Vector databases use hashing to partition data across"shards,"logical units for distributed execution. This placement, however, destroys semantic locality, forcing each query into scatter-gather limited by the slowest shard. Vector-space clustering can help, but scientific evidence is often connected by factual relatio...
Song Young Oh, Amal Gueroudji, Seth Ockerman et al.· 0 citations
Agents can now externalize experience into memory, consolidating historical traces into semantic knowledge and procedural shortcuts that persist between sessions. Such memory is typically private to a single agent. We argue that agentic memory benefits from being collective, because trustworthy knowledge of the shared...
Ryan Chard, Gus Ellerm, Alexander Brace et al.· 0 citations
Cloud-native LLM serving frameworks have made deployment routine in data centers, yet deploying them on leadership-class supercomputers remains an engineering challenge requiring scheduler integration, MPI launch, accelerator selection, node-local weight staging, and platform-specific patches. We present ExaServe, a pi...
Wen-Yi Wang, Shu Shi, Y. Babuji et al.· 0 citations
This work presents Avatar, an actor-based architecture comprising an orchestrator, an executor, and a provenance monitor, and observes that Avatar's rule mode reproduces native execution, with a single unchanged core running all three.
Suman Raj, H. Nguyen, Haochen Pan et al.· 0 citations
This work designs Diamond Agent, an agentic system that enables intelligent execution of HPC workflows across heterogeneous clusters with typed skills as the interface, and reduces the median additional completion time relative to the fastest observed placement from 42 seconds to 4 seconds, a 10.5x reduction.
Haotian Xie, Jun-Lin Chen, Ming-Kai Zheng et al.· 0 citations
This work applies differential privacy (which adds noise to the model weights before sending across the network) as an added privacy measure to protect sensitive data from being reconstructed in Federated Learning.
OmniLens is presented, which applies a single lens family to any model-width activation, whether residual stream, attention, or MLP, and combines two independent scaling techniques, which reproduces key published results at substantially lower cost.
Jordan Pettyjohn, Mansi Sakarvadia, Nathaniel Hudson 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.
