arXiv:2603.24714v1 Announce Type: new Abstract: Analog design often slows down because even small changes to device sizes or biases require expensive simulation cycles, and high-quality solutions typically occupy only a narrow part of a very large search space. While existing…
arXiv:2511.16992v2 Announce Type: replace Abstract: Aligning Large Language Models (LLMs) with human values often involves balancing multiple, conflicting objectives such as helpfulness and harmlessness. Training these models is computationally intensive, and centralizing the process raises significant data privacy concerns. Federated…
arXiv:2603.24744v1 Announce Type: new Abstract: Weather data, comprising multiple variables, poses significant challenges due to its high dimensionality and multimodal nature. Creating low-dimensional embeddings requires compressing this data into a compact, shared latent space. This compression is required to improve…
arXiv:2603.16951v2 Announce Type: replace Abstract: Identifying physical laws from noisy observational data is a central challenge in scientific machine learning. We present Minimum-Action Learning (MAL), a framework that selects symbolic force laws from a pre-specified basis library by minimizing a…
arXiv:2603.24746v1 Announce Type: new Abstract: Grokking — the delayed onset of generalization after early memorization — is often described with phase-transition language, but that claim has lacked falsifiable finite-size inputs. Here we supply those inputs by treating the group order…
arXiv:2410.12476v3 Announce Type: replace-cross Abstract: Machine learning (ML) holds great promise for clinical applications but is often hindered by limited access to high-quality data due to privacy concerns, high costs, and long timelines associated with clinical trials. While large language…
arXiv:2603.24753v1 Announce Type: new Abstract: Standard vision models treat objects as independent points in Euclidean space, unable to capture hierarchical structure like parts within wholes. We introduce Worldline Slot Attention, which models objects as persistent trajectories through spacetime worldlines, where…
arXiv:2509.21385v2 Announce Type: replace-cross Abstract: Concept Bottleneck Models (CBMs) use a set of human-interpretable concepts to predict the final task label, enabling domain experts to not only validate the CBM’s predictions, but also intervene on incorrect concepts at test time.…
arXiv:2603.24780v1 Announce Type: new Abstract: Effective problem solving with Large Language Models (LLMs) can be enhanced when they are paired with external search algorithms. By viewing the space of diverse ideas and their follow-up possibilities as a tree structure, the…
arXiv:2603.15017v2 Announce Type: replace-cross Abstract: Because human preferences are too complex to codify, AIs operate with misspecified objectives. Optimizing such objectives often produces undesirable outcomes; this phenomenon is known as reward hacking. Such outcomes are not necessarily catastrophic. Indeed, most…
