LLM-INSTRUCT at UZH Shared Task 2026: Constraint-Aware Retrieval and Selective Debate for Paragraph-Level Argument Mining
cs.CL, cs.AI
Submitted: 2026-05-10
Updated: 2026-05-10
Comments: Accepted to the 13th Workshop on Argument Mining (ArgMining 2026) at ACL 2026
Journal ref: Proceedings of the 13th Workshop on Argument Mining (ArgMining 2026), pages 119-124
DOI: 10.18653/v1/2026.argmining-1.11
Code: https://github.com/LLMInstruct-at-UZH-Shared-Task-2026/Method
License: http://creativecommons.org/licenses/by/4.0/
The gist: We present LLM-INSTRUCT, the winning system for the UZH Shared Task at ArgMining 2026 on paragraph-level argument mining in UN and UNESCO resolutions.
Terminology
Abstract
We present LLM-INSTRUCT, the winning system for the UZH Shared Task at ArgMining 2026 on paragraph-level argument mining in UN and UNESCO resolutions. The task requires paragraph-type classification, prediction of a subset of 141 official tags, and directed relation prediction under a strict JSON schema setting using only open-weight models up to 8B parameters. We frame the task as constrained structured prediction. The system first narrows the candidate tag space with metadata-aware dense retrieval, then applies constrained decoding with per-dimension caps, escalates only uncertain cases to a three-agent debate branch, and finally validates the output schema. On the official leaderboard, LLM-INSTRUCT ranked 1st overall, with 1st in F1 and 5th in LLM-as-a-Judge. During development, our configuration search further improved Task 1b Micro-F1 from 35.83% to 40.08% while keeping the internal Task 2 score at 4.421. The main lesson is simple: reducing the decision space before generation improves both accuracy and submission robustness. Our code and supporting scripts are publicly available at: https://github.com/LLM-Instruct-at-UZH-Shared-Task-2026/Method
Sources
- The Llama 3 Herd of Models
- Structured Prediction as Translation between Augmented Natural Languages
- Text Embeddings by Weakly-Supervised Contrastive Pre-training
- Qwen3 Technical Report
Related papers
- ALARM: Audio-Language Alignment for Reasoning Models
- ThinkSum: Probabilistic reasoning over sets using large language models
- Coherence boosting: When your pretrained language model is not paying enough attention
- From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents
- Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization
- The Art of Asking: Multilingual Prompt Optimization for Synthetic Data