KUET at StanceNakba Shared Task: StanceMoE: Mixture-of-Experts Architecture for Stance Detection

Dimensions overall score 7.0

• A Mixture-of-Experts model built on BERT for actor-level stance detection, leveraging specialized modules to capture diverse linguistic signals. Although transformer-based models have achieved relatively good performance in stance classification, they typically rely on unified representations that may not sufficiently capture heterogeneous linguistic signals, such as contrastive discourse structures, framing cues, and salient lexical indicators.problem 35%
  Evidencepartial

  A Mixture-of-Experts model built on BERT for actor-level stance detection, leveraging specialized modules to capture diverse linguistic signals. Although transformer-based models have achieved relatively good performance in stance classification, they typically rely on unified representations that may not sufficiently capture heterogeneous linguistic signals, such as contrastive discourse structures, framing cues, and salient lexical indicators.

  Implicationpartial

  Abstract-backed fallback claim; anchored extraction has not materialized a public claim row yet.

  Verificationpartial

  partial

• Actor-level stance detection aims to determine an author expressed position toward specific geopolitical actors mentioned or implicated in a text. Although transformer-based models have achieved relatively good performance in stance classification, they typically rely on unified representations that may not sufficiently capture heterogeneous linguistic signals, such as contrastive discourse structures, framing cues, and salient lexical indicators.method 35%
  Evidencepartial

  Actor-level stance detection aims to determine an author expressed position toward specific geopolitical actors mentioned or implicated in a text. Although transformer-based models have achieved relatively good performance in stance classification, they typically rely on unified representations that may not sufficiently capture heterogeneous linguistic signals, such as contrastive discourse structures, framing cues, and salient lexical indicators.

  Implicationpartial

  Abstract-backed fallback claim; anchored extraction has not materialized a public claim row yet.

  Verificationpartial

  partial

• ScienceToStartup currently rates this 7.0/10 on the public viability pass. StanceMoE achieves a macro-F1 score of 94.26%, outperforming traditional baselines, and alternative BERT-based variants. Code availability is flagged in the production record; the public repository link still needs proof alignment.result 35%
  Evidencepartial

  ScienceToStartup currently rates this 7.0/10 on the public viability pass. StanceMoE achieves a macro-F1 score of 94.26%, outperforming traditional baselines, and alternative BERT-based variants. Code availability is flagged in the production record; the public repository link still needs proof alignment.

  Implicationpartial

  Abstract-backed fallback claim; anchored extraction has not materialized a public claim row yet.

  Verificationpartial

  partial

• Stance Detection moved forward this cycle; last verified April 2026. Public score 7.0/10. Production flags indicate code availability.why now 35%
  Evidencepartial

  Stance Detection moved forward this cycle; last verified April 2026. Public score 7.0/10. Production flags indicate code availability.

  Implicationpartial

  Abstract-backed fallback claim; anchored extraction has not materialized a public claim row yet.

  Verificationpartial

  partial

Competitive landscape

A Mixture-of-Experts model built on BERT for actor-level stance detection, leveraging specialized modules to capture diverse linguistic signals.