Lune

Profiles2024-2026

Jaehyung Kim

Yonsei University

18Papers
10Last author
15Unique co-authors
1First author
0Solo

First and last author come from byline position, which some venues assign alphabetically.

Recent papers18

  1. Enhancing Instruction Following of LLMs via Activation Steering with Dynamic RejectionAuthor 2 of 2ICLR 2026 · 6 citations
  2. Align to Misalign: Automatic LLM Jailbreak with Meta-Optimized LLM JudgesAuthor 3 of 3ICLR 2026 · 4 citations
  3. RPM: Reasoning-Level Personalization for Black-Box Large Language ModelsAuthor 4 of 5ICLR 2026 · 3 citations
  4. TiTok: Transfer Token-level Knowledge via Contrastive Excess to Transplant LoRAAuthor 2 of 2ICLR 2026 · 2 citations
  5. Gap-K%: Measuring Top-1 Prediction Gap for Detecting Pretraining DataAuthor 2 of 2ACL 2026 · 1 citation
  6. Embracing Anisotropy: Turning Massive Activations into Interpretable Control Knobs for Large Language ModelsAuthor 3 of 3ACL 2026
  7. EMCEE: Improving Multilingual Capability of LLMs via Bridging Knowledge and Reasoning with Extracted Synthetic Multilingual ContextAuthor 2 of 2ACL 2026
  8. Learning from the Undesirable: Robust Adaptation of Language Models Without ForgettingAuthor 2 of 3AAAI 2026
  9. Prior-based Noisy Text Data Filtering: Fast and Strong Alternative For PerplexityAuthor 3 of 4ICLR 2026
  10. Revisit What You See: Revealing Visual Semantics in Vision Tokens to Guide LVLM DecodingAuthor 2 of 2ACL 2026
  11. Revisiting the Uniform Information Density Hypothesis in LLM ReasoningAuthor 3 of 3ACL Findings 2026
  12. Fast and Fluent Diffusion Language Models via Convolutional Decoding and Rejective Fine-tuningAuthor 3 of 4NeurIPS 2025 · 8 citations
  13. Improving Chemical Understanding of LLMs via SMILES ParsingAuthor 2 of 3EMNLP 2025 · 2 citations
  14. Debiasing Online Preference Learning via Preference Feature PreservationAuthor 4 of 6ACL Findings 2025
  15. Personalized Language Models via Privacy-Preserving Evolutionary Model MergingAuthor 3 of 3EMNLP 2025
  16. Spread Preference Annotation: Direct Preference Judgment for Efficient LLM AlignmentAuthor 4 of 4ICLR 2025
  17. Structural Reasoning Improves Molecular Understanding of LLMAuthor 2 of 3ACL 2025
  18. Learning to Correct for QA Reasoning with Black-box LLMsAuthor 1 of 4EMNLP 2024
Dusk over the sea between two cliffs drawn in fine vertical lines