noisy-labels

Tag

Cards List
#noisy-labels

Reliability-Aware LLM Alignment from Inconsistent Human Feedback

arXiv cs.AI · 2026-07-24 Cached

Proposes Reliability-Guided Preference Optimization (RGPO) to handle inconsistent human feedback in LLM alignment by estimating annotator reliability and dynamically modulating training based on consensus, achieving superior performance over standard RLHF methods.

0 favorites 0 likes
#noisy-labels

Design-Based Supervised Learning with Noisy Human Labels

arXiv cs.AI · 2026-07-20 Cached

Proposes Partially Adjudicated Design-Based Supervised Learning (PA-DSL), a method that corrects noisy human labels using a small set of adjudicated cases to debias automated classifiers, achieving nominal coverage and reducing RMSE by 10-17% in experiments.

0 favorites 0 likes
#noisy-labels

Metadata-Free Meta-Reweighted Direct Preference Optimization under Noisy Preference Labels

arXiv cs.LG · 2026-07-14 Cached

This paper proposes a bilevel optimization framework for Direct Preference Optimization under noisy preference labels, introducing a metadata-free meta-reweighting method that uses central-difference approximation and LoRA fine-tuning to improve alignment performance.

0 favorites 0 likes
#noisy-labels

MOLAR: Learning Multimodal Molecular Representations from Noisy Labels

arXiv cs.LG · 2026-06-18 Cached

MOLAR proposes a noise-aware framework for learning multimodal molecular representations from noisy labels by separating clean-property inference from observed label noise, outperforming baselines on molecular benchmarks.

0 favorites 0 likes
#noisy-labels

Symmetrization of Loss Functions for Robust Training of Neural Networks in the Presence of Noisy Labels

arXiv cs.LG · 2026-05-21 Cached

This paper studies symmetrization of loss functions for robust training under label noise, introducing SGCE and alpha-MAE loss functions that interpolate between multi-class unhinged loss and Mean Absolute Error, with theoretical guarantees and competitive empirical performance.

0 favorites 0 likes
← Back to home

Submit Feedback