Newest papers

Knowing When to Stop Thinking: A Dissection of Expected Free Energy as a Test-Time Compute Controller for Latent World-Model Planning

Tong Zhang, Zhiyuan Shi, Yun Peng, Tao Xie · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.28011 · cited by 0 works (OpenAlex) · open-access PDF

Korean

배운 세계 모델을 사용하여 계획자는 예측된 결과를 향상시킬 수 있지만 더 나쁜 행동을 선택할 수 있습니다. 예상되는 무료 에너지 (EFE)가 계획에 얼마나 많은 계산을 지출하는지 결정하는 데 도움이 될 때 연구합니다. EFE 영감을받은 목표 거리와 스토카스틱 예측 사이의 불일치를 결합합니다. 우리는 후보의 점수, 중지 및 온라인 예측 업데이트를 별도로 테스트합니다. 엘리트 비용 스택이 목표에 도달하는 데 필요한 지역에 대한 정교함을 멈추면 성공 플라이트와 깊이, 고정 깊이 참조에 가까운 평균 성공을 감소시킵니다. 더 큰 테스트 된 무게에서 보상 불일치가 냉동 모델의 성공을 낮추고 있습니다. 처벌은 일부 작업 설정과 체중

Machine translation (Meta M2M100); the English original follows.

Abstract

A planner using a learned world model can improve its predicted outcome while choosing a worse action. We study when expected free energy (EFE) can help decide how much computation to spend on a plan. An EFE-inspired objective combines goal distance with disagreement among stochastic predictions. We test candidate scoring, stopping, and online predictor updates separately. Stopping when elite cost stalls reduces refinement on tasks whose success plateaus with depth, with mean success close to fixed-depth references. At the larger tested weights, rewarding disagreement lowers success with a frozen model. Penalizing it helps at some task and weight settings but harms others. When training data omit a region needed to reach the goal, online updates recover part of the resulting loss. The tested cost-stall rule can fail when further optimization exploits model errors, because predicted cost keeps falling after task success peaks. Cost progress can guide computation on a plateau, but protecting performance after a peak requires information about the consequences of the chosen actions.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Task (project management), Computer science, Computation, Exploit, Outcome (game theory), Energy (signal processing)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange