Year: 2026
Type: conference-paper
Abstract: Subtask 1 of the UZH Shared Task 2026 asks for paragraph-level classification of UN resolutions as preambular or operative and multilabel tagging from a 141-code, 15-dimension taxonomy, scored by tag F1 and an open-weight LLM-as-Judge on reasoning quality.Two earlier pipelines we built failed in opposite ways.An embedding-retrieval system dropped relevant tags before the LLM saw them.A perdimension prompting system was accurate but too slow to iterate.The submitted system fixes both.A deterministic French-English lexical classifier assigns paragraph types at type macro-F1 of 0.910 on the official silver standard with no LLM calls.DeepSeek-R1-0528-Qwen3-8B (DeepSeek-AI, 2025) predicts tags through a single merged prompt that exposes the full taxonomy, hints likely dimensions, and walks the model through an eight-point checklist.Re-solveNow places 7th overall, with 2nd on LLMas-Judge and the 9th-place F1 rank explained by the absence of any Subtask 2 submission rather than by tagging quality on Subtask 1 in isolation.Code:
Authors: Keying Yang, Hao Wang, Chengtao Jian, Kai Yang
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
OpenAlex ID: W7166799084