2 papers
cs.AI2026
Coverage, Not Credit: Failure-Credit Routing of Zeroth-Order Perturbation Budgets Does Not Improve On-Pool Sample Efficiency for LLM Agents
Yuxu Ge
Trajectory-level credit assignment can localize which module of a tool-using LLM agent causes failures using only verifiable signals. We ask whether such failure credit should rout…
cs.LG2026
BigMac: Breaking the Pareto Frontier of Compute and Memory in Multimodal LLM Training
Zili Zhang, Chengxu Yang, Shenglong Zhang +8
Training multimodal large language models (MLLMs) is challenged by both model and data heterogeneity. Existing systems redesign the training pipeline to address these challenges, b…