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..

SaFiRe: Saccade-Fixation Reiteration with Mamba for Referring Image Segmentation

Authors: Zhenjie Mao, Yang Yuhuan, Chaofan Ma, Dongsheng Jiang, Jiangchao Yao, Ya Zhang, Yanfeng Wang

NeurIPS 2025 | Venue PDF | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on both standard and proposed datasets demonstrate the superiority of Sa Fi Re over state-of-the-art baselines.
Researcher Affiliation Collaboration 1Cooperative Medianet Innovation Center, Shanghai Jiao Tong University 2School of Artificial Intelligence, Shanghai Jiao Tong University 3Institute of Artificial Intelligence for Medicine, Shanghai Jiao Tong University School of Medicine 4Huawei Inc.
Pseudocode No The paper describes the model architecture and operations (Saccade and Fixation) in Sections 3.1, 3.2, and 3.3 using natural language and mathematical equations, but does not include any clearly labeled pseudocode or algorithm blocks.
Open Source Code No Project page: https://zhenjiemao.github.io/Sa Fi Re/. (Abstract); We will release the code and dataset on publication. (Neur IPS Paper Checklist, Q5 Justification)
Open Datasets Yes Following prior works [7, 16, 9], we systematically assess model performance on the widely used Ref COCO benchmarks: Ref COCO [13], Ref COCO+ [13] and Ref COCOg [37]. These datasets are all grounded in the MSCOCO [38] visual corpus but differ significantly in their linguistic properties.
Dataset Splits Yes Following prior works [7, 16, 9], we systematically assess model performance on the widely used Ref COCO benchmarks: Ref COCO [13], Ref COCO+ [13] and Ref COCOg [37]...U indicates UMD partition of Ref COCOg. The best performances are in bold. (Table 1 caption)
Hardware Specification Yes All experiments are conducted on 2 NVIDIA A100 GPUs.
Software Dependencies No The paper mentions software components like Swin-Transformer, BERT, Adam W optimizer, and a cosine learning rate scheduler, but does not provide specific version numbers for these software dependencies.
Experiment Setup Yes The model is trained end-to-end for 50 epochs using the Adam W optimizer with a learning rate of 5e-5 and weight decay of 1e-4, along with a cosine learning rate scheduler. All experiments are conducted on 2 NVIDIA A100 GPUs.