using https://events.vtools.ieee.org/meetings/xml/0/30/asc/6/OREGON
33278 bytes
4 meetings
- Title:
- High Performance Inferencing for LLMs
- Date:
- December 11th
6:00 PM (1 hour) - Abstract:
Inferencing has become ubiquitous across cloud, regional, edge, and device environments, powering a wide spectrum of AI use cases spanning vision, language, and traditional machine learning applications. In recent years, Large Language Models (LLMs), initially developed for natural language tasks, have expanded to multimodal applications including vision speech, reasoning and planning each demanding distinct service-level objectives (SLOs). Achieving high-performance inferencing for such diverse workloads requires both model-level and system-level optimizations.
This talk focuses on system-level optimization techniques that maximize token throughput , achieve user experience metrics and inference service-provider efficiency. We review several recent innovations including KV caching, Paged/Flash/Radix Attention, Speculative Decoding, P/D Disaggregation, KV Routing and Parallelism, and explain how these mechanisms enhance performance by reducing latency, memory footprint, and compute overhead. These techniques are implemented in leading open-source inference frameworks such as vLLM, SGLang, Hugging Face TGI, and NVIDIA’s TensorRT-llm, which form the backbone of large-scale public and private LLM serving platforms.
Attendees will gain a practical understanding of the challenges in delivering scalable, low-latency LLM inference, and of the architectural and algorithmic innovations driving next-generation high-performance inference systems.
- Title:
- IEEE CS Webinar: IEEE Oregon Section Technical Seminar - Beyond Games: Real-World Applications of (Deep) Reinforcement Learning
- Date:
- December 11th
6:00 PM (1 hour) - Abstract:
We hope to have you for another interesting talk by one of the experts that we invite from academia, industry, and government.
* As this online event is free and open to non-IEEE members, please feel free to share it with your colleagues, students, classmates, etc.
* For the abstract and biography of the speaker, please refer to the speakers section below.
* Please note that you will receive a registration confirmation email after you register for the event and you will receive a separate email containing the invite to the meeting later. You can add the link to the meeting invite to your calendar manually as the calendar invite does not get updated automatically.
- Title:
- 2025 Oregon Section Chapter Awards Banquet
- Date:
- December 13th
3:30 PM (2.5 hours) - Location:
- 3211 SW Cedar Hills Blvd
Beaveront - Abstract:
Please join us for the 2025 Oregon Section Chapter Chairs Awards Dinner
Chairs are welcome to attend with one guest or send a representative for your chapter.
!Please limit to only Oregon Chapter Chairs and their guest!
Dinner reservation is at 4:00 pm at the Lake Oswego Grille in Beaverton
3211 SW Cedar Hills Blvd, Beaverton, OR 97005
Cost will be covered by the Section, Thank You for volunteering your time all year.
Choose your Dinner entree from the 4 options or let us know if you have other dietary needs to accommodate prior to the dinner.
- Title:
- Vehicle-to-Everything (V2X) Communications in Unlicensed Spectrum Can Be Safe and Efficient
- Date:
- December 17th
6:00 PM (1.5 hours) - Location:
- Bannan
Seattle, WA - Abstract:
To meet the communications demands of connected vehicles, the wireless devices deployed in vehicles and on roadside infrastructure may need access to more spectrum than is available today. This presentation describes a novel approach that allows connected vehicle devices using V2X technology (e.g., C-V2X or NR-V2X) to share spectrum with Wi-Fi and other unlicensed devices, thereby gaining access to more spectrum. Each vehicle dynamically and independently adjusts to its environment in a manner that gives connected V2X devices access to enough of the shared spectrum to meet their quality-of-service requirements, while leaving as much spectrum as possible for Wi-Fi. As a result, this approach uses spectrum more efficiently than the current approach of establishing one spectrum band exclusively for connected vehicles and another spectrum band exclusively for Wi-Fi and other unlicensed devices. The proposed approach uses a backward-compatible form of implicit beaconing that requires no change to Wi-Fi technology, so there is no need to replace Wi-Fi devices that have been deployed, and the approach requires only modest modifications to V2X which reduces cost and complexity. The approach also does not require involvement from a cellular operator or other centralized controller. Simulation results show that it is possible to protect quality of service for both V2X and Wi-Fi communications in a shared band, while achieving high spectrum efficiency. One spectrum band where this approach could be especially beneficial is adjacent to the Intelligent Transportation System (ITS) band, where this approach could help meet the needs of both connected vehicles and Wi-Fi 6.
4 meetings. Generated Friday, December 12 2025, at 2:43:09 PM. All times America/Los_Angeles