9 papers
Meta: Recursive Self-Improvement through Emergent Depth
Zae Myung Kim, Young-Jun Lee, Seungyeon Jwa +1
Self-improving LLM agents refine answers, not the process that produces those answers. Systems that add a meta-level hold that level fixed, and those that edit themselves must leav…
Structure Liberates: How Constrained Sensemaking Produces More Novel Research Output
James Mooney, Zae Myung Kim, Young-Jun Lee +1
Scientific discovery is an extended process of ideation--surveying prior work, forming hypotheses, and refining reasoning--yet existing approaches treat this phase as a brief pream…
The Amazing Agent Race: Strong Tool Users, Weak Navigators
Zae Myung Kim, Dongseok Lee, Jaehyung Kim +2
Existing tool-use benchmarks for LLM agents are overwhelmingly linear: our analysis of six benchmarks shows 55 to 100% of instances are simple chains of 2 to 5 steps. We introduce…
Align to Structure: Aligning Large Language Models with Structural Information
Zae Myung Kim, Anand Ramachandran, Farideh Tavazoee +3
Generating long, coherent text remains a challenge for large language models (LLMs), as they lack hierarchical planning and structured organization in discourse generation. We intr…
Anchors Aweigh! Sail for Optimal Unified Multi-Modal Representations
Minoh Jeong, Zae Myung Kim, Min Namgung +3
A unified representation space in multi-modal learning is essential for effectively integrating diverse data sources, such as text, images, and audio, to enhance efficiency and per…
Toward Evaluative Thinking: Meta Policy Optimization with Evolving Reward Models
Zae Myung Kim, Chanwoo Park, Vipul Raheja +2
Reward-based alignment methods for large language models (LLMs) face two key limitations: vulnerability to reward hacking, where models exploit flaws in the reward signal; and reli…