Hidetsune at SemEval-2026 Task 11: Adapting Pretrained Reasoning Models with Deep Supervision and Inference Refinement for Content-Independent Validity Classification
Year: 2026
Type: conference-paper
Abstract: This paper presents a system that adopts several training and inference strategies for SemEval-2026 Task 11 Subtask 1, which focuses on binary classification for content-independent validity reasoning in syllogistic inference.Building on fine-tuning of standard language models, additional approaches were explored, including layer-wise deep supervision and in-context learning.Furthermore, models that had been previously trained on datasets related to logical reasoning were adapted to the task through additional fine-tuning.Finally, refinement was performed at the inference stage by adjusting the softmax-based decision threshold of the selected model.The experimental results illustrate how model selection, training strategies, and threshold adjustment affect not only validity accuracy but also robustness against plausibility-driven bias, thereby contributing to improved logical integrity.
Author: Hidetsune Takahashi
Institution: Waseda University
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7166853543