Citation Analysis

Fake Date Tests: \\ Can We Trust In-sample Accuracy of \\LLMs in Macroeconomic Forecasting?
Alexander Eliseev, Sergei Seleznev
https://arxiv.org/abs/2601.07992
4
Citation mentions
4
Cited references
10
Sections
10,517
Words (approx)

References by Citation Intensity

Ordered by composite index (descending). Higher values indicate more intensive citation.

# Reference Year Mentions Breadth Sec. Wtd Share Composite Main %
1 Korinek, Anton 2023 1 1 1.0 0.250 0.406 100%
2 Korinek, Anton 2024 1 1 1.0 0.250 0.406 100%
3 Korinek, Anton 2025 1 1 1.0 0.250 0.406 100%
4 DeepSeek-AI 2025 1 1 2.0 0.250 0.406 100%
Measures: Mentions = total in-text citations; Breadth = distinct sections; Sec. Wtd = section-weighted count (body ×2, lit review/appendix ×0.5); Share = mentions / total citations in paper; Composite = geometric mean of normalised count, breadth, and main-text ratio; Main % = fraction of mentions in main text (excl. appendix). (self) = self-citation.