Learning What to Skip: Counterfactual Credit Assignment for Efficient Multi-Agent LLM WorkflowsResearchSep 28, 2026
NeuralCert: certified computational discovery of extremal mathematical constructionsResearchSep 28, 2026
All In Good Time: Causality-Aware Framework for LLM-Based Simultaneous Speech-to-Speech TranslationResearchSep 28, 2026
Guarded Gradient-Based Activation Steering of Shutdown Responses in Qwen3.5-0.8B: A Minimum-Step PolicyResearchSep 28, 2026