Similar Items: Formalize, Don't Optimize: The Heuristic Trap in LLM-Generated Combinatorial Solvers
- AI-Generated Smells: An Analysis of Code and Architecture in LLM and Agent-Driven Development
- To Call or Not to Call: A Framework to Assess and Optimize LLM Tool Calling
- RHyVE: Competence-Aware Verification and Phase-Aware Deployment for LLM-Generated Reward Hypotheses
- AdaMeZO: Adam-style Zeroth-Order Optimizer for LLM Fine-tuning Without Maintaining the Moments
- Shepherd: A Runtime Substrate Empowering Meta-Agents with a Formalized Execution Trace
- NeuroAgent: LLM Agents for Multimodal Neuroimaging Analysis and Research