PAST-TIDE：基於原型錨定之語句調校與主題不變歸一化之立場偵測

PAST-TIDE: Prototype-Anchored Statement Tuning with Topic-Invariant Normalization for Stance Detection

July 6, 2026

作者: Md. Shakhoyat Rahman Shujon, MD Jahid Hasan Jim, Md. Milon Islam, Md Rezwanul Haque, Fakhri Karray

cs.AI

摘要

我們介紹PAST-TIDE，這是一個立場檢測系統，旨在解決NakbaNLP@LREC-COLING 2026的StanceNakba共享任務中的兩個子任務。核心思路是語句微調。我們將立場重新定義為完形填空式的遮罩語言模型（MLM），透過詞彙映射器將標籤詞對應到立場類別，並利用預訓練的MLM頭部，而非附加隨機初始化的分類頭部。我們進一步結合原型對比學習，該方法使用可學習的類別原型進行獨立於批次大小的對比訓練，並採用主題條件層正規化來處理跨主題的阿拉伯語立場檢測。PAST-TIDE在官方排行榜上分別於子任務A和子任務B達到0.75和0.74的巨集F1分數，顯示在低資源環境中，對預訓練模型進行最小的架構調整仍能保持競爭力。

English

We introduce PAST-TIDE, our stance detection system addressing both subtasks of the StanceNakba Shared Task at NakbaNLP@LREC-COLING 2026. The main idea is statement tuning. We redefine stance as cloze-style masked language modeling (MLM), letting a verbalizer map label words to stance categories through the pre-trained MLM head rather than appending a randomly initialized classification head. We complement this with prototypical contrastive learning, which uses learnable class prototypes for batch-size independent contrastive training, and topic-conditional layer normalization for cross-topic Arabic stance detection. PAST-TIDE achieves macro-F1 scores of 0.75 for Subtask A and 0.74 for Subtask B on the official leaderboard, indicating that minimal architectural additions to a pre-trained model can remain competitive in low-resource settings.