Internalizing the Future: A Unified Agentic Training Paradigm for World Model Planning
The paper introduces a unified training paradigm that equips large language model agents with internal world modeling for future-aware planning. It outlines a three-stage process—World Model Agentic Mid-Training, Format-Eliciting SFT, and Foresight-Conditioned Reinforcement Learning—to embed and calibrate predictive capabilities. Experiments on search and mathematical reasoning benchmarks demonstrate performance gains over existing baselines.
- ▪The authors propose training a single autoregressive model to generate prospective state rollouts and plan-conditioned success estimates, providing a textual analogue of Q-values.
- ▪A three-stage training pipeline (WM-AMT, FE-SFT, FC-RL) is designed to inject, structure, and refine latent predictive abilities within the policy.
- ▪Evaluations on search and mathematical reasoning tasks show the method consistently outperforms other training baselines.
arXiv cs.AI files mainly under ai research. We currently carry 1,128 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | arXiv.org |
| Canonical URL | https://arxiv.org/abs/2606.27483 |
| Publication time | Mon, 29 Jun 2026 00:00:00 -0400 |
| Retrieval time | 2026-06-29T07:20:58.024Z |
| Last seen | 2026-06-29T07:20:58.024Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | fOJlvwIxDf25 |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Artificial Intelligence arXiv:2606.27483 (cs) [Submitted on 25 Jun 2026] Title:Internalizing the Future: A Unified Agentic Training Paradigm for World Model Planning Authors:Xuan Zhang, Zhijian Zhou, Lingfeng Qiao, Yulei Qin, Ke Li, Xing Sun, Xiaoyu Tan, Chao Qu, Yuan Qi View a PDF of the paper titled Internalizing the Future: A Unified Agentic Training Paradigm for World Model Planning, by Xuan Zhang and 8 other authors View PDF Abstract:Large language model (LLM) agents have demonstrated strong capability in sequential decision-making, yet they remains fundamentally reactive in long-horizon tasks. Unlike humans who employ "what-if" reasoning to evaluate potential plans before commitment, standard agents lack an internal world model to simulate future outcomes.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv.org.
Discussion
0 commentsMore from arXiv.org
