LLM pairwise evaluation is recast as low-rank tensor completion, yielding semiparametric efficient estimators and asymptotic normality for ability functionals via a score-whitening correction for anisotropic operators.
arXiv preprint arXiv:2509.01847 , year=
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
years
2026 5representative citing papers
Low-rank matrix modeling with cross-fitted debiased estimators and multiplier bootstrap yields stable task-specific LLM rankings and asymptotically valid simultaneous confidence sets from sparse pairwise data.
A learnable continuous perturbation framework for LLM token prefixes via latent vector transformations, optimized through unbiased estimating equations, yields gains in out-of-domain performance.
HJA ranking separates consensus ranking, judge sensitivity, and residual disagreement as distinct inferential targets with identifiability conditions and an anchored alternating algorithm, yielding better recovery and uncertainty calibration than pooled baselines on synthetic and real data.
Perturbing the prefix before next-token prediction, during both training and inference, improves out-of-distribution language-model generation and yields a conditional extrapolation guarantee.
citing papers explorer
-
LLM Evaluation as Tensor Completion: Low Rank Structure and Semiparametric Efficiency
LLM pairwise evaluation is recast as low-rank tensor completion, yielding semiparametric efficient estimators and asymptotic normality for ability functionals via a score-whitening correction for anisotropic operators.
-
Low Rank for Rank: Uncertainty-Aware Task-Specific LLM Ranking under Sparse Pairwise Comparisons
Low-rank matrix modeling with cross-fitted debiased estimators and multiplier bootstrap yields stable task-specific LLM rankings and asymptotically valid simultaneous confidence sets from sparse pairwise data.
-
Learning Perturbations to Extrapolate Your LLM
A learnable continuous perturbation framework for LLM token prefixes via latent vector transformations, optimized through unbiased estimating equations, yields gains in out-of-domain performance.
-
Heterogeneous Judge-Aware Ranking with Sensitivity, Disagreement, and Confidence
HJA ranking separates consensus ranking, judge sensitivity, and residual disagreement as distinct inferential targets with identifiability conditions and an anchored alternating algorithm, yielding better recovery and uncertainty calibration than pooled baselines on synthetic and real data.
-
Perturbation is All You Need for Extrapolating Language Models
Perturbing the prefix before next-token prediction, during both training and inference, improves out-of-distribution language-model generation and yields a conditional extrapolation guarantee.