The second issue of the sixth year of the Italian Journal of Computational Linguistics (IJCoL) is a miscellaneous volume which reports original findings achieved in the currently investigated lines of research within the national computational linguistics community. The contributions collected in the volume cover recent and fruitful developments in computational linguistics research, ranging from the computational modelling of lexical semantic change from different perspectives (both diachronic and synchronic), to unsupervised methods for identifying difficult-to-parse syntactic constructions and Natural Language Generation, to the EVALITA evaluation campaign specifically devoted to Natural Language Processing and Speech tools for Italian, its past and current achievements.

Editorial Note
Roberto Basili, Simonetta Montemagni.

Embeddings-based detection of word use variation in Italian newspapers
Michele Cafagna, Lorenzo De Mattei, Malvina Nissim.

Analyzing Gaussian distribution of semantic shifts in Lexical Semantic Change Models
Pierluigi Cassotti, Pierpaolo Basile, Marco de Gemmis, Giovanni Semeraro.

Linguistically-driven Selection of Difficult-to-Parse Dependency Structures
Chiara Alzetta, Felice Dell’Orletta, Simonetta Montemagni, Giulia Venturi.

ETC-NLG: End-to-end Topic-Conditioned Natural Language Generation
Ginevra Carbone, Gabriele Sarti.

Lessons Learned from EVALITA 2020 and Thirteen Years of Evaluation of Italian Language Technology
Lucia C. Passaro, Maria Di Maro, Valerio Basile, Danilo Croce.

EVALITA4ELG: Italian Benchmark Linguistic Resources, NLP Services and Tools for the ELG Platform
Viviana Patti, Valerio Basile, Cristina Bosco, Rossella Varvara, Michael Fell, Andrea Bolioli, Alessio Bosca.