy](https://mlsanity.com/paper/2403.14610/) (2024)
- [WiderPerson: A Diverse Dataset for Dense Pedestrian Detection in the Wild](https://mlsanity.com/paper/1909.12118/) (2019)
- [MoAI: Mixture of All Intelligence for Large Language and Vision Models](https://mlsanity.com/paper/2403.07508/) (2024)
- [ChatRex: Taming Multimodal LLM for Joint Perception and Understanding](https://mlsanity.com/paper/2411.18363/) (2024)
- [RefDrone: A Challenging Benchmark for Referring Expression Comprehension in Drone Scenes](https://mlsanity.com/paper/2502.00392/) (2025)
- [HazyDet: Open-Source Benchmark for Drone-View Object Detection with Depth-Cues in Hazy Scenes](https://mlsanity.com/paper/2409.19833/) (2024)
- [Language-guided Learning for Object Detection Tackling Multiple Variations in Aerial Images](https://mlsanity.com/paper/2505.23193/) (2025)

## Abstract (arXiv)

While Multimodal Large Language Models (MLLMs) have enhanced grounding capabilities in general scenes, their robustness in crowded scenes remains underexplored. Crowded scenes entail visual challenges (i.e., occlusion and small objects), which impair object semantics and degrade grounding performance. In contrast, language expressions are immune to such degradation and preserve object semantics. In light of these observations, we propose a novel method that overcomes such constraints by leveraging Language-Guided Semantic Cues (LGSCs). Specifically, our approach introduces a Semantic Cue Extractor (SCE) to derive semantic cues of objects from the visual pipeline of an MLLM. We then guide these cues using corresponding text embeddings to produce LGSCs as linguistic semantic priors. Subsequently, they are reintegrated into the original visual pipeline to refine object semantics. Extensive experiments and analyses demonstrate that incorporating LGSCs into an MLLM effectively improves grounding accuracy in crowded scenes.