余席宇 博士
Employers on the program
- 2026OPPO1 talk
Events1
Conferences1
Talks1
Years active2026
The year around now
Talks in the past 12 months and announced talks ahead, one row per conference, January to December. Hover a mark for the talk.
Appearances by year
- AICon 全球人工智能开发与应用大会 (AICon Global AI Development and Application Conference)
Talks
| Date | Event | Conference | Role | Title | Links |
|---|---|---|---|---|---|
| Aug 21, 2026 | AICon 全球人工智能开发与应用大会 | 深圳 | AICon 全球人工智能开发与应用大会 (AICon Global AI Development and Application Conference) | speaker | OPPO 端侧多模态大模型工程化实践 as OPPO 随着端侧多模态场景扩展与长上下文需求增长，端侧大模型在 prefill / decoding 时延、工程链路复杂度、模型迭代周期上面临多重约束。本次分享围绕 OPPO 端侧化全栈实践，介绍以 QAT 量化感知训练 为核心的协同训练体系，覆盖稀疏化压缩、长上下文 Cache Eviction、解码加速训练，并结合…Show less随着端侧多模态场景扩展与长上下文需求增长，端侧大模型在 prefill / decoding 时延、工程链路复杂度、模型迭代周期上面临多重约束。本次分享围绕 OPPO 端侧化全栈实践，介绍以 QAT 量化感知训练 为核心的协同训练体系，覆盖稀疏化压缩、长上下文 Cache Eviction、解码加速训练，并结合 QALFT 自动化工程化 与 0 阶端侧训练，打通从模型压缩、编译部署到垂域落地与本地个性化的完整链路。 演讲提纲 端侧大模型的核心挑战 推理效率瓶颈：长上下文场景下，prefill 与 decoding 时延显著上升 工程链路复杂：端侧化涉及 … |
Frequent co-speakers
No one else spoke at two or more of the same events.
Speakers with the same name are merged into one entry, so this page may combine more than one person. Affiliations are shown as each conference listed them.