4 papers indexed here
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.
HAServe is presented, which turns HorizonSchema, a workload abstraction that connects exact prefix paths, permitted serializations, and released consumers as shared inputs to both decisions, into a scheduling--memory co-design method of systematic optimization for online LLM serving.
STEPS is proposed, a selective OPSD framework that controls the location, direction, and duration of distillation that improves mean accuracy over GRPO and support brief, targeted self-distillation as an efficient complement to reward-driven reasoning training.
Two metrics to measure the exploration bias in instruction following are proposed and a two-stage framework to alleviate it is introduced: Behavioral Bootstrapping, a lightweight rejection sampling fine-tuning stage before RL to activate hard instructions; and Scarcity-Aware Rewards, a new RL reward function that assig...
CacheBridge is introduced, which co-designs architecture-indexed mapper support, attention-aligned calibration, and bounded mapper construction while retaining a closed-form affine interface for online deployment.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
