--- library_name: transformers license: mit language: - en metrics: - accuracy --- # Model Card for Logic2Vision Logic2Vision is a [LLaVA-1.5-13B](https://huggingface.co/llava-hf/llava-1.5-13b-hf) model finetuned on [VisReas dataset](https://arxiv.org/abs/2403.10534) for complex visual reasoning tasks. ![results](https://huggingface.co/RE-N-Y/logic2vision/resolve/main/Code2Vision.png) ## Model Details ### Model Description Logic2Vision is a [LLaVA-1.5-13B](https://huggingface.co/llava-hf/llava-1.5-13b-hf) model finetuned on [VisReas dataset](https://arxiv.org/abs/2403.10534) for complex visual reasoning tasks. The model has been finetuned using LoRA to generate python pseudocode outputs to solve a complex visual reasoning tasks. - **Developed by:** Sangwu Lee and Syeda Akter - **Model type:** Multimodal (Text + Image) - **Language(s) (NLP):** English - **License:** MIT - **Finetuned from model:** [LLaVA-1.5-13B](https://huggingface.co/llava-hf/llava-1.5-13b-hf) ### Model Sources - **Repository:** TBD - **Paper:** [VisReas dataset](https://arxiv.org/abs/2403.10534) ## Uses The inference method is similar to [LLaVA-1.5-13B](https://huggingface.co/llava-hf/llava-1.5-13b-hf). ### Example images **Question** ![zebras.jpg](https://huggingface.co/RE-N-Y/logic2vision/resolve/main/zebras.jpg) - Q: What else in the image is striped as the rope and the mane to the left of the white clouds? - A: Question is problematic. There are a couple zebras (i.e. mane/horses) at the back left of the clouds. But, there are clearly no ropes in the image. **Question** ![room.jpg](https://huggingface.co/RE-N-Y/logic2vision/resolve/main/room.jpg) - Q: What material do the chair and the table have in common? - A: Wood. ### Code ```python import torch from transformers import LlavaProcessor, LlavaForConditionalGeneration import requests from PIL import Image class LLaVACodeTemplate: prompt = """ USER: Executes the code and logs the results step-by-step to provide an answer to the question. Question {question} Code {codes} ASSISTANT: Log """ answer = """ {logs} Answer: {answer} """ template = LLaVACodeTemplate() model = LlavaForConditionalGeneration.from_pretrained("RE-N-Y/logic2vision", torch_dtype=torch.bfloat16, low_cpu_mem_usage=True, cache_dir="/data/tir/projects/tir6/general/sakter/cache") model.to("cuda") processor = LlavaProcessor.from_pretrained("RE-N-Y/logic2vision") processor.tokenizer.pad_token = processor.tokenizer.eos_token processor.tokenizer.padding_side = "left" image = Image.open(requests.get("https://huggingface.co/RE-N-Y/logic2vision/resolve/main/zebras.jpg", stream=True).raw) question = "What else in the image is striped as the rope and the mane to the left of the white clouds?" codes = """selected_clouds = select(clouds) filtered_clouds = filter(selected_clouds, ['white']) related_mane = relate(mane, to the left of, o, filtered_clouds) selected_rope = select(rope) pattern = query_pattern(['selected_rope', 'related_mane']) result = select(objects, attr=pattern) """ prompt = template.prompt.format(question=question, codes=codes) inputs = processor(images=image, text=prompt, return_tensors="pt") inputs.to("cuda") generate_ids = model.generate(**inputs, max_new_tokens=256) output = processor.batch_decode(generate_ids, skip_special_tokens=True) print(output[0]) # USER: # Executes the code and logs the results step-by-step to provide an answer to the question. # Question # What else in the image is striped as the rope and the mane to the left of the white clouds? # Code # selected_clouds = select(clouds) # filtered_clouds = filter(selected_clouds, ['white']) # related_mane = relate(mane, to the left of, o, filtered_clouds) # selected_rope = select(rope) # pattern = query_pattern(['selected_rope', 'related_mane']) # result = select(objects, attr=pattern) # ASSISTANT: # Log # ('clouds', ['white']) # ('clouds', ['white']) # ('mane', ['striped']) # ('rope', ['no object']) # ['the question itself is problematic'] # ['the question itself is problematic'] # Answer: # the question itself is problematic image = Image.open(requests.get("https://huggingface.co/RE-N-Y/logic2vision/resolve/main/room.jpg", stream=True).raw) question = "What material do the chair and the table have in common?" codes = """selected_chair = select(chair) selected_table = select(table) materials = query_material([selected_chair, selected_table]) common_material = common(materials) """ prompt = template.prompt.format(question=question, codes=codes) inputs = processor(images=image, text=prompt, return_tensors="pt") inputs.to("cuda") generate_ids = model.generate(**inputs, max_new_tokens=256) output = processor.batch_decode(generate_ids, skip_special_tokens=True) print(output[0]) # USER: # Executes the code and logs the results step-by-step to provide an answer to the question. # Question # What material do the chair and the table have in common? # Code # selected_chair = select(chair) # selected_table = select(table) # materials = query_material([selected_chair, selected_table]) # common_material = common(materials) # ASSISTANT: # Log # ('chair', ['wood']) # ('table', ['wood']) # [['wood'], ['wood']] # ['wood'] # Answer: # wood ``` ## Bias, Risks, and Limitations The model has been mostly trained on VisReas dataset which is generated from [Visual Genome](https://homes.cs.washington.edu/~ranjay/visualgenome/index.html) dataset. Furthermore, since the VLM was mostly finetuned to solve visual reasoning tasks by "generating python pseudocode" outputs provided by the user. Hence, it may struggle to adopt to different prompt styles and code formats. ## Training / Evaluation Details The model has been finetuned using 2 A6000 GPUs on CMU LTI's Babel cluster. The model has been finetuned using LoRA (`r=8, alpha=16, dropout=0.05, task_type="CAUSAL_LM"`). LoRA modules were attached to `["q_proj", "v_proj"]`. We use DDP for distributed training and BF16 to speed up training. For more details, check [our paper](https://arxiv.org/abs/2403.10534)! ### Results ![results](https://huggingface.co/RE-N-Y/logic2vision/resolve/main/results.png) ## Citation **BibTeX:** ``` @misc{akter2024visreas, title={VISREAS: Complex Visual Reasoning with Unanswerable Questions}, author={Syeda Nahida Akter and Sangwu Lee and Yingshan Chang and Yonatan Bisk and Eric Nyberg}, year={2024}, eprint={2403.10534}, archivePrefix={arXiv}, primaryClass={cs.CV} } ``` ## Model Card Authors - Sangwu Lee - [Google Scholar](https://scholar.google.com/citations?user=FBJeGpAAAAAJ) - [Github](https://github.com/RE-N-Y) - [LinkedIn](https://www.linkedin.com/in/sangwulee/) - Syeda Akter - [Google Scholar](https://scholar.google.com/citations?hl=en&user=tZFFHYcAAAAJ) - [Github](https://github.com/snat1505027) - [LinkedIn](https://www.linkedin.com/in/syeda-nahida-akter-989770114/)