Research map / Learning theory and algorithms / Submodular and game algorithms
Dynamic User Profiling for Streams of Short Texts
Shangsong Liang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use UET topic model (dynamic, one topic per short document) with SKDA instead of best baseline topic model GSDMM | SPA · MGT monthly | Pre@10: .339 -> .365 | Improved |
| UET vs GSDMM | SPA · AGT monthly | NDCG@10: .257 -> .282 | Improved |
| UET vs GSDMM | SPA · MGT monthly | MRR@10: .755 -> .813 | Improved |
| UET vs GSDMM on semantic MAP | SPA · MGT monthly | MAP-S@10: .254 -> .262 | Improved |
| one topic per document sampling (SPA, GSDMM) instead of multiple topics per document (LDA etc.) | topic-model profiling · AGT monthly | Pre@10: .239 -> .275 (LDA vs GSDMM) | Improved |
| use topic models instead of tfidf/PLM | user profiling · AGT monthly | Pre@10: .216 -> .239 (tfidf vs LDA) | Improved |
| increase time period length from week to month (ablation) | SPA and baselines · AGT | standard metrics | Improved |
| increase time period beyond a month (quarter to year) (ablation) | SPA · AGT | standard metrics | No clear effect |
| UET vs baseline topic models for generalization | UET · Twitter test data | perplexity | Improved |
Closest papers by meaning
- FedSum: Data-Efficient Federated Learning Under Data Scarcity Scenario for Text Summarization (AAAI 2025 · similarity 0.80)
- PersonalSum: A User-Subjective Guided Personalized Summarization Dataset for Large Language Models (NeurIPS 2024 · similarity 0.79)
- Streaming Stochastic Submodular Maximization with On-Demand User Requests (NeurIPS 2025 · similarity 0.79)
- Physics of Language Models: Part 3.1, Knowledge Storage and Extraction (ICML 2024 · similarity 0.78)
- UDA: A Benchmark Suite for Retrieval Augmented Generation in Real-World Document Analysis (NeurIPS 2024 · similarity 0.78)
- HYDRA: Model Factorization Framework for Black-Box LLM Personalization (NeurIPS 2024 · similarity 0.78)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
