Difference between revisions of "Resource:Seminar"

From MobiNetS
Jump to: navigation, search
 
(3 intermediate revisions by one other user not shown)
Line 1: Line 1:
{{SemNote
{{SemNote
|time='''2026-01-09 10:30'''
|time='''2026-01-30 10:30'''
|addr=4th Research Building A518
|addr=4th Research Building A518
|note=Useful links: [[Resource:Reading_List|📚 Readling list]]; [[Resource:Seminar_schedules|📆 Schedules]]; [[Resource:Previous_Seminars|🧐 Previous seminars]].
|note=Useful links: [[Resource:Reading_List|📚 Readling list]]; [[Resource:Seminar_schedules|📆 Schedules]]; [[Resource:Previous_Seminars|🧐 Previous seminars]].
Line 8: Line 8:


{{Latest_seminar
{{Latest_seminar
|abstract = DistServe improves the performance of large language models (LLMs) serving by disaggregating the prefill and decoding computation. Existing LLM serving systems colocate the two phases and batch the computation of prefill and decoding across all users and requests. We find that this strategy not only leads to strong prefill-decoding interferences but also couples the resource allocation and parallelism plans for both phases. LLM applications often emphasize individual latency for each phase: time to first token (TTFT) for the prefill phase and time per output token (TPOT) of each request for the decoding phase. In the presence of stringent latency requirements, existing systems have to prioritize one latency over the other, or over-provision compute resources to meet both. DistServe assigns prefill and decoding computation to different GPUs, hence eliminating prefill-decoding interferences. Given the application's TTFT and TPOT requirements, DistServe co-optimizes the resource allocation and parallelism strategy tailored for each phase. DistServe also places the two phases according to the serving cluster's bandwidth to minimize the communication caused by disaggregation. As a result, DistServe significantly improves LLM serving performance in terms of the maximum rate that can be served within both TTFT and TPOT constraints on each GPU. Our evaluations show that on various popular LLMs, applications, and latency requirements, DistServe can serve 7.4× more requests or 12.6× tighter SLO, compared to state-of-the-art systems, while staying within latency constraints for > 90% of requests.
|abstract = LoRa technology promises to enable Internet of Things applications over large geographical areas. However, its performance is often hampered by poor channel quality in urban environments, where blockage and multipath effects are prevalent. Our study uncovers that a slight shift in the position or attitude of the receiving antenna can substantially improve the received signal quality. This phenomenon can be attributed to the rich multipath characteristics of wireless signal propagation in urban environments, wherein even small antenna movement can alter the dominant signal path or reduce the polarization angular difference between transceivers. Leveraging these key observations, we propose and implement MoLoRa, an intelligent mobile antenna system designed to enhance LoRa packet reception. At its core, MoLoRa represents the position and attitude of an antenna as a state and employs a statistical optimization method to search for states that offer optimal signal quality efficiently. Through extensive evaluation, we demonstrate that MoLoRa achieves a maximum Signal-to-Noise Ratio (SNR) gain of 13 dB in a few attempts, enabling formerly problematic blind spots to reconnect and strengthening links for other nodes.
|confname =OSDI'24
|confname =SenSys'25
|link = https://www.usenix.org/conference/osdi24/presentation/zhong-yinmin
|link = https://dl.acm.org/doi/10.1145/3715014.3722075
|title= DistServe: Disaggregating Prefill and Decoding for Goodput-optimized Large Language Model Serving
|title= MoLoRa: Intelligent Mobile Antenna System for Enhanced LoRa Reception in Urban Environments
|speaker=Ruizheng
|speaker=Kai Chen
|date=2026-1-09
|date=2026-1-30
}}
}}
{{Latest_seminar
{{Latest_seminar
|abstract =In virtual machine (VM) allocation systems, caching repetitive and similar VM allocation requests and associated resolution rules is crucial for reducing computational costs and meeting strict latency requirements. While modern allocation systems distribute requests among multiple allocator agents and use caching to improve performance, current schedulers often neglect the cache state and latency considerations when assigning each new request to an agent. Due to the high variance in costs of cache hits and misses and the associated processing overheads of updating the caches, simple load-balancing and cache-aware mechanisms result in high latencies. We introduce Kamino, a high-performance, latency-driven and cache-aware request scheduling system aimed at minimizing end-to-end latencies. Kamino employs a novel scheduling algorithm grounded in theory which uses partial indicators from the cache state to assign each new request to the agent with the lowest estimated latency. Evaluation of Kamino using a high-fidelity simulator on large-scale production workloads shows a 42% reduction in average request latencies. Our deployment of Kamino in the control plane of a large public cloud confirms these improvements, with a 33% decrease in cache miss rates and 17% reduction in memory usage.
|abstract =Large language models (LLMs) achieve superior performance in generative tasks. However, due to the natural gap between language model generation and structured information extraction in three dimensions: task type, output format, and modeling granularity, they often fall short in structured information extraction, a crucial capability for effective data utilization on the web. In this paper, we define the generation process of the language model as the controllable state transition, aligning the generation and extraction processes to ensure the integrity of the output structure and adapt to the goals of the information extraction task. Furthermore, we propose the Structure2Text decider to help the language model understand the fine-grained extraction information, which converts the structured output into natural language and makes state decisions, thereby focusing on the task-specific information kernels, and alleviating language model hallucinations and incorrect content generation. We conduct extensive experiments and detailed analyses on myriad information extraction tasks, including named entity recognition, relation extraction, and event argument extraction. Our method not only achieves significant performance improvements but also considerably enhances the model's capability to generate precise and relevant content, making the extracted content easy to parse.
|confname =OSDI'25
|confname =WWW'25
|link = https://www.usenix.org/conference/osdi25/presentation/domingo
|link = https://dl.acm.org/doi/abs/10.1145/3696410.3714571
|title= Kamino: Efficient VM Allocation at Scale with Latency-Driven Cache-Aware Scheduling
|title= Bridging the Gap: Aligning Language Model Generation with Structured Information Extraction via Controllable State Transition
|speaker=Chenli
|speaker=Daobin
|date=2026-1-09
|date=2026-1-30
}}
}}
{{Resource:Previous_Seminars}}
{{Resource:Previous_Seminars}}

Latest revision as of 10:51, 30 January 2026

Time: 2026-01-30 10:30
Address: 4th Research Building A518
Useful links: 📚 Readling list; 📆 Schedules; 🧐 Previous seminars.

Latest

  1. [SenSys'25] MoLoRa: Intelligent Mobile Antenna System for Enhanced LoRa Reception in Urban Environments, Kai Chen
    Abstract: LoRa technology promises to enable Internet of Things applications over large geographical areas. However, its performance is often hampered by poor channel quality in urban environments, where blockage and multipath effects are prevalent. Our study uncovers that a slight shift in the position or attitude of the receiving antenna can substantially improve the received signal quality. This phenomenon can be attributed to the rich multipath characteristics of wireless signal propagation in urban environments, wherein even small antenna movement can alter the dominant signal path or reduce the polarization angular difference between transceivers. Leveraging these key observations, we propose and implement MoLoRa, an intelligent mobile antenna system designed to enhance LoRa packet reception. At its core, MoLoRa represents the position and attitude of an antenna as a state and employs a statistical optimization method to search for states that offer optimal signal quality efficiently. Through extensive evaluation, we demonstrate that MoLoRa achieves a maximum Signal-to-Noise Ratio (SNR) gain of 13 dB in a few attempts, enabling formerly problematic blind spots to reconnect and strengthening links for other nodes.
  2. [WWW'25] Bridging the Gap: Aligning Language Model Generation with Structured Information Extraction via Controllable State Transition, Daobin
    Abstract: Large language models (LLMs) achieve superior performance in generative tasks. However, due to the natural gap between language model generation and structured information extraction in three dimensions: task type, output format, and modeling granularity, they often fall short in structured information extraction, a crucial capability for effective data utilization on the web. In this paper, we define the generation process of the language model as the controllable state transition, aligning the generation and extraction processes to ensure the integrity of the output structure and adapt to the goals of the information extraction task. Furthermore, we propose the Structure2Text decider to help the language model understand the fine-grained extraction information, which converts the structured output into natural language and makes state decisions, thereby focusing on the task-specific information kernels, and alleviating language model hallucinations and incorrect content generation. We conduct extensive experiments and detailed analyses on myriad information extraction tasks, including named entity recognition, relation extraction, and event argument extraction. Our method not only achieves significant performance improvements but also considerably enhances the model's capability to generate precise and relevant content, making the extracted content easy to parse.

History

2024

2023

2022

2021

2020

  • [Topic] [ The path planning algorithm for multiple mobile edge servers in EdgeGO], Rong Cong, 2020-11-18

2019

2018

2017

Instructions

请使用Latest_seminar和Hist_seminar模板更新本页信息.

    • 修改时间和地点信息
    • 将当前latest seminar部分的code复制到这个页面中
    • 将{{Latest_seminar... 修改为 {{Hist_seminar...，并增加对应的日期信息|date=
    • 填入latest seminar各字段信息
    • link请务必不要留空，如果没有link则填本页地址 https://mobinets.org/index.php?title=Resource:Seminar
  • 格式说明
    • Latest_seminar:

{{Latest_seminar
|confname=
|link=
|title=
|speaker=
}}

    • Hist_seminar

{{Hist_seminar
|confname=
|link=
|title=
|speaker=
|date=
}}

Retrieved from "http://mobinets.cn/site/index.php?title=Resource:Seminar&oldid=3490"