Difference between revisions of "Resource:Seminar"

From MobiNetS
Jump to: navigation, search
 
(66 intermediate revisions by 3 users not shown)
Line 1: Line 1:
{{SemNote
{{SemNote
|time='''2024-10-25 10:30-12:00'''
|time='''2025-12-05 10:30'''
|addr=4th Research Building A533
|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 = Sparsely-activated Mixture-of-Expert (MoE) layers have found practical applications in enlarging the model size of large-scale foundation models, with only a sub-linear increase in computation demands. Despite the wide adoption of hybrid parallel paradigms like model parallelism, expert parallelism, and expert-sharding parallelism (i.e., MP+EP+ESP) to support MoE model training on GPU clusters, the training efficiency is hindered by communication costs introduced by these parallel paradigms. To address this limitation, we propose Parm, a system that accelerates MP+EP+ESP training by designing two dedicated schedules for placing communication tasks. The proposed schedules eliminate redundant computations and communications and enable overlaps between intra-node and inter-node communications, ultimately reducing the overall training time. As the two schedules are not mutually exclusive, we provide comprehensive theoretical analyses and derive an automatic and accurate solution to determine which schedule should be applied in different scenarios. Experimental results on an 8-GPU server and a 32-GPU cluster demonstrate that Parm outperforms the state-of-the-art MoE training system, DeepSpeed-MoE, achieving 1.13× to 5.77× speedup on 1296 manually configured MoE layers and approximately 3× improvement on two real-world MoE models based on BERT and GPT-2.
|abstract = Intermediate reasoning or acting steps have successfully improved large language models (LLMs) for handling various downstream natural language processing (NLP) tasks. When applying LLMs for code generation, recent works mainly focus on directing the models to articulate intermediate natural-language reasoning steps, as in chain-of-thought (CoT) prompting, and then output code with the natural language or other structured intermediate steps. However, such output is not suitable for code translation or generation tasks since the standard CoT has different logical structures and forms of expression with the code. In this work, we introduce the universal code (UniCode) as the intermediate representation. It is a description of algorithm steps using a mix of conventions of programming languages, such as assignment operator, conditional operator, and loop. Hence, we collect an instruction dataset UniCoder-Instruct to train our model UniCoder on multi-task learning objectives. UniCoder-Instruct comprises natural-language questions, code solutions, and the corresponding universal code. The alignment between the intermediate universal code representation and the final code solution significantly improves the quality of the generated code. The experimental results demonstrate that UniCoder with the universal code significantly outperforms the previous prompting methods by a large margin, showcasing the effectiveness of the structural clues in pseudo-code.
|confname =INFOCOM‘24
|confname =ACL'24
|link = https://ieeexplore.ieee.org/abstract/document/10621327
|link = https://arxiv.org/abs/2406.16441
|title= Parm: Efficient Training of Large Sparsely-Activated Models with Dedicated Schedules
|title= UniCoder: Scaling Code Large Language Model via Universal Code
|speaker=Mengqi
|speaker=Bairong Liu
|date=2024-11-1
|date=2025-12-05
}}
}}
{{Latest_seminar
{{Latest_seminar
|abstract = HD map is a key enabling technology towards fully autonomous driving. We propose VI-Map, the first system that leverages roadside infrastructure to enhance real-time HD mapping for autonomous driving. The core concept of VI-Map is to exploit the unique cumulative observations made by roadside infrastructure to build and maintain an accurate and current HD map. This HD map is then fused with on-vehicle HD maps in real time, resulting in a more comprehensive and up-to-date HD map. By extracting concise bird-eye-view features from infrastructure observations and utilizing vectorized map representations, VI-Map incurs low compute and communication overhead. We conducted end-to-end evaluations of VI-Map on a real-world testbed and a simulator. Experiment results show that VI-Map can construct decentimeter-level (up to 0.3 m) HD maps and achieve real-time (up to a delay of 42 ms) map fusion between driving vehicles and roadside infrastructure. This represents a significant improvement of 2.8× and 3× in map accuracy and coverage compared to the state-of-the-art online HD mapping approaches. A video demo of VI-Map on our real-world testbed is available at https://youtu.be/p2RO65R5Ezg.
|abstract =LoRaWANs are envisioned to connect billions of IoT devices through thousands of physically overlapping yet logically orthogonal channels (termed logical channels). These logical channels hold significant potential for enabling highly concurrent scalable IoT connectivity. Large-scale deployments however face strong interference between logical channels. This practical issue has been largely overlooked by existing works but becomes increasingly prominent as LoRaWAN scales up. To address this issue, we introduce Canas, an innovative gateway design that is poised to orthogonalize the logical channels by eliminating mutual interference. To this end, Canas develops a series of novel solutions to accurately extract the meta-information of individual ultra-weak LoRa signals from the received overlapping channels. The meta-information is then leveraged to accurately reconstruct and subtract the LoRa signals over thousands of logical channels iteratively. Real-world evaluations demonstrate that Canas can enhance concurrent transmissions across overlapping logical channels by 2.3× compared to the best known related works.
|confname=Mobicom'23
|confname =TMC'25
|link = https://dl.acm.org/doi/abs/10.1145/3570361.3613280
|link = https://ieeexplore.ieee.org/abstract/document/11160677
|title= VI-Map: Infrastructure-Assisted Real-Time HD Mapping for Autonomous Driving
|title= Resolving Inter-Logical Channel Interference for Large-scale LoRa Deployments
|speaker=Wangyang
|speaker=Mengyu
|date=2024-11-1
|date=2025-12-05
}}
}}
{{Resource:Previous_Seminars}}
{{Resource:Previous_Seminars}}

Latest revision as of 09:25, 5 December 2025

Time: 2025-12-05 10:30
Address: 4th Research Building A518
Useful links: 📚 Readling list; 📆 Schedules; 🧐 Previous seminars.

Latest

  1. [ACL'24] UniCoder: Scaling Code Large Language Model via Universal Code, Bairong Liu
    Abstract: Intermediate reasoning or acting steps have successfully improved large language models (LLMs) for handling various downstream natural language processing (NLP) tasks. When applying LLMs for code generation, recent works mainly focus on directing the models to articulate intermediate natural-language reasoning steps, as in chain-of-thought (CoT) prompting, and then output code with the natural language or other structured intermediate steps. However, such output is not suitable for code translation or generation tasks since the standard CoT has different logical structures and forms of expression with the code. In this work, we introduce the universal code (UniCode) as the intermediate representation. It is a description of algorithm steps using a mix of conventions of programming languages, such as assignment operator, conditional operator, and loop. Hence, we collect an instruction dataset UniCoder-Instruct to train our model UniCoder on multi-task learning objectives. UniCoder-Instruct comprises natural-language questions, code solutions, and the corresponding universal code. The alignment between the intermediate universal code representation and the final code solution significantly improves the quality of the generated code. The experimental results demonstrate that UniCoder with the universal code significantly outperforms the previous prompting methods by a large margin, showcasing the effectiveness of the structural clues in pseudo-code.
  2. [TMC'25] Resolving Inter-Logical Channel Interference for Large-scale LoRa Deployments, Mengyu
    Abstract: LoRaWANs are envisioned to connect billions of IoT devices through thousands of physically overlapping yet logically orthogonal channels (termed logical channels). These logical channels hold significant potential for enabling highly concurrent scalable IoT connectivity. Large-scale deployments however face strong interference between logical channels. This practical issue has been largely overlooked by existing works but becomes increasingly prominent as LoRaWAN scales up. To address this issue, we introduce Canas, an innovative gateway design that is poised to orthogonalize the logical channels by eliminating mutual interference. To this end, Canas develops a series of novel solutions to accurately extract the meta-information of individual ultra-weak LoRa signals from the received overlapping channels. The meta-information is then leveraged to accurately reconstruct and subtract the LoRa signals over thousands of logical channels iteratively. Real-world evaluations demonstrate that Canas can enhance concurrent transmissions across overlapping logical channels by 2.3× compared to the best known related works.

History

|abstract =The rapid expansion of large language models (LLMs) requires the development of extensive GPU clusters, with companies deploying clusters with tens to hundreds of thousands of GPUs. This growth significantly expands the design space for LLM training systems, requiring thorough exploration of different parallelization strategies, communication parameters, congestion control, fabric topology, etc. Current methods require up to 10k simulation experiments to identify optimal configurations, with inadequate exploration leading to significant degradation of training performance. In this paper, we tackle the overlooked problem of efficiently conducting parallel simulation experiments for design space exploration. Our

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=3454"