Abstract: The sulfur and nitrogen elements in low-rank coal significantly impactits processing, conversion, and utilization. A clear understanding of the occurrence forms, structure, and migration patterns of sulfur and nitrogen during thermal conversion not only mitigates their potential environmental hazards but also enables the targeted conversion of these elements intosulfur- ornitrogen-containing chemicals and the development of sulfur-/nitrogen-doped novel carbon materials for high-value applications. To this end, this paper systematically summarizes the occurrence forms of sulfur and nitrogen in low-rank coal; analyzes the effects of pyrolysis atmosphere, pyrolysis temperature, catalysts, and other factors on the distribution characteristics and migration pathways of sulfur- and nitrogen-containing compounds in coal pyrolysis products; and explores the application of machine learning methods such as Random Forest and LightGBM in predicting pyrolysis products. Sulfur in low-rank coal primarily exists as organic sulfur, while nitrogen is predominantly present as four types of organic nitrogen: pyrrolic nitrogen, pyridinic nitrogen, quaternary nitrogen, andoxidized pyridinic nitrogen. Specifically, within vitrinite, the predominant sulfur-containing functional groups are thiophene, thiol, and thioether, while the nitrogen-containing functional groups are primarily pyridinic and benzonitrile derivatives. In contrast, within inertinite, the forms of sulfur-containing functional groups are similar to those in vitrinite, but the nitrogen is predominantly present as amine and pyrrolic structures.Increasing pyrolysis temperature promotes the decomposition of sulfur and nitrogen elements. Slower heating rates favor the removal of organic sulfur, whilefaster heating rates are more conducive to the migration of nitrogen to gaseous products. H₂, water vapor, and CO₂ atmospheres all promote the decomposition of sulfur- and nitrogen-containing compounds. Specifically, H₂ and water vapor provide hydrogen radicals that attack sulfur and nitrogen atoms in heterocyclic aromatics, thereby accelerating their decomposition. The CO₂ atmosphere promotes C—S, C—C, and C—N bond cleavage, accelerating the formation of gaseous sulfur- and nitrogen-containing compounds. Both calcium-based and iron-based catalysts exhibit sulfur fixationcapabilities while also influencing the conversion of nitrogen, typically promoting its release as gaseous species such as HCN and NH₃. During pyrolysis, inorganic sulfur （primarily pyrite） is transformed into pyrrhotite, which further reacts with active hydrogen and CO to form gaseous products like H₂S and COS; the undecomposed fraction remains in the char. Organic sulfur decomposition primarily occurs through C—S bond cleavage. The resulting sulfur-containing radicals react with hydrogen atoms or other hydrogen donors to form gaseous products like H₂S and SO₂. Other sulfur-containing groups polymerize or combine with aromatic rings to form polycyclic sulfur-containing aromatics, migrating into tar and char. Nitrogen in pyrolysis gas originates from the ring-opening reactions of nitrogen-containing heterocycles like pyridine and quinoline. Nitrogen in tar derives from the elimination and reorganization of heterocyclic compounds such as pyridines and pyrroles, while highly stable organic nitrogen remains in the char. Using out-of-bag estimation for hyperparameter optimization of the random forest algorithm reduced the prediction deviation for naphthobenzothiopheneto 0.11%. The LightGBM model built on raw coal physical parameters achieved a prediction accuracy with a coefficient of determination （R²） of 0.91 for morphological forms of sulfur. Further hyperparameter optimization using Hyperopt not only reduced the computation time by 60% but alsoincreased model’s R² to 0.96. In summary, elucidating the migration and transformation characteristics and mechanisms of sulfur and nitrogen during the pyrolysis of low-rank coal, and constructing machine learning prediction models with multi-source feature parameter inputs, provide significant theoretical and practical guidancefor the targeted transformation of sulfur- and nitrogen-containing structural units in coal, their high-value utilization, and the development of technologies for reducing pollutant emissions.