Tag
论文提出 ProACT，一个性能驱动的自适应 VR 消防训练系统，通过闭环调节火灾蔓延动力学与物体可燃性来动态调整训练难度，实验表明其在客观训练效果和主观体验上均优于基线方法。
A paper introduces A-RSI (Adaptive Task Refresh for Self-Improvement), a method enabling zero-data multimodal self-improvement by adaptively refreshing tasks during training without requiring new human-labeled data.
The paper proposes TAPS (Trajectory Adaptive Progress–Fluctuation Scheduler), which adaptively adjusts the step size of recurrent updates in looped transformers by decomposing terminal loss sensitivity into persistent progress and centered fluctuation, improving accuracy without retraining and delivering up to 1.56× wall-clock speedup when applied during training.
D³-MOPD is a zero-overhead scheduler for multi-teacher distillation that dynamically adjusts domain sampling ratios based on per-domain reverse-KL trajectories, improving convergence efficiency and closing most of the student-to-teacher performance gap.
This paper introduces AdaMTP, an adaptive training paradigm for multi-token prediction that dynamically aligns prediction horizons with sequence predictability using entropy-based segmentation, consistently outperforming standard MTP on math, code, and general benchmarks across three LLM backbones.
LLMZero uses LLM agents to search over training trajectories via tree search, discovering adaptive multi-parameter transitions for RL post-training that outperform fixed schedules and grid search across diverse tasks.
Adaptive Teacher Exposure for Self-Distillation (ATESD) improves LLM reasoning by dynamically adjusting how much of the reference reasoning the teacher shows the student during training, using a learnable policy controller and a discounted learning-progress reward. Experiments on math benchmarks show consistent improvements over existing self-distillation and RL baselines.