A seemingly reasonable push in Sokoban can make the puzzle unsolvable. To avoid such mistakes, an agent has to anticipate the consequences of its actions before execution, especially when feedback from the real environment is sparse or actions are irreversible. We introduce Agentic Imagination (Agima), a training frame...
Jia-Xin Ge, Junyi Zhang, Tian-Ning Chai et al.· 0 citations
RGB-infrared (RGB-IR) object detection combines visible appearance cues with infrared thermal information and supports perception under complex illumination, weak texture, occlusion, and large scale variation. However, existing fusion detectors still face three coupled problems: modality-specific structures may be comp...
Hai-Wen Feng, Xiao-Chuan Qian· Journal of King Saud Univers...· 0 citations
This work proposes training models to fail gracefully: to preserve successful behavior whenever possible while learning to execute a safe fallback policy when acting would be unreliable, and shows that reinforcement learning can not only incentivize models to succeed, but also to reason about the limits of their capabi...
Daman Arora, Xue-Chen Liu, H. Singh et al.· 0 citations
Training a model to generate visual content can encourage it to learn rich perceptual capabilities related to geometry, spatial relationships, and objectness; yet, its benefits for visual understanding remain unclear. We ask: when and how does visual generation supervision improve visual understanding? We study control...
Jia-Xin Ge, Yi-Ming Qin, Ji Xie et al.· 0 citations
Learning to complete tasks in unfamiliar environments with unknown rules remains a key challenge for LLM agents. Current LLM agents often record their discoveries in prose, which may not provide a compact, explicit account of how the environment works. Inspired by how scientists organize observations into testable, pre...
Guan-Ning Zeng, Jia-Ni Wang, Wenjie Ma et al.· arXiv.org· 0 citations
Tail-Likelihood Reinforcement Learning (TailRL), which maximizes the log-probability of exceeding a randomly chosen reward threshold, which gives more weight to rare, high-reward rollouts and can be interpreted as a mixture of Best-of-k gradients.
Shrinivas Ramasubramanian, Daman Arora, Fahim Tajwar et al.· 2 citations
Reinforcement learning typically optimizes average reward. For generative policies, the average can hide an important distinction: two policies can achieve the same mean reward while having very different chances of producing a rare but high-reward rollout. This matters as sampling increases during training and inferen...
Shrinivas Ramasubramanian, Daman Arora, Fahim Tajwar et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.