DualEval jointly calibrates LLM abilities and item difficulties/sharpness in a shared latent space using static labels and reward-model scores to unify benchmark and arena-style evaluation.
arXiv preprint arXiv:2510.14966 , year=
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
DualEval: Joint Model-Item Calibration for Unified LLM Evaluation
DualEval jointly calibrates LLM abilities and item difficulties/sharpness in a shared latent space using static labels and reward-model scores to unify benchmark and arena-style evaluation.