18 [1] 山田健太, 青田雅輝 (2024). マルチモーダルな深層学習手法を用いた政治資金収支報告書の判読 の試み. 2024年度日本選挙学会総会・研究会. [2] 大村和正, 白井穂乃, 石原祥太郎, 澤紀彦 (2023). 極性と重要度を考慮した決算短信からの業績要 因文の抽出. 言語処理学会第29回年次大会発表論文集. [3] Shotaro Ishihara, Hiromu Takahashi, and Hono Shirai (2023). Quantifying Diachronic Language Change via Word Embeddings: Analysis of Social Events using 11 Years News Articles in Japanese and English. 9th International Conference on Computational Social Science.
[7] • ユーザ入力画像を用いた記事推薦 [8] • 生成的推薦による人気バイアスの分析 [9] 提供:閲覧数以外も考慮した推薦システム 22 [7] Atom Sonoda, Fujio Toriumi, and Hiroto Nakajima (2024). User Experiments on the Effect of the Diversity of Consumption on News Services. IEEE Access. [8] Kota Tanabe, Shotaro Ishihara, Kenta Yamada, Masaki Aota, and Yasutsuna Matayoshi (2025). Making News Familiar: News Recommendation from Daily Scenery. Proceedings of the KES2025.
[10] Norihiko Sawa (2020). Test headlines on News Media by Multi-Armed Bandit: Case Study of Multi-Armed Bandit to raise CTR of Articles. Computation + Journalism Symposium 2020.
Models: A Survey. Proceedings of the TrustNLP 2023. コーパスの前処理,事前学習の工夫,モデルの後処理など[13] Defense: Training Defense: Pre-processing Defense: Post-processing data deduplication data sanitization regularization differential privacy filterling confidence masking knowledge distillation 50 どう対応するべき?