We have 3 of 18 papers
We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.
Not the right person? Other researchers publish under this name.
GeoMesh is presented, a synchronous geo-distributed training framework for heterogeneous GPUs that balances per-worker workloads by assigning each GPU a suitable batch size and number of inner steps, so faster GPUs do more useful work instead of waiting.
This paper proposes CONA, a new training method that introduces online strategy chaining, which ranks candidate strategies during training using a surrogate metric built from compute throughput and gradient statistics, and switches the current strategy to a new strategy with a higher metric.
Bidirectional prefix caching, BiCache, the first KV caching technique for shared prefixes in DLMs is proposed, which dynamically identifies a safe layer depth for reusing shared prefix KVs and eliminates redundant computation.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
