LLMs exhibit a persistent comprehension-execution gap in end-to-end mathematical modeling tasks, with a new stage-wise framework showing better alignment to human expert judgments than prior schemes.
Conversions are performed on a per-page basis to retain layout cues and to avoid cross-page con- catenation errors
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
How Far Are We? Systematic Evaluation of LLMs vs. Human Experts in Mathematical Contest in Modeling
LLMs exhibit a persistent comprehension-execution gap in end-to-end mathematical modeling tasks, with a new stage-wise framework showing better alignment to human expert judgments than prior schemes.