guyulongcs/Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising
A curated map of how tech giants actually rank stuff
A living bibliography that traces deep learning for search, recommendations, and ads from Word2vec to LLM-based ranking, with a bias for papers that shipped at scale.
Not currently ranked — collecting fresh signals.
star history
What it does This repository is a curated, chronologically organized reading list of deep learning papers for industrial search, recommendation, and advertising systems. It covers the full pipeline: embedding, matching, pre-ranking, CTR/CVR prediction, post-ranking, relevance, reinforcement learning, and increasingly, LLM-based ranking. Each entry links directly to PDFs hosted in the repo.
The interesting bit The curation has a clear industrial slant. Papers from Google, Alibaba, Amazon, Baidu, Airbnb, and Pinterest are marked with asterisks, and the maintainer’s own 2026 TOIS survey paper frames the collection. This is less “comprehensive literature review” and more “what actually worked at billion-user scale.”
Key highlights
- Covers 10+ sub-areas from embeddings to RL, with explicit pipeline stages (matching → ranking → post-ranking)
- Heavy representation from Chinese tech giants (Alibaba, Baidu, Tencent) alongside US counterparts
- Includes foundational methods (Word2vec, DeepWalk, matrix factorization) alongside recent LLM ranking work
- Direct PDF links mean no chasing paywalls for included papers
- Explicitly designed to support citation of the maintainer’s TOIS 2026 overview paper
Caveats
- README is a flat list with minimal annotation; you need to know the field to navigate it
- PDF hosting in-repo may raise copyright questions for some papers
- No code or reproductions included — pure bibliography
Verdict Worth bookmarking if you build or research production ranking systems and want a shortcut to the papers that mattered industrially. Skip it if you need tutorials, implementations, or theoretical depth beyond applied deep learning.
Frequently asked
- What is guyulongcs/Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising?
- A living bibliography that traces deep learning for search, recommendations, and ads from Word2vec to LLM-based ranking, with a bias for papers that shipped at scale.
- Is Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising open source?
- Yes — guyulongcs/Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising is an open-source project tracked on heatdrop.
- What language is Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising written in?
- guyulongcs/Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising is primarily written in Python.
- How popular is Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising?
- guyulongcs/Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising has 2.6k stars on GitHub.
- Where can I find Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising?
- guyulongcs/Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising is on GitHub at https://github.com/guyulongcs/Awesome-Deep-Learning-Papers-for-Search-Recommendation-Advertising.