Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
October 16, 2025Open Access

Dynamic Topic Evolution with Temporal Decay and Attention in Large Language Models

View Full Paper
Ask AI
Bookmark
Share

Authors

DWDi WuSPShaowei Pan

Discussion

Loading...

Member takes

Overview

This approach enhances topic coherence and diversity using temporal decay and an attention mechanism, demonstrating effective topic modeling.

Key Points

  • The method improves topic coherence and diversity by adjusting semantic unit importance over time, capturing topic evolution.
  • Experiments show notable improvements in topic modeling compared to existing models across various metrics.
  • A joint optimization objective ensures both semantic representation and temporal consistency for better interpretations.
  • The framework supports complex text analysis tasks, enriching the research on dynamic semantic patterns.

Cite This Study

Wu et al. (2025) studied this question.

synapsesocial.com/papers/68f04920e559138a1a06d78ahttps://doi.org/10.21203/rs.3.rs-7831142/v1
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1A temporal adaptive dictionary-constrained LDA and Bi-calibrated dual granularity DTM framework for dynamic topic evolution analysis in academic papers2026 · 1 citations
  2. 2Dynamic Manifold Evolution Theory: Modeling and Stability Analysis of Latent Representations in Large Language Models2025
  3. 3Modeling Dynamic Topics in Chain-Free Fashion by Evolution-Tracking Contrastive Learning and Unassociated Word Exclusion2024
  4. 4Enhancing Contextual Understanding in Large Language Models with Dynamic Dependency Structures: A Methodological Approach2024
  5. 5Topic Modeling for Evolving Textual Data Using LDA, HDP, NMF, BERTOPIC, and DTM With a Focus on Research Papers2024 · 2 citations