aiqbench analysis
This research introduces Future-Aware Recall (FAR), a novel framework for world models that dynamically selects relevant episodic memories by learning which retrieval cues—such as time, pose, or vision—are most predictive for future outcomes. By optimizing recall based on conditional log-likelihood rather than static heuristics, the system effectively adapts to changing environments without relying on fixed criteria. This advancement addresses a critical bottleneck in building robust world models capable of leveraging long-term historical data efficiently.
Learning What to Recall: Adaptive Multi-Cue Episodic Memory for World Models
arXiv:2609.34677v1 Announce Type: new Abstract: World models predict future observations from current experience and actions, yet prediction can depend on observations seen far in the past. Episodic memory preserves past observations for later recall; however, as memory accumulates, it raises a fundamental question: which memories are useful for the current prediction, and which available retrieval cues should be trusted to find them? This is challenging because fixed criteria based on recency, pose overlap, or visual similarity can be unreliable across environments and queries. We propose Future-Aware Recall (FAR), a framework that learns episodic recall from future-aware predictive supervision and adaptive multi-cue scoring. During training, FAR measures predictive utility by the conditional log-likelihood of the realized future given recalled context, approximated by negative diffusion prediction loss, and uses it to train a retriever that remains future-blind at inference. The retriever learns cue-specific relevance and automatically determines which available retrieval cues, such as time, pose, vision, and audio, to trust for each query when selecting memories. Across three complementary settings, FAR outperforms hand-designed recall even with the same retrieval cues, automatically adapts which available cues to trust, and recalls the right history as the world changes. Together, these results establish FAR as a flexible, principled approach to episodic memory access in world models.
Abstract:World models predict future observations from current experience and actions, yet prediction can depend on observations seen far in the past. Episodic memory preserves past observations for later recall; however, as memory accumulates, it raises a fundamental question: which memories are useful for the current prediction, and which available retrieval cues should be trusted to find them? This is challenging because fixed criteria based on recency, pose overlap, or visual similarity can be unreliable across environments and queries. We propose Future-Aware Recall (FAR), a framework that learns episodic recall from future-aware predictive supervision and adaptive multi-cue scoring. During training, FAR measures predictive utility by the conditional log-likelihood of the realized future given recalled context, approximated by negative diffusion prediction loss, and uses it to train a retriever that remains future-blind at inference. The retriever learns cue-specific relevance and automatically determines which available retrieval cues, such as time, pose, vision, and audio, to trust for each query when selecting memories. Across three complementary settings, FAR outperforms hand-designed recall even with the same retrieval cues, automatically adapts which available cues to trust, and recalls the right history as the world changes. Together, these results establish FAR as a flexible, principled approach to episodic memory access in world models.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
arXiv:2609.34677 [cs.LG]
(or arXiv:2609.34677v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2609.34677 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 309 words
Related coverage
What's Continual Learning, and Why Might We Expect To See It In Advanced LLM Agents?
LessWrong · Jun 12, 2026
Concrete Evaluations to Investigate the OpenAI Model That Hacked Hugging Face
AI Alignment Forum · Aug 3, 2026
[AINews] not much happened today
Latent Space · Sep 10, 2026
Runway’s WorldPrompt and the Engineering of Real-Time Worlds
Latent Space · Sep 25, 2026