Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2507.08679 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908542314741760 |
|---|---|
| author | Roy, Rajarshi Das, Devleena Banerjee, Ankesh Bhattacharjee, Arjya Dasgupta, Kousik Tripathi, Subarna |
| author_facet | Roy, Rajarshi Das, Devleena Banerjee, Ankesh Bhattacharjee, Arjya Dasgupta, Kousik Tripathi, Subarna |
| contents | We introduce ByDeWay, a training-free framework designed to enhance the performance of Multimodal Large Language Models (MLLMs). ByDeWay uses a novel prompting strategy called Layered-Depth-Based Prompting (LDP), which improves spatial reasoning and grounding without modifying any model parameters. It segments the scene into closest, mid-range, and farthest layers using monocular depth estimation, then generates region-specific captions with a grounded vision-language model. These structured, depth-aware captions are appended to the image-question prompt, enriching it with spatial context. This guides MLLMs to produce more grounded and less hallucinated responses. Our method is lightweight, modular, and compatible with black-box MLLMs. Experiments on hallucination-sensitive (POPE) and reasoning-intensive (GQA) benchmarks show consistent improvements across multiple MLLMs, validating the effectiveness of depth-aware prompting in a zero-training setting. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_08679 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | ByDeWay: Boost Your multimodal LLM with DEpth prompting in a Training-Free Way Roy, Rajarshi Das, Devleena Banerjee, Ankesh Bhattacharjee, Arjya Dasgupta, Kousik Tripathi, Subarna Computer Vision and Pattern Recognition We introduce ByDeWay, a training-free framework designed to enhance the performance of Multimodal Large Language Models (MLLMs). ByDeWay uses a novel prompting strategy called Layered-Depth-Based Prompting (LDP), which improves spatial reasoning and grounding without modifying any model parameters. It segments the scene into closest, mid-range, and farthest layers using monocular depth estimation, then generates region-specific captions with a grounded vision-language model. These structured, depth-aware captions are appended to the image-question prompt, enriching it with spatial context. This guides MLLMs to produce more grounded and less hallucinated responses. Our method is lightweight, modular, and compatible with black-box MLLMs. Experiments on hallucination-sensitive (POPE) and reasoning-intensive (GQA) benchmarks show consistent improvements across multiple MLLMs, validating the effectiveness of depth-aware prompting in a zero-training setting. |
| title | ByDeWay: Boost Your multimodal LLM with DEpth prompting in a Training-Free Way |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2507.08679 |