On the afternoon of May 13, 2026, at the invitation of the School of Foreign Languages of Nanjing Normal University, Professor Yang Kun, doctoral supervisor from the School of Foreign Languages at Southwest University, delivered an academic lecture entitled “Theoretical Breakthroughs and Innovative Pathways in Construction Grammar Research in the Age of Artificial Intelligence” in Room 101 of the Audio-Visual Building on the Suiyuan Campus. The lecture was hosted by Professor Zhang Hui and attracted the enthusiastic participation of numerous MA and PhD students as well as young faculty members.
As a syntactic theory within Cognitive Linguistics, Construction Grammar has achieved substantial progress over the past four decades in areas such as linguistic knowledge representation, constructional networks, language acquisition, and language evolution. However, Professor Yang pointed out that, due to the limitations of traditional research approaches and methodologies, several key theoretical issues have remained unresolved for a long time. The rapid development of artificial intelligence in big data mining, pattern recognition, and inferential modeling has brought new opportunities to Construction Grammar research.
During the lecture, Professor Yang systematically reviewed the theoretical strengths of Construction Grammar and the pressing issues that call for breakthroughs. He focused on how artificial intelligence can empower Construction Grammar research in four major areas: the automatic identification and classification of constructions; the in-depth exploration of semantic constraints and local productivity; the diachronic tracing and dynamic tracking of constructions; and constructional typology supported by large-scale cross-linguistic data. Drawing on numerous concrete examples—including the Chinese “V de O” split construction, the semi-schematic “X + assassin” construction, and the extraction of semantic constraints on the English ditransitive construction—he vividly demonstrated how AI helps researchers move beyond the limitations of small-scale corpora and introspective judgments.
Professor Yang particularly emphasized that artificial intelligence not only broadens the scope of Construction Grammar research but also facilitates the transition from theoretical explanation to the exploration of verifiable and modelable cognitive mechanisms. He noted that AI’s distributional learning and probabilistic modeling are highly compatible with the usage-based perspective of Construction Grammar, providing new methodological support for revealing gradient structures, center–periphery organization, and local productivity within constructional networks. At the same time, he objectively discussed the current limitations of AI in areas such as physical world knowledge, social cognition, and the extraction of complex grammatical rules, reminding researchers to maintain an “AI-assisted” rather than “AI-dependent” research stance.
In conclusion, Professor Yang remarked: “Artificial intelligence is not a replacement for linguistic research, but a powerful tool. Ultimately, research quality is determined by scholars’ research questions, theoretical competence, and explanatory capacity.” He encouraged faculty and students to actively embrace the methodological benefits brought by AI and to achieve new advances in theoretical innovation and interdisciplinary applications in Construction Grammar.
The lecture was forward-looking, rich in examples, and clearly structured, presenting a new vision for Construction Grammar research in the age of artificial intelligence while offering valuable academic inspiration and methodological guidance for all attendees.