← Search

Heecheol Yun

2 accepted papers

2026

When to Ensemble: Identifying Token-Level Points for Stable and Fast LLM Ensembling

ICLR 2026poster

Ensembling Large Language Models (LLMs) has gained attention as a promising approach to surpass the performance of individual models by leveraging their complementary strengths. In particular, aggregating models’ next-token probability distributions to select the next token has been shown to be effe…

Cited by 0SourceScholar
2024

PruNeRF: Segment-Centric Dataset Pruning via 3D Spatial Consistency

ICML 2024poster

Neural Radiance Fields (NeRF) have shown remarkable performance in learning 3D scenes. However, NeRF exhibits vulnerability when confronted with distractors in the training images -- unexpected objects are present only within specific views, such as moving entities like pedestrians or birds. Excludi…

Cited by 0SourcePDFScholar