long-tail

SevDiff is a severity-conditioned diffusion model for generating vehicle conflict trajectories with controlled time-to-collision values, achieving high hit-rate on a real-world dataset for ADAS evaluation.

This paper presents a training-free LLM-based candidate generation pipeline for vacation rental marketplaces, using an off-the-shelf LLM to synthesize semantic queries and dense retrieval to complement collaborative filtering, significantly improving coverage for long-tail properties while maintaining performance on well-served ones.

This paper studies the fairness problem of thresholded subgroup underdiagnosis in long-tailed chest X-ray classification, demonstrating that rare-label fairness depends jointly on the finding, subgroup, and operating threshold, not on label frequency or ranking metrics alone.

This paper proposes a class-frequency guided noise schedule for diffusion models that assigns larger-scale noises to low-frequency classes to improve generation quality on imbalanced datasets, demonstrating substantial improvements over baselines.

In a tweet, Sarah Hooker argues that GPUs are ill-suited for the long-tail distribution of real-world data, suggesting a need for alternative AI hardware.

This paper introduces a distribution-aware reinforcement learning framework that enhances MLLM performance in long-tailed numerical regression tasks using batch-level comparison-based supervision.