Metraipolis › archive › dc_be7399ab4bc6
A Reality Check of Inference-Time Context Augmentation Methods — arXiv:2610.30983
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.30983
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-25T08:36:24.047Z
- SHA-256 of the archived text
34f28288ddd8ebb00547e19a276cbf36a02fdc49d07a3738dcf507de80549d44- Stable URL of this page
https://metraipolis.com/documento/dc_be7399ab4bc6
How to cite
A Reality Check of Inference-Time Context Augmentation Methods — arXiv:2610.30983 — arXiv (en), https://arxiv.org/abs/2610.30983 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_be7399ab4bc6 (Metraipolis archive dc_be7399ab4bc6, sha256 34f28288ddd8ebb00547e19a276cbf36a02fdc49d07a3738dcf507de80549d44)
BibTeX
@misc{dcbe7399ab4bc6,
title = {A Reality Check of Inference-Time Context Augmentation Methods — arXiv:2610.30983},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.30983},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_be7399ab4bc6; sha256 34f28288ddd8ebb00547e19a276cbf36a02fdc49d07a3738dcf507de80549d44},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.30983. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large language models (LLMs) often fail to utilise the context given in the input, resulting in hallucinations or a fallback to parametric knowledge. Context augmentation methods aim to improve model performance by modifying the input, from inserting filler tokens that add computational time to adding relevant content such as elaboration of a context during inference. However, these methods have been evaluated in isolation and only by their performance gains, leaving it unclear how they compare and whether they reliably improve performance. In this paper, we present an evaluation framework spanning five methods that differ in how informative the inserted content is and apply them at the sentence level across seven models spanning multiple sizes on three datasets. We find that accuracy gains are not always genuine improvements but undirected perturbations: the methods correct some predictions while breaking others, particularly when the inserted text allows only for more computational time. Our analysis reveals that methods that bring no additional content primarily change model predictions near the decision boundary. Our findings highlight the need to evaluate the reliability of the accuracy gains before adopting them in practice. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2610.30983 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-25T08:31:27.779Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).