---
license: apache-2.0
library_name: transformers
pipeline_tag: image-text-to-text
base_model:
- LongVA/LongVA-7B
tags:
- vision-language
- remote-sensing
- ultra-high-resolution
- query-guided-token-compression
- qwen2
- longva
- uhr-bat
---

# UHR-BAT

**UHR-BAT: Budget-Aware Token Compression Vision-Language model for Ultra-High-Resolution Remote Sensing**

**ICML 2026**

<p align="center">
  <a href="https://yunkaidang.github.io/bibliography/dang2026uhr-bat/">
    <img src="https://img.shields.io/badge/Project-Page-2f855a?style=for-the-badge" alt="Project Page">
  </a>
  <a href="https://arxiv.org/abs/2604.13565">
    <img src="https://img.shields.io/badge/arXiv-2604.13565-b31b1b?style=for-the-badge&logo=arxiv&logoColor=white" alt="arXiv">
  </a>
  <a href="https://github.com/Yunkaidang/UHR-BAT">
    <img src="https://img.shields.io/badge/Code-GitHub-181717?style=for-the-badge&logo=github&logoColor=white" alt="GitHub Code">
  </a>
  <a href="https://huggingface.co/FelixKAI/UHR-BAT">
    <img src="https://img.shields.io/badge/Model-Hugging%20Face-ffcc4d?style=for-the-badge" alt="Hugging Face Model">
  </a>
  <img src="https://img.shields.io/badge/Conference-ICML%202026-4c6fff?style=for-the-badge" alt="ICML 2026">
</p>

[Project Page](https://yunkaidang.github.io/bibliography/dang2026uhr-bat/) | [Paper](https://arxiv.org/abs/2604.13565) | [Code](https://github.com/Yunkaidang/UHR-BAT)

UHR-BAT is a budget-aware vision-language framework for ultra-high-resolution remote sensing imagery. It targets the setting where kilometer-scale scenes contain query-critical evidence that may occupy only a few pixels. Instead of relying on direct downsampling, dense tiling, or generic global pruning, UHR-BAT uses query-guided multi-scale token selection and region-faithful compression to preserve small decisive evidence under a strict context budget.

## Highlights

- **Query-guided token compression:** visual token budgets are allocated according to the current instruction, helping preserve small but decisive evidence.
- **Multi-scale input:** the model encodes remote-sensing images at multiple target scales to retain both global context and fine-grained local details.
- **Region-faithful preserve and merge:** informative regional tokens are kept, while redundant background tokens are merged into compact representatives.
- **Efficient UHR understanding:** the method is designed for quality under memory and latency constraints, not only raw benchmark accuracy.

## Main Results

The project page reports strong ultra-high-resolution remote-sensing results under strict token budgets:

- **XLRS-Bench:** 44.0 weighted average accuracy.
- **MMERealworld-RS:** 33.33 mean score.
- **RSHR-Bench:** 29.2 on Perception and 45.0 on Reasoning.

## Model Details

This checkpoint contains the full multimodal UHR-BAT model:

- Qwen2/LongVA language backbone
- CLIP ViT-L/14-336 vision tower
- multimodal projector
- multiscale token MLP
- scale positional residual weights
- Hugging Face remote-code wrappers for direct loading

The model repository includes `configuration_uhr_bat.py` and `modeling_uhr_bat.py`, so `trust_remote_code=True` is required when loading the full architecture.

## Quick Start

```python
import importlib
import torch
from PIL import Image
from transformers import AutoImageProcessor
from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "FelixKAI/UHR-BAT"
image_path = "your_remote_sensing_image.jpg"
question = "Describe this remote-sensing image."

tokenizer = AutoTokenizer.from_pretrained(model_id)
if tokenizer.pad_token_id is None:
    tokenizer.pad_token = tokenizer.eos_token
image_processor = AutoImageProcessor.from_pretrained(model_id)

model = AutoModelForCausalLM.from_pretrained(
    model_id,
    trust_remote_code=True,
    torch_dtype="auto",
    device_map="auto",
).eval()

# Reuse the preprocessing helpers shipped with the model's remote code.
uhrbat = importlib.import_module(model.__class__.__module__)
image = Image.open(image_path).convert("RGB")

prompt = (
    "<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n"
    f"<|im_start|>user\n<image>\n{question}<|im_end|>\n"
    "<|im_start|>assistant\n"
)
image_token_id = getattr(model.config, "image_token_index", -200)
input_ids = uhrbat.tokenizer_image_token(
    prompt,
    tokenizer,
    image_token_id,
    return_tensors="pt",
).unsqueeze(0).to(model.device)
attention_mask = torch.ones_like(input_ids)

target_sizes = [672, 1344, 2688, 4032]
multiscale_pixels = [
    uhrbat.split_image_to_multiscale_tiles(
        image,
        image_processor,
        target_sizes=target_sizes,
        tile_size=336,
    )
]

with torch.inference_mode():
    output = model.generate(
        inputs=input_ids,
        attention_mask=attention_mask,
        image_sizes=[image.size],
        modalities=["image"],
        multiscale_pixels=multiscale_pixels,
        multiscale_masks=[{}],
        multiscale_topk=[80, 320, 600, 2000],
        multiscale_target_sizes=target_sizes,
        do_sample=False,
        max_new_tokens=256,
        return_dict_in_generate=True,
        output_scores=True,
        pad_token_id=tokenizer.pad_token_id or tokenizer.eos_token_id,
        eos_token_id=tokenizer.eos_token_id,
    )

prompt_len = output.sequences.shape[1] - len(output.scores)
answer_ids = output.sequences[:, prompt_len:].clone()
answer_ids[answer_ids < 0] = tokenizer.pad_token_id or tokenizer.eos_token_id
answer = tokenizer.decode(answer_ids[0], skip_special_tokens=True).strip()
print(answer)
```

## Citation

```bibtex
@inproceedings{dang2026uhrbat,
  title={UHR-BAT: Budget-Aware Token Compression Vision-Language model for Ultra-High-Resolution Remote Sensing},
  author={Dang, Yunkai and Dai, Minxin and Yang, Yuekun and Li, Zhangnan and Li, Wenbin and Miao, Feng and Gao, Yang},
  booktitle={International Conference on Machine Learning (ICML)},
  year={2026}
}
```