Han Zhao
where else to find them
no profile links on file
what they talk about
where they keep going back
no series more than once — all 4 in the filters
on a stage with
all 17 in the filters
who talks about the same things
nobody close enough yet
| title | conference | date |
|---|---|---|
| FLARE: Anomaly Diagnostics for Divergent LLM Training in GPU Clusters of Thousand-Plus... | NSDI '26 | 2026-05-04 |
FLARE integrates a lightweight tracing daemon that instruments Python APIs and GPU kernels without backend patches, and a diagnostic engine that performs… | ||
| Efficient Performance-Aware GPU Sharing with Compatibility and Isolation through... | USENIX ATC '25 | 2025-07-07 |
KRYPTON implements kernel-space command-buffer interception and a spatio-temporal orchestrator on A100 and 4090 hardware, using offline profiling, CPU tokens… | ||
| Revisiting Scalarization in Multi-Task Learning | Joint IFML/MPG Symposium 2024 | 2024-11-19 |
Claims that linear scalarization cannot in general trace the Pareto front for linear multi‑task models, proving concretely that when hidden width q >= k the… | ||
| DVABatch: Diversity-aware Multi-Entry Multi-Exit Batching for Efficient... | USENIX ATC '22 | 2022-07-11 |
DVABatch is a GPU runtime that slices DNNs into staged executors and exposes three meta-operations new/stress/split to implement a multi-entry multi-exit… | ||
| nothing matches | ||