…S- GFN provides more stable training while maintaining the optimal policy of GFN . We demonstrate the overwhelming attack performance and diversity of S- GFN across various settings. View arXiv page View PDF…
…compute-optimal allocation advice. Beyond a point, further repetition is counterproductive and compute is better spent on model capacity. We show that following our law's recommended configuration improves performance in data…
…Following this analysis, we adopt a revised PPO-EWMA method, which achieves significant gains in both training speed and optimization performance. Code at https://github.com/millioniron/ROLL. View arXiv page View…
…Your Language Model is Secretly Optimizing a Preference Graph (2026) EvoPref: Multi-Objective Evolutionary Optimization Discovers Diverse LLM Alignments Beyond Gradient Descent (2026) Parameter-Efficient Neuroevolution for Diverse LLM Generation: Quality-Diversity…
…While existing video VAEs achieve commendable reconstruction quality, continued optimization of reconstruction does not necessarily translate into improved generative performance . How to enhance the diffusability of video latents remains a critical and…
…Our unification reveals a shared optimization principle: Global Trajectory Score Matching (GTSM), for which gradient boosting (in an idealized version) is asymptotically optimal. We underscore the conceptual value of our work through…
…Bo Yin , Qi Li , Abstract FATE is an on-policy framework that uses failure trajectories to improve agent safety and performance through self-evolution and Pareto-aware optimization. AI-generated summary Tool…
…We introduce General Dijkstra Search (GDS), and prove that globally optimal goal-reaching policies can be recovered through temporal composition of intermediate optimal sub-policies . Motivated by the "search, select, update" principle…
…Yulong Huang , Xiang Liu , , , , , , Abstract Linear attention models face challenges with information decay and convergence, which are addressed through a momentum-based approach that improves training efficiency and performance over existing models…
…balance efficiency and performance. Existing rank-adaptive solutions such as DyLoRA mitigate this by sampling ranks during the training from a predefined distribution. However, they often yield sub-optimal results at higher…