# Literature verification notes

Canonical bibliography support for Series I / Paper 03.

Verification date: 2026-08-14.

Primary publication sources checked where available.

1. Joachims et al. 2005. Accurately Interpreting Clickthrough Data as Implicit Feedback. SIGIR 2005, 154-161.
2. Craswell et al. 2008. An Experimental Comparison of Click Position-Bias Models. WSDM 2008, 87-94. DOI 10.1145/1341531.1341545.
3. Ilievski et al. 2018. Systematic Study of Long Tail Phenomena in Entity Linking. COLING 2018.
4. Hoveyda et al. 2024. Real World Conversational Entity Linking Requires More Than Zero-Shots. Findings ACL 2024. DOI 10.18653/v1/2024.findings-acl.829.
5. Kim et al. 2024. DynamicER: Resolving Emerging Mentions to Dynamic Entities for RAG. EMNLP 2024. DOI 10.18653/v1/2024.emnlp-main.762.
6. Aggarwal et al. 2024. GEO: Generative Engine Optimization. KDD 2024, 5-16. DOI 10.1145/3637528.3671900.
7. Vollmers et al. 2025. Contextual Augmentation for Entity Linking using Large Language Models. COLING 2025.
8. Wu et al. 2025. WebWalker: Benchmarking LLMs in Web Traversal. ACL 2025. DOI 10.18653/v1/2025.acl-long.508.
9. Deshmukh et al. 2025. All Entities are Not Created Equal: Examining the Long Tail for Ultra-Fine Entity Typing. *SEM 2025. DOI 10.18653/v1/2025.starsem-1.15.
10. Ghonim et al. 2025. RAED: Retrieval-Augmented Entity Description Generation for Emerging Entity Linking and Disambiguation. EMNLP 2025. DOI 10.18653/v1/2025.emnlp-main.1746.
11. Li et al. 2025. Leveraging the Power of Large Language Models in Entity Linking via Adaptive Routing and Targeted Reasoning. EMNLP Industry 2025. DOI 10.18653/v1/2025.emnlp-industry.59.
12. Sawada et al. 2026. entity-linkings: A Unified Library for Entity Linking. EACL 2026. DOI 10.18653/v1/2026.eacl-demo.42.

Boundary note:
- The paper's "Machine-First Recognition Regime" and "Latent Machine Recognizability" are proposed theoretical constructs.
- Existing literature supports the separability of search exposure, long-tail machine resolution, retrieval-based entity knowledge, and user attention, but does not by itself establish the full PPSS decoupling model.