ACL Findings2026
An LLM-Embedding Semantic Adaptation Network for Post-level Semantic Drift Evaluation
Ning Chen, Mingyu Kang, Jie Li, Linyuan Lü
Abstract
Evaluating semantic drift is essential for understanding dynamical discourse evolution and opinion formation in online discussions. However, sparse and uneven distributions of eventspecific keywords prevent traditional models from capturing post-level semantic drift. Thus, to address this issue, we propose an LLM-embedding Semantic Adaptation Network (LLM-SAN), which is a hybrid semantic drift evaluation model with an LLM-Embedding gated recurrent unit (GRU) module, an LLM-Embedding graph convolutional network (GCN) module and a multi-expert adaptive fusion module. The GRU module is used to extract features from event related posts, and The GCN is used to extract features from temporal graphical topic posts. Then, the features are merged by the multi-expert adaptive fusion module. Finally, this module predicts the future post embedding, and the prediction error is used to evaluate and detect the semantic drift points. Extensive experiments are conducted, and the results show that LLM-SAN achieves the state-of-the-art performance on the semantic drift evaluation task, compared to the other baselines. Ablation experiments are also conducted to show the effectiveness of each module in LLM-SAN. Code is available at: https://github.com/mengtahua/LLMSAN.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
- Dynamic Structured Neural Topic Model with Self-Attention Mechanism cites Dynamic Topic Models for Temporal Document Networks
- DynaMiTE: Discovering Explosive Topic Evolutions with User Guidance cites Dynamic Topic Models for Temporal Document Networks
Builds on5
- Dynamic Topic Models for Temporal Document NetworksDelvin Ce Zhang, Hady W. LauwICML 2022 · 26 citations
- Unsupervised Multi-Granularity SummarizationMing Zhong, Yang Liu, Suyu Ge, Yuning Mao et al.EMNLP Findings 2022 · 12 citations
- Dynamic Structured Neural Topic Model with Self-Attention MechanismNozomu Miyamoto, Masaru Isonuma, Sho Takase, Junichiro Mori et al.ACL Findings 2023 · 10 citations
- Event Ontology Completion with Hierarchical Structure Evolution NetworksPengfei Cao, Yupu Hao, Yubo Chen, Kang Liu et al.EMNLP 2023 · 3 citations
- DynaMiTE: Discovering Explosive Topic Evolutions with User GuidanceNishant Balepur, Shivam Agarwal, Karthik Venkat Ramanan, Susik Yoon et al.ACL Findings 2023 · 2 citations
Related papers
- Integrating Semantic and Structural Information with Graph Convolutional Network for Controversy DetectionLei Zhong, Juan Cao, Qiang Sheng, Junbo Guo et al.ACL 2020 · 26 citations
- Semantic Evolvement Enhanced Graph Autoencoder for Rumor DetectionXiang Tao, Liang Wang, Qiang Liu, Shu Wu et al.WWW 2024 · 20 citations
- Characterizing and Measuring Linguistic Dataset DriftTyler A. Chang, Kishaloy Halder, Neha Anna John, Yogarshi Vyas et al.ACL 2023 · 2 citations
- METER: A Dynamic Concept Adaptation Framework for Online Anomaly DetectionJiaqi Zhu, Shaofeng Cai, Fang Deng, Beng Chin Ooi et al.VLDB 2024 · 18 citations
- Revealing COVID-19's Social Dynamics: Diachronic Semantic Analysis of Vaccine and Symptom Discourse on TwitterZeqiang Wang, Jiageng Wu, Yuqi Wang, Wei Wang et al.EMNLP Findings 2024 · 1 citation
