Diversity-Driven Offline Multi-Objective Optimization via Nested Pareto Set Learning. A risk-aware framework that learns diverse Pareto solutions from offline data by combining energy-based OOD risk control with nested preference-model optimization and a novel IGD_off selection strategy. ICML 2026. - GitHub - YaolinWen/DOMOO: Diversity-Driven Offline Multi-Objective Optimization via Nested Pareto Set Learning. A risk-aware framework that learns diverse Pareto solutions from offline data by combining energy-based OOD risk control with nested preference-model optimization and a novel IGD_off selection strategy. ICML 2026.
[ICLR-2026] The PyTorch Implementation of FastMix. The algorithms are based on our paper: "Fast Data Mixture Optimization via Gradient Descent". FastMix is designed to rapidly identify optimal data mixing ratios for large language models (LLMs). - GitHub - hrtan/fastmix: [ICLR-2026] The PyTorch Implementation of FastMix. The algorithms are based on our paper: "Fast Data Mixture Optimization via Gradient Descent". FastMix is designed to rapidly identify optimal data mixing ratios for large language models (LLMs).
[ICML 2026] A neuro-symbolic framework for abductive reasoning that grounds multi-agent collaboration in structured belief states using a causal graph and state machine.
thinkingbox is a framework for defining tool as MCP servers, running LLM agents against them, and evaluating agent behavior — for offline training-data generation, reinforcement-learning training loops, and model evaluation.
