<aside> 📌 이번 차수 계획

일시: 2025년 3~5월 매주 목요일 오후 6시 장소: 한양대학교 IT/BT관 506호 발표자: 김강산, 윤예진, 서동건, 이정연, 김창현, 김지수, 신영우, 서기정, 손유리, 이휘영, 김민서, 김승희, 임혜림, 황의지 각 1회 발표 방법: 자료는 영어로 작성, 발표는 한국어/영어 자유 발표 주제: 아래 지정 논문

</aside>

Papers to be presented

  1. s1: Simple test-time scaling
  2. INFERENCE SCALING LAWS: AN EMPIRICAL ANALYSIS OF COMPUTE-OPTIMAL INFERENCE FOR LLM PROBLEM-SOLVING
  3. Towards Large Reasoning Models: A Survey of Reinforced Reasoning with Large Language Models
  4. S2R: Teaching LLMs to Self-verify and Self-correct via Reinforcement Learning
  5. LM2: Large Memory Models
  6. Towards an AI co-scientist
  7. Large Language Diffusion Models
  8. The Danger of Overthinking: Examining the Reasoning-Action Dilemma in Agentic Tasks
  9. RM-Bench: Benchmarking Reward Models of Language Models with Subtlety and Style
  10. RETRIEVAL HEAD MECHANISTICALLY EXPLAINS LONG-CONTEXT FACTUALITY
  11. MAP: Multi-Human-Value Alignment Palette
  12. MEASURING AND ENHANCING TRUSTWORTHINESS OF LLMS IN RAG THROUGH GROUNDED ATTRIBUTIONS AND LEARNING TO REFUSE
  13. SPREAD PREFERENCE ANNOTATION: DIRECT PREFER- ENCE JUDGMENT FOR EFFICIENT LLM ALIGNMENT
  14. Unlocking the Power of Function Vectors for Characterizing and Mitigating Catastrophic Forgetting in Continual Instruction Tuning

Schedule