Research map / Image generation / Text-to-image diffusion
TemPEST: Soft Template-Based Personalized EDM Subject Generation through Collaborative Summarization
Yu-Hsiu Chen, Pin-Yu Chen, Hong-Han Shuai, Wen-Chih Peng
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 TemPEST instead of BiSET | summarization · KKday | BLEU-1: 22.06 -> 24.36 | Improved |
| use TemPEST instead of BiSET | summarization · KKday | BLEU-4: 1.52 -> 1.97 | Improved |
| use TemPEST instead of BiSET | summarization · KKday | ROUGE-1 F1: 55.54 -> 65.47 | Improved |
| use TemPEST instead of BiSET | summarization · KKday | ROUGE-L F1: 8.70 -> 8.33 | Hurt |
| add selective template mechanism (BiSET) over S2S+Att | summarization · KKday | ROUGE-L F1: 5.65 -> 8.70 | Improved |
| use TemPEST instead of BiSET for summary similarity | summarization · KKday | summary similarity: 0.580 -> 0.592 | Improved |
| use TemPEST instead of BiSET for user similarity | summarization · KKday | user similarity: 0.393 -> 0.456 | Improved |
| use BiSET instead of S2S+Att | summarization · KKday | user similarity: 0.409 -> 0.393 | Hurt |
| use TemPEST user-specific article representation instead of CDAE | recommendation · KKday | Recall@10: 0.0629 -> 0.1921 | Improved |
| use TemPEST user-specific article representation instead of GATE embeddings | recommendation · KKday | Recall@10: 0.1871 -> 0.1921 | Improved |
| use TemPEST instead of GATE | recommendation · KKday | NDCG@10: 0.0619 -> 0.0632 | Improved |
| user-based TemPEST subjects instead of BiSET subjects | user study · KKday, 81 users | preference share: 61.4% vs 38.6% | Improved |
Closest papers by meaning
- Towards Personalized Review Summarization via User-Aware Sequence Network (AAAI 2019 · similarity 0.84)
- PersonalSum: A User-Subjective Guided Personalized Summarization Dataset for Large Language Models (NeurIPS 2024 · similarity 0.82)
- Principled Fine-tuning of LLMs from User-Edits: A Medley of Preference, Supervision, and Reward (NeurIPS 2025 · similarity 0.81)
- Rethinking Personalized Generation: Test-time Alignment via Factorized Ranking Models (NeurIPS 2026 · similarity 0.80)
- PersonalLLM: Tailoring LLMs to Individual Preferences (ICLR 2025 · similarity 0.80)
- Towards Unified Multi-Modal Personalization: Large Vision-Language Models for Generative Recommendation and Beyond (ICLR 2024 · similarity 0.80)
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.
