Introduces a decomposition of LLM token efficiency into completion rate, conditional correctness given completion, and normalized length to distinguish logic-limited, context-limited, and verbosity-limited failure modes across benchmarks.
Title resolution pending
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
-
Beyond Accuracy: Decomposing the Reasoning Efficiency of LLMs
Introduces a decomposition of LLM token efficiency into completion rate, conditional correctness given completion, and normalized length to distinguish logic-limited, context-limited, and verbosity-limited failure modes across benchmarks.