LLM popularity judgments align more closely with pretraining data exposure counts than with Wikipedia popularity, with stronger effects in pairwise comparisons and larger models.
Title resolution pending
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
years
2026 8verdicts
UNVERDICTED 8roles
background 2polarities
background 2representative citing papers
A structural reduction to local medians enables new constant-round MPC (2-α) and (2-ζ) approximation algorithms for 1-median rank aggregation under multiple distances, plus an improved 1.968-approximation for Ulam that extends to weighted cases.
Introduces claim-conditioned re-scoring (SIFT) and warranted supports proportion (WSP) metric, reporting accuracy recovery up to 27.6 points and WSP calibration at AUC 0.92 on FEVER, SciFact and other benchmarks.
RBFN projection heads serve as competitive replacements for MLP heads in SSL and enable SNS, a label-free metric from RBF parameters that correlates strongly with logistic regression evaluation.
Neuron-level inference-time intervention reduces multiple biases in reward models, enabling 2B and 7B models to match 70B performance on LLM alignment benchmarks without trade-offs.
PREFAB applies preference learning grounded in the peak-end rule to let users annotate only key affective change segments while interpolating the rest, reducing workload and improving confidence in a 25-participant study.
Higher sequence probability predicts correctness across different answers in a dataset but does not reliably improve accuracy when decoding methods or hyperparameters are changed, nor does it indicate correctness for repeated responses to one prompt.
Kernel ridge regression combined with mRMR feature selection improves prediction of full benchmark scores from question subsets over existing efficient benchmarking techniques.
citing papers explorer
-
Pretraining Exposure Explains Popularity Judgments in Large Language Models
LLM popularity judgments align more closely with pretraining data exposure counts than with Wikipedia popularity, with stronger effects in pairwise comparisons and larger models.
-
The Warrant Gap: Claim-Conditioned Re-scoring for Fact-Checking
Introduces claim-conditioned re-scoring (SIFT) and warranted supports proportion (WSP) metric, reporting accuracy recovery up to 27.6 points and WSP calibration at AUC 0.92 on FEVER, SciFact and other benchmarks.
-
Debiasing Reward Models via Causally Motivated Inference-Time Intervention
Neuron-level inference-time intervention reduces multiple biases in reward models, enabling 2B and 7B models to match 70B performance on LLM alignment benchmarks without trade-offs.