2026-06-07T14:11:09Zhttps://iris.cnr.it/oai/request

oai:iris.cnr.it:20.500.14243/5805822026-05-12T13:55:21Zou_ou239

Steering Pragmatic Interpretation in LLMs: A Diagnostic Evaluation of Few-Shot and Reasoning-Based Prompting for Indirect Speech Acts Massimiliano Orsini Dominique Brunato Orsini, Massimiliano Brunato, Dominique Italian Indirect Speech Acts Few-shot Prompting Large Language Models Evaluation Pragmatic competence presents a persistent challenge for Large Language Models (LLMs), as it requires contextdependent inference beyond literal meaning. This study examines whether few-shot prompting can reliably steer LLMs toward appropriate interpretations of indirect speech acts under small-data conditions. Focusing on Italian, we evaluate three LLMs on a small dataset that captures pragmatic ambiguity through graded plausibility judgments. We compare a zero-shot baseline with multiple few-shot prompting configurations that vary in the number and composition of demonstrations, as well as in the presence of explicit pragmatic guidance. Results show that few-shot prompting does not yield robust or monotonic improvements overall. While performance improves substantially for conventionalized indirect speech acts, gains for non-conventionalized indirect speech acts are unstable and limited. In contrast, introducing explicit pragmatic reasoning along with demonstrations through guided chain-of-thought prompting appears more promising. Overall, these findings highlight the limits of example-based steering for pragmatic inference and suggest that explicitly modeling pragmatic reasoning may be a more effective direction in small-data settings. 2026 info:eu-repo/semantics/conferenceObject https://hdl.handle.net/20.500.14243/580582 info:eu-repo/semantics/altIdentifier/isbn/978-2-493814-80-7 eng ispartofbook:Proceedings of the Workshop on Learning Non-Literal Expressions with Small Data, LREC 2026 Workshop on Learning Non-Literal Expressions with Small Data @ LREC 2026 firstpage:12 lastpage:20 numberofpages:9