Research map / Graph learning / Graph self-supervised learning
MixScentNet: A Multiscale Graph-based Framework for Predicting Scent Mixture Perception
Xingran Liao, Mingliang Zhou, Weisi Lin
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes MixScentNet, a multiscale graph-based framework for predicting the perceived odor of scent mixtures, addressing limitations of single-molecule encoders with mean pooling, concatenation, or self-attention aggregation.
Problem
Predicting olfactory perception of scent mixtures is hard; existing methods encode each component with a single-molecule encoder pretrained on annotated olfactory data and aggregate features by mean pooling, concatenation, or self-attention, which faces data scarcity and limited modeling of mixture
Approach
A multiscale graph-based framework that models scent mixtures at multiple scales rather than aggregating independent single-molecule embeddings; full method details are not available in the provided abstract.
Closest papers by meaning
- Can Transformers Smell Like Humans? (NeurIPS 2024 · similarity 0.86)
- MixMin: Finding Data Mixtures via Convex Minimization (ICML 2025 · similarity 0.82)
- CheMixHub: Datasets and Benchmarks for Chemical Mixture Property Prediction (NeurIPS 2025 · similarity 0.81)
- Efficient Mixture Learning in Black-Box Variational Inference (ICML 2024 · similarity 0.80)
- SCENT: Robust Spatiotemporal Learning for Continuous Scientific Data via Scalable Conditioned Neural Fields (ICML 2025 · similarity 0.80)
- On the Scalability of GNNs for Molecular Graphs (NeurIPS 2024 · similarity 0.80)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
