We introduce SMProbLog, a generalization of the probabilistic logic\nprogramming language ProbLog. A ProbLog program defines a distribution over\nlogic programs by specifying for each clause the probability that it belongs to\na randomly sampled program, and these probabilities are mutually independent.\nThe semantics of ProbLog is given by the success probability of a query, which\ncorresponds to the probability that the query succeeds in a randomly sampled\nprogram. It is well-defined when each random sample uniquely determines the\ntruth values of all logical atoms. Argumentation problems, however, represent\nan interesting practical application where this is not always the case.\nSMProbLog generalizes the semantics of ProbLog to the setting where multiple\ntruth assignments are possible for a randomly sampled program, and implements\nthe corresponding algorithms for both inference and learning tasks. We then\nshow how this novel framework can be used to reason about probabilistic\nargumentation problems. Therefore, the key contribution of this paper are: a\nmore general semantics for ProbLog programs, its implementation into a\nprobabilistic programming framework for both inference and parameter learning,\nand a novel approach to probabilistic argumentation problems based on such\nframework.\n