Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].
CLIPAway: Harmonizing focused embeddings for removing objects via diffusion models
Authors: Yiğit Ekin, Ahmet Burak Yildirim, Erdem Eren Çağlar, Aykut Erdem, Erkut Erdem, Aysegul Dundar
NeurIPS 2024 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We provide comprehensive evaluations on a standard dataset, demonstrating consistent improvements over state-of-the-art methods. |
| Researcher Affiliation | Academia | Department of Computer Engineering, Bilkent University, Ankara, Turkey Department of Computer Engineering, Koç University, Istanbul, Turkey KUIS AI Center, Koç University, Istanbul, Turkey Department of Computer Engineering, Hacettepe University, Ankara, Turkey |
| Pseudocode | Yes | Algorithm 1 Algorithm for our training and inference |
| Open Source Code | Yes | Code and models are available via our project website: https://yigitekin.github.io/CLIPAway/. |
| Open Datasets | Yes | We evaluate the models on the COCO 2017 validation dataset [15], which provides us with the collection of images indoor and outdoor and instance level annotations. |
| Dataset Splits | Yes | We evaluate the models on the COCO 2017 validation dataset [15], which provides us with the collection of images indoor and outdoor and instance level annotations. |
| Hardware Specification | Yes | The MLP model is trained on a single NVIDIA A40 GPU with batch size 8. |
| Software Dependencies | No | No specific version numbers for software dependencies (e.g., Python, PyTorch, CUDA) are explicitly provided in the paper. |
| Experiment Setup | Yes | Adam, optimizer is used with learning rate and weight decay are set to 1e 5 and 1e 4, respectively. The MLP model is trained on a single NVIDIA A40 GPU with batch size 8. |