AI ResearchApr 20, 2026, 9:00 AM

Gradient-based Planning for World Models at Longer Horizons

TickrWire Editorial Desk·Apr 20, 2026, 9:00 AM·1 min read AI-assisted, human-reviewed

Reported by BAIR Blog: Gradient-based Planning for World Models at Longer Horizons. Analysis and context written by TickrWire.

30-second summary

Berkeley researchers introduce GRASP, a gradient-based planner for learned world models that extends planning horizons, improving long-term decision-making in AI systems.

TickrWire
Gradient-based Planning for World Models at Longer Horizons
Full story

.grasp-results-table table { font-size: 0.875rem; line-height: 1.35; width: 100%; }

.grasp-results-table th,

.grasp-results-table td { padding: 0.35rem 0.5rem; }

/* Consistent whitespace between major sections (this post is long and hr-heavy) */

article.post-content h2 {

margin-top: 2.75rem;

margin-bottom: 0.75rem;

}

article.post-content h2:first-of-type {

margin-top: 2.25rem;

}

article.post-content h3 {

margin-top: 1.65rem;

margin-bottom: 0.5rem;

}

article.post-content hr {

margin-top: 2.5rem;

margin-bottom: 2.5rem;

}

GRASP is a new gradient-based planner for learned

Sources · 1
Read next
More stories