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 Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
Multi-Instance Causal Representation Learning for Instance Label Prediction and Out-of-Distribution Generalization
Authors: Weijia Zhang, Xuanhui Zhang, hanwen deng, Min-Ling Zhang
NeurIPS 2022 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments on synthetic and real-world datasets demonstrate that our approach significantly outperforms various baselines on instance label prediction and out-of-distribution generalization tasks. In this section, we first evaluate the instance label prediction performances of Causal MIL against MIL baselines... Then, we evaluate the out-of-distribution generalization ability of Causal MIL by comparing with supervised learning algorithms... |
| Researcher Affiliation | Academia | Weijia Zhang1 , Xuanhui Zhang2, Han-Wen Deng1, Min-Ling Zhang1 1School of Computer Science and Engineering, Southeast University, Nanjing 210096, China 2 School of Information Management, Nanjing University, Nanjing 210023, China EMAIL, EMAIL, EMAIL |
| Pseudocode | No | No section or figure explicitly labeled 'Pseudocode' or 'Algorithm' was found in the paper. |
| Open Source Code | Yes | The implementation code is publicly available at https://github. com/Weijia Zhang24/Causal MIL. |
| Open Datasets | Yes | Our quantitative evaluation first utilizes 30 multi-instance classification datasets generated from the MNIST [19], Fashion MNIST [38], and Kuzushiji MNIST [7] datasets... We also evaluate Causal MIL on a hematoxylin and eosin (H&E) stained Colon Cancer histopathology task [32]. |
| Dataset Splits | Yes | Experiments are repeated for 5 times and the average metric standard deviation of 5-fold cross validations are reported. |
| Hardware Specification | Yes | We implemented Causal MIL using Py Torch and conducted most of the experiments with a single NVIDIA RTX3090 GPU. |
| Software Dependencies | No | We implemented Causal MIL using Py Torch and conducted most of the experiments with a single NVIDIA RTX3090 GPU. |
| Experiment Setup | Yes | Detailed settings, parameters, and data for reproducing the results are provided in the Appendices. |