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..

Monotone and Separable Set Functions: Characterizations and Neural Models

Authors: SOUTRIK SARANGI, Yonatan Sverdlov, Nadav Dym, Abir De

NeurIPS 2025 | Venue PDF | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimentally, we consider a variety of set containment tasks. The experiments show the benefit of using our MASNET model, in comparison with standard set models which do not incorporate set containment as an inductive bias. Our implementation is available in https://github.com/structlearning/MASNET. 5 Experiments We evaluate the MASNET variants (Section 4) on synthetic, text and point-cloud datasets to characterize monotonicity and separability. Specifically, we focus on (exact and approximate) set containment.
Researcher Affiliation Academia Soutrik Sarangi IIT Bombay Yonatan Sverdlov Technion Nadav Dym Technion Abir De IIT Bombay
Pseudocode No The paper describes methods in prose and mathematical equations, but does not include any clearly labeled pseudocode or algorithm blocks.
Open Source Code Yes Our implementation is available in https://github.com/structlearning/MASNET.
Open Datasets Yes We use Model Net40 [34], a dataset of 3D CAD models across 40 categories.
Dataset Splits Yes In each case, we split the dataset into 5:2:2 train, test, and dev folds.
Hardware Specification Yes All experiments were performed in a compute server, running the OS of GNU Linux Version 12, equipped with a 16 core Intel(R) Xeon(R) Gold 6130 CPU @ 2.10GHz CPU architecture and equipped with a cluster of 6 NVIDIA RTX A6000 GPUs with a memory of 49GB each.
Software Dependencies No The paper mentions tools like 'BERT embeddings' but does not specify version numbers for any software, libraries, or frameworks used in the experiments.
Experiment Setup Yes We minimize the following fixed-margin hinge loss that enforces vector dominance, to train the parameters of MASNET. X S,T (1 y(S, T)) mini [m] [F(S)[i] F(T)[i] + δ]+ + y(S, T) maxi [m] [F(S)[i] F(T)[i] + δ]+ (9) We chose d = 4 and the set embeddings have m = 256 dimensions, the class-ratio of subsets to non-subsets was taken to be 1:1. Gaussian noise with 0.01 std was used to generate noisy pairs. The class ratio of subsets to non-subsets was taken to be 1:1. for inexact set containment. We use a pointcloud encoder(such as pointnet) followed by a set-to-vector embedding model to produce m = 50 dimensional embeddings, and each point in the dataset is a 3D coordinate with d = 3 dimensions. We trained using MSE loss, and report test MAE results in Table 7