Bomin Chuang

Bomin ChuangPersonal website of Bomin Chuang, an R&D engineer building reliable AI with RAG and text embeddings.https://bominc0624.dev/[Publication] DualCSE: 文の明示的・暗黙的意味の埋め込み学習https://bominc0624.dev/publications/dualcse/https://bominc0624.dev/publications/dualcse/Published in 言語処理学会第32回年次大会, pp. 3761–3766Sun, 01 Mar 2026 00:00:00 GMT[Publication] One Sentence, Two Embeddings: Contrastive Learning of Explicit and Implicit Semantic Representationshttps://bominc0624.dev/publications/one-sentence-two-embeddings/https://bominc0624.dev/publications/one-sentence-two-embeddings/Published in Findings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2026), short paper, pp.1444-1452Sun, 01 Mar 2026 00:00:00 GMTEncoder-based言語モデルによるテキスト生成：RoBERTa Diffusionの実装と検証https://bominc0624.dev/posts/roberta-diffusion/https://bominc0624.dev/posts/roberta-diffusion/RoBERTaのMasked Language Modelingを用いたテキスト拡散モデルの実装と、per-batch・per-sampleマスク戦略の比較。Mon, 22 Dec 2025 00:00:00 GMT[Publication] 文の明示的・暗黙的な意味に対する埋め込みの同一空間上での表現https://bominc0624.dev/publications/explicit-implicit-embeddings-yans2025/https://bominc0624.dev/publications/explicit-implicit-embeddings-yans2025/Published in 第20回言語処理若手シンポジウム (YANS2025)Fri, 19 Sep 2025 00:00:00 GMT2024年の振り返りhttps://bominc0624.dev/posts/2024-review/https://bominc0624.dev/posts/2024-review/2024年の出来事と、東京で社会人として過ごした一年の振り返り。Thu, 30 Jan 2025 00:00:00 GMT[Publication] Identification of Opinion and Ground in Customer Review Using Heterogeneous Datasetshttps://bominc0624.dev/publications/identification-of-opinion-and-ground/https://bominc0624.dev/publications/identification-of-opinion-and-ground/Published in Proceedings of 16th International Conference on Agents and Artificial Intelligence (ICAART 2024), pp. 68-78Thu, 01 Feb 2024 00:00:00 GMTgzip is all you need (gzip + kNN outperforms Transformers)https://bominc0624.dev/posts/gzip-knn-text-classification/https://bominc0624.dev/posts/gzip-knn-text-classification/Notes on parameter-free text classification with gzip compression and k-nearest neighbors.Tue, 15 Aug 2023 00:00:00 GMTExploiting Cloze Questions for Few Shot Text Classification and Natural Language Inferencehttps://bominc0624.dev/posts/pattern-exploiting-training/https://bominc0624.dev/posts/pattern-exploiting-training/A reading note on Pattern-Exploiting Training (PET), a semi-supervised prompt-based method for few-shot text classification.Tue, 11 Jul 2023 00:00:00 GMTBreadth-First Search (using Python)https://bominc0624.dev/posts/breadth-first-search/https://bominc0624.dev/posts/breadth-first-search/An introduction to breadth-first search, with a Python implementation, complexity analysis, and traversal example.Sun, 09 Jul 2023 00:00:00 GMTSystem Design - YouTubehttps://bominc0624.dev/posts/system-design-youtube/https://bominc0624.dev/posts/system-design-youtube/Notes on designing a large-scale video platform, covering requirements, capacity estimation, APIs, storage, and scalability.Sun, 09 Apr 2023 00:00:00 GMT