Tag
This paper introduces a dataset for monster-level prediction in Pathfinder Second Edition and compares machine learning models for ordinal regression, showing tree-based ensembles achieve near-perfect ranking accuracy.
ConOrd proposes a contrastive learning framework for ordinal regression that integrates contrastive learning and order learning, achieving state-of-the-art performance on facial age estimation, image quality assessment, and video quality assessment.
This paper introduces LOPA, a lightweight framework for spoken language assessment that uses latent ordinal prototype alignment and semantic-anchored layer routing on a frozen Whisper encoder, achieving performance comparable to billion-parameter models without LLM fine-tuning.
This paper proposes a probabilistic framework for Alzheimer's disease progression forecasting that combines ordinal diagnosis prediction, multi-horizon trajectory generation, and decomposed uncertainty estimation using a Temporal Fusion Transformer encoder and an autoregressive Mixture Density Network. The model outperforms baselines on ADNI data, achieving near-nominal 90% credible interval coverage with clinically meaningful uncertainty signals.
DiffoR proposes a novel continuous generative framework for ordinal regression using diffusion models, overcoming limitations of discrete methods. Extensive experiments on 12 benchmarks demonstrate state-of-the-art performance across four domains.