AI Agent Papers2026
Lets a main agent synthesize executable workflows to coordinate and verify sub-agents as a task unfolds. The paper reports 88.08% overall accuracy in Forced mode across six benchmarks.
Estimates how committed tokens improve predictions at remaining masked positions, then uses those contributions to refine reinforcement-learning signals for diffusion language models. On LLaDA-8B-Instruct, it raises MATH500 and GSM8K scores by up to 9.0 and 5.3 points.