Instructions to use CheneyWong/povl-dpo-2ep with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use CheneyWong/povl-dpo-2ep with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="CheneyWong/povl-dpo-2ep", trust_remote_code=True)
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("CheneyWong/povl-dpo-2ep", trust_remote_code=True)
model = AutoModelForMultimodalLM.from_pretrained("CheneyWong/povl-dpo-2ep", trust_remote_code=True, device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use CheneyWong/povl-dpo-2ep with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "CheneyWong/povl-dpo-2ep"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CheneyWong/povl-dpo-2ep",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/CheneyWong/povl-dpo-2ep

SGLang

How to use CheneyWong/povl-dpo-2ep with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "CheneyWong/povl-dpo-2ep" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CheneyWong/povl-dpo-2ep",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "CheneyWong/povl-dpo-2ep" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CheneyWong/povl-dpo-2ep",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use CheneyWong/povl-dpo-2ep with Docker Model Runner:
```
docker model run hf.co/CheneyWong/povl-dpo-2ep
```

povl-dpo-2ep

What this checkpoint is

Direct preference optimisation on top of the corrective SFT, 2 epochs over 8,383 preference pairs. Intended to move the model from 'roughly right structure' to 'exactly right structure'.

Lineage

Fine-tuned from CheneyWong/povl-sft-2d-merge, which is its own immediate predecessor in the same run.

PaddleOCR-VL-1.6 (base)
  └─ povl-dpo-2ep   <-- this checkpoint

Results on public datasets

Base = the unmodified base model. Public datasets only; the project's self-built benchmarks are excluded.

Dataset	n	Base TEDS	Ours TEDS	d TEDS	Base TEDS-S	Ours TEDS-S	d TEDS-S
PubTables-1M	2000	0.9015	0.9403	+0.0388	0.9243	0.9503	+0.0259
PubTables-v2	2000	0.9026	0.9556	+0.0530	0.9239	0.9644	+0.0405
FinTabNet (OTSL)	2000	0.9442	0.9651	+0.0209	0.9524	0.9726	+0.0202
PubTabNet / ICDAR	2000	0.9297	0.9293	-0.0004	0.9464	0.9425	-0.0038
SynthTabNet	2000	0.8093	0.9194	+0.1100	0.8474	0.9316	+0.0842
TableVerse-5K	?	?	?	?	?	?	?
OmniDocBench v1.5	?	?	?	?	?	?	?
OmniDocBench v1.6	?	?	?	?	?	?	?
CC-OCR	?	?	?	?	?	?	?
OCRBench-v2	?	?	?	?	?	?	?
Real5 (re-shot OmniDoc)	?	?	?	?	?	?	?
SPRINT-MUSTARD	?	?	?	?	?	?	?
TABLET	?	?	?	?	?	?	?
Overall (equal weight)	-	0.8975	0.9419	+0.0445	0.9189	0.9523	+0.0334

Note on SPRINT-MUSTARD: its ground truth carries no cell text, so its TEDS (~0.22 here) measures how well the model reproduces empty cells and should be ignored -- a perturbation test shows a structurally perfect prediction that correctly reads the visible text scores 0.33 TEDS / 1.00 TEDS-S. Only the TEDS-S column is meaningful for that dataset.

Known limitations

The gain is within sampling noise (+/- 0.5 pt) and should be treated as no effect. Worth recording precisely because it is evidence about where the remaining headroom is not.

Task format: a single table image plus the prompt Table Recognition:, one pass, emitting OTSL markup (<fcel> <ecel> <ucel> <lcel> <xcel> <nl>). Evaluation numbers are tracked separately and are not part of this card.

Downloads last month: 29

Safetensors

Model size

0.9B params

Tensor type

BF16