PulseAugur
EN
LIVE 07:39:31

PAST-TIDE system enhances stance detection with novel tuning methods · 3 sources tracked

By PulseAugur Editorial · [3 sources] ·

Researchers have developed PAST-TIDE, a novel system for stance detection that addresses both subtasks of the StanceNakba Shared Task. The system utilizes statement tuning, redefining stance detection as a masked language modeling task and employing a verbalizer to map labels to stance categories. It also incorporates prototypical contrastive learning and topic-conditional layer normalization for improved performance, particularly in low-resource settings for Arabic language. AI

IMPACT This research offers a novel approach to stance detection, potentially improving performance in low-resource scenarios and cross-topic analysis.

RANK_REASON The cluster contains an academic paper detailing a new method for stance detection.

Read on arXiv cs.CL →

AI-generated summary · Google Gemini · from 3 sources. How we write summaries →

PAST-TIDE system enhances stance detection with novel tuning methods · 3 sources tracked

COVERAGE [3]

  1. arXiv cs.CL TIER_1 English(EN) · Md. Shakhoyat Rahman Shujon, MD Jahid Hasan Jim, Md. Milon Islam, Md Rezwanul Haque, Fakhri Karray ·

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

    arXiv:2607.04690v1 Announce Type: new Abstract: 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 …

  2. arXiv cs.CL TIER_1 English(EN) · Fakhri Karray ·

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

    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 s…

  3. Hugging Face Daily Papers TIER_1 English(EN) ·

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

    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 s…

RELATED ENTITIES

RELATED TOPICS