# Beyond Referring Expressions: Scenario Comprehension Visual Grounding

arXiv: https://arxiv.org/abs/2604.02323 · HTML: https://mlsanity.com/paper/2604.02323/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 28 of its 39 references that are in the map.
- First posted on arXiv 2 Apr 2026.

## Builds on (most cited first, 28 of 28)

- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection](https://mlsanity.com/paper/2303.05499/) (2023)
- [Self-Instruct: Aligning Language Models with Self-Generated Instructions](https://mlsanity.com/paper/2212.10560/) (2022)
- [Flickr30k Entities: Collecting Region-to-Phrase Correspondences for Richer Image-to-Sentence Models](https://mlsanity.com/paper/1505.04870/) (2015)
- [LVIS: A Dataset for Large Vocabulary Instance Segmentation](https://mlsanity.com/paper/1908.03195/) (2019)
- [WizardLM: Empowering large pre-trained language models to follow complex instructions](https://mlsanity.com/paper/2304.12244/) (2023)
- [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)
- [How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suites](https://mlsanity.com/paper/2404.16821/) (2024)
- [Shikra: Unleashing Multimodal LLM's Referential Dialogue Magic](https://mlsanity.com/paper/2306.15195/) (2023)
- [Generation and Comprehension of Unambiguous Object Descriptions](https://mlsanity.com/paper/1511.02283/) (2015)
- [MDETR -- Modulated Detection for End-to-End Multi-Modal Understanding](https://mlsanity.com/paper/2104.12763/) (2021)
- [Ferret: Refer and Ground Anything Anywhere at Any Granularity](https://mlsanity.com/paper/2310.07704/) (2023)
- [TransVG: End-to-End Visual Grounding with Transformers](https://mlsanity.com/paper/2104.08541/) (2021)
- [Ferret-UI: Grounded Mobile UI Understanding with Multimodal LLMs](https://mlsanity.com/paper/2404.05719/) (2024)
- [iGibson 1.0: a Simulation Environment for Interactive Tasks in Large Realistic Scenes](https://mlsanity.com/paper/2012.02924/) (2020)
- [PhraseCut: Language-based Image Segmentation in the Wild](https://mlsanity.com/paper/2008.01187/) (2020)
- [GroundingGPT:Language Enhanced Multi-modal Grounding Model](https://mlsanity.com/paper/2401.06071/) (2024)
- [UniVG-R1: Reasoning Guided Universal Visual Grounding with Reinforcement Learning](https://mlsanity.com/paper/2505.14231/) (2025)
- [CLEVR-Ref+: Diagnosing Visual Reasoning with Referring Expressions](https://mlsanity.com/paper/1901.00850/) (2019)
- [Graph-Structured Referring Expression Reasoning in The Wild](https://mlsanity.com/paper/2004.08814/) (2020)
- [RoboVerse: Towards a Unified Platform, Dataset and Benchmark for Scalable and Generalizable Robot Learning](https://mlsanity.com/paper/2504.18904/) (2025)
- [Cops-Ref: A new Dataset and Task on Compositional Referring Expression Comprehension](https://mlsanity.com/paper/2003.00403/) (2020)
- [Griffon v2: Advancing Multimodal Perception with High-Resolution Scaling and Visual-Language Co-Referring](https://mlsanity.com/paper/2403.09333/) (2024)
- [FineCops-Ref: A new Dataset and Task for Fine-Grained Compositional Referring Expression Comprehension](https://mlsanity.com/paper/2409.14750/) (2024)
- [Advancing Visual Grounding with Scene Knowledge: Benchmark and Method](https://mlsanity.com/paper/2307.11558/) (2023)
- [Visual Intention Grounding for Egocentric Assistants](https://mlsanity.com/paper/2504.13621/) (2025)

## Abstract (arXiv)

Existing visual grounding benchmarks primarily evaluate alignment between image regions and literal referring expressions, where models can often succeed by matching a prominent named category. We explore a complementary and more challenging setting of scenario-based visual grounding, where the target must be inferred from roles, intentions, and relational context rather than explicit naming. We introduce Referring Scenario Comprehension (RSC), a benchmark designed for this setting. The queries in this benchmark are paragraph-length texts that describe object roles, user goals, and contextual cues, including deliberate references to distractor objects that often require deep understanding to resolve. Each instance is annotated with interpretable difficulty tags for uniqueness, clutter, size, overlap, and position which expose distinct failure modes and support fine-grained analysis. RSC contains approximately 31k training examples, 4k in-domain test examples, and a 3k out-of-distribution split with unseen object categories. We further propose ScenGround, a curriculum reasoning method serving as a reference point for this setting, combining supervised warm-starting with difficulty-aware reinforcement learning. Experiments show that scenario-based queries expose systematic failures in current models that standard benchmarks do not reveal, and that curriculum training improves performance on challenging slices and transfers to standard benchmarks.