Context-Aware Search and Retrieval Under Token Erasure
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866913048149622784 |
|---|---|
| author | Ghasvarianjahromi, Sara Barr, Joshua Yakimenka, Yauhen Kliewer, Jörg |
| author_facet | Ghasvarianjahromi, Sara Barr, Joshua Yakimenka, Yauhen Kliewer, Jörg |
| contents | This paper introduces and analyzes a search and retrieval model for RAG-like systems under {token} erasures. We provide an information-theoretic analysis of remote document retrieval when query representations are only partially preserved. The query is represented using term-frequency-based features, and semantically adaptive redundancy is assigned according to feature importance. Retrieval is performed using TF-IDF-weighted similarity. We characterize the retrieval error probability by showing that the vector of similarity margins converges to a multivariate Gaussian distribution, yielding an explicit approximation and computable upper bounds. Numerical results support the analysis, while a separate data-driven evaluation using embedding-based retrieval on real-world data shows that the same importance-aware redundancy principles extend to modern retrieval pipelines. Overall, the results show that assigning higher redundancy to semantically important query features improves retrieval reliability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_18424 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Context-Aware Search and Retrieval Under Token Erasure Ghasvarianjahromi, Sara Barr, Joshua Yakimenka, Yauhen Kliewer, Jörg Information Retrieval Information Theory This paper introduces and analyzes a search and retrieval model for RAG-like systems under {token} erasures. We provide an information-theoretic analysis of remote document retrieval when query representations are only partially preserved. The query is represented using term-frequency-based features, and semantically adaptive redundancy is assigned according to feature importance. Retrieval is performed using TF-IDF-weighted similarity. We characterize the retrieval error probability by showing that the vector of similarity margins converges to a multivariate Gaussian distribution, yielding an explicit approximation and computable upper bounds. Numerical results support the analysis, while a separate data-driven evaluation using embedding-based retrieval on real-world data shows that the same importance-aware redundancy principles extend to modern retrieval pipelines. Overall, the results show that assigning higher redundancy to semantically important query features improves retrieval reliability. |
| title | Context-Aware Search and Retrieval Under Token Erasure |
| topic | Information Retrieval Information Theory |
| url | https://arxiv.org/abs/2604.18424 |