ProsaBuddy: Assisting Mechanized Real-Time Schedulability Analysis with LLM-based AgentsResearchOct 6, 2026
An Executable Benchmark for LLM-Based HLS Repair:Design Complexity and Repair UnderconstraintResearchOct 6, 2026
Learning without Overwriting: A Theory of Self-Distillation and Supervised Fine-Tuning in Continual ReasoningResearchOct 6, 2026
SALUS: Automated Auditing of NL-to-SQL Benchmarks through Weak Supervision of Multi-Agent OutputResearchOct 6, 2026
Don't Judge an LLM Only by Its Activations: Discovering Suppressed Safety Features via Counterfactual Activation PotentialResearchOct 6, 2026
Fast Convergence through Distributed Augmentation for Class-Imbalanced Federated LearningResearchOct 6, 2026
ColdDDI: Evaluating Knowledge Utilization in Cold-Start Drug-Drug Interaction PredictionResearchOct 6, 2026