LLMbench supplies a side-by-side workbench with probability overlays and variation modes that treat LLM output as a research object whose counterfactual history is made legible.
Different variants of the same text, whether produced by different models or by the same model at different temperatures, are analytically productive to explore
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CY 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
LLMbench: A Comparative Close Reading Workbench for Large Language Models
LLMbench supplies a side-by-side workbench with probability overlays and variation modes that treat LLM output as a research object whose counterfactual history is made legible.