Difference between revisions of "Resource:Seminar"

From MobiNetS
Jump to: navigation, search
 
(123 intermediate revisions by 4 users not shown)
Line 1: Line 1:
{{SemNote
{{SemNote
|time='''Thursday 9:00-10:30'''
|time='''2025-12-12 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]].
}}
}}


===Latest===
===Latest===
{{Latest_seminar
{{Latest_seminar
|abstract=The massive connection of LoRa brings serious collision interference. Existing collision decoding methods cannot effectively deal with the adjacent collisions that occur when the collided symbols are adjacent in the frequency spectrum. The decoding features relied on by the existing methods will be corrupted by adjacent collisions. To address these issues, we propose Paralign, which is the first LoRa collision decoder supporting decoding LoRa collisions with confusing symbols via parallel alignment. The key enabling technology behind Paralign is tha there is no spectrum leakage with periodic truncation of a chirp. Paralign leverages the precise spectrum obtained by aligning the de-chirped periodic signals from each packet in parallel for spectrum filtering and power filtering. To aggregate correlation peaks in different windows of the same symbol, Paralign matches the peaks of multiple interfering windows to the interested window based on the time offset between collided packets. Moreover, a periodic truncation method is proposed to address the multiple candidate peak problem caused by side lobes of confusing symbols. We evaluate Paralign using USRP N210 in a 20-node network. Experimental results demonstrate that Paralign can significantly improve network throughput, which is over 1.46× higher than state-of-the-art methods.
|abstract = Code translation is a crucial activity in the software development and maintenance process, and researchers have recently begun to focus on using pre-trained large language models (LLMs) for code translation. However, existing LLMs only learn the contextual semantics of code during pre-training, neglecting executability information closely related to the execution state of the code, which results in unguaranteed code executability and unreliable automated code translation. To address this issue, we propose ExeCoder, an LLM specifically designed for code translation, aimed at utilizing executability representations such as functional semantics, syntax structures, and variable dependencies to enhance the capabilities of LLMs in code translation. To evaluate the effectiveness of ExeCoder, we manually enhanced the widely used benchmark TransCoder-test, resulting in a benchmark called TransCoder-test-X that serves LLMs. Evaluation of TransCoder-test-X indicates that ExeCoder achieves state-of-the-art performance in code translation, surpassing existing open-source code LLMs by over 10.88% to 38.78% and over 27.44% to 42.97% on two metrics, and even outperforms the renowned closed-source LLM GPT-4o.  
|confname=ToSN '23
|confname =EMNLP'25
|link=https://dl.acm.org/doi/10.1145/3571586
|link = https://arxiv.org/abs/2501.18460
|title=Decoding LoRa Collisions via Parallel Alignment
|title= ExeCoder: Empowering Large Language Models with Executability Representation for Code Translation
|speaker=Kai Chen
|speaker=Youwei Ran
|date=2024-01-04}}
|date=2025-12-12
}}
{{Latest_seminar
{{Latest_seminar
|abstract=Human activity recognition is important for a wide range of applications such as surveillance systems and human-computer interaction. Computer vision based human activity recognition suffers from performance degradation in many real-world scenarios where the illumination is poor. On the other hand, recently proposed WiFi sensing that leverage ubiquitous WiFi signal for activity recognition is not affected by illumination but has low accuracy in dynamic environments. In this paper, we propose WiMix, a lightweight and robust multimodal system that leverages both WiFi and vision for human activity recognition. To deal with complex real-world environments, we design a lightweight mix cross attention module for automatic WiFi and video weight distribution. To reduce the system response time while ensuring the sensing accuracy, we design an end-to-end framework together with an efficient classifier to extract spatial and temporal features of two modalities. Extensive experiments are conducted in the real-world scenarios and the results demonstrate that WiMix achieves 98.5% activity recognition accuracy in 3 scenarios, which outperforms the state-of-the-art 89.6% sensing accuracy using WiFi and video modalities. WiMix can also reduce the inference latency from 1268.25ms to 217.36ms, significantly improving the response time.
|abstract =Imitation learning from human demonstrations has shown impressive performance in robotics. However, most results focus on table-top manipulation, lacking the mobility and dexterity necessary for generally useful tasks. In this work, we develop a system for imitating mobile manipulation tasks that are bimanual and require whole-body control. We first present Mobile ALOHA, a low-cost and whole-body teleoperation system for data collection. It augments the ALOHA system with a mobile base, and a whole-body teleoperation interface. Using data collected with Mobile ALOHA, we then perform supervised behavior cloning and find that co-training with existing static ALOHA datasets boosts performance on mobile manipulation tasks. With 50 demonstrations for each task, co-training can increase success rates by up to 90%, allowing Mobile ALOHA to autonomously complete complex mobile manipulation tasks such as sauteing and serving a piece of shrimp, opening a two-door wall cabinet to store heavy cooking pots, calling and entering an elevator, and lightly rinsing a used pan using a kitchen faucet. We will open-source all the hardware and software implementations upon publication.
|confname=MASS '23
|confname =CoRL'24
|link=https://ieeexplore.ieee.org/abstract/document/10298524
|link = https://openreview.net/forum?id=FO6tePGRZj
|title=WiMix: A Lightweight Multimodal Human Activity Recognition System based on WiFi and Vision
|title= Mobile ALOHA: Learning Bimanual Mobile Manipulation using Low-Cost Whole-Body Teleoperation
|speaker=Haotian
|speaker=Yi Zhou
|date=2024-01-04}}
|date=2025-12-12
{{Latest_seminar
}}
|abstract=In wireless networks, the Named Data Networking (NDN) architecture maximizes contents’ availability throughout multiple network paths based on multicast-based communication combined with stateful forwarding and caching in intermediate nodes. Despite its benefits, the downside of the architecture resides in the packet flooding not efficiently prevented by current forwarding strategies and mainly associated with the native flooding of the architecture or malicious packets from Interest Flooding Attack (IFA). This work introduces iFLAT, a multicriteria-based forwarding strategy for i nterest FL ooding mitig AT ion on named data wireless networking. It follows a cross-layer approach that considers: (i) the Received Signal Strength (RSS) to adjust itself to the current status of the wireless network links, (ii) the network traffic ( meInfo ) to detect flooding issues and traffic anomalies, and (iii) a fake Interest Blacklist to identify IFA-related flooding. In doing so, the strategy achieves better efficiency on Interest flooding mitigation, addressing both native and IFA types of flooding. We evaluated the proposed strategy by reproducing a Flying Ad hoc Network (FANET) composed of Unmanned Aerial Vehicles (UAVs) featured by the NDN stack deployed in all nodes. Simulation results show that iFLAT can effectively detect and mitigate flooding, regardless of its origin or nature, achieving greater packet forwarding efficiency than competing strategies. In broadcast storm and IFA scenarios, it achieved 25.75% and 37.37% of traffic reduction, whereas Interest satisfaction rates of 16.36% and 51.78% higher, respectively.
|confname=TMC '23
|link=https://ieeexplore.ieee.org/document/9888056
|title=A Multicriteria-Based Forwarding Strategy for Interest Flooding Mitigation on Named Data Wireless Networking
|speaker=Zhenghua
|date=2024-01-04}}
{{Latest_seminar
|abstract=While recent work explored streaming volumetric content on-demand, there is little effort on live volumetric video streaming that bears the potential of bringing more exciting applications than its on-demand counterpart. To fill this critical gap, in this paper, we propose MetaStream, which is, to the best of our knowledge, the first practical live volumetric content capture, creation, delivery, and rendering system for immersive applications such as virtual, augmented, and mixed reality. To address the key challenge of the stringent latency requirement for processing and streaming a huge amount of 3D data, MetaStream integrates several innovations into a holistic system, including dynamic camera calibration, edge-assisted object segmentation, cross-camera redundant point removal, and foveated volumetric content rendering. We implement a prototype of MetaStream using commodity devices and extensively evaluate its performance. Our results demonstrate that MetaStream achieves low-latency live volumetric video streaming at close to 30 frames per second on WiFi networks. Compared to state-of-the-art systems, MetaStream reduces end-to-end latency by up to 31.7% while improving visual quality by up to 12.5%.
|confname=MobiCom '23
|link=https://dl.acm.org/doi/abs/10.1145/3570361.3592530
|title=MetaStream: Live Volumetric Content Capture, Creation, Delivery, and Rendering in Real Time
|speaker=Jiale
|date=2024-01-04}}
{{Resource:Previous_Seminars}}
{{Resource:Previous_Seminars}}

Latest revision as of 23:32, 11 December 2025

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

Latest

  1. [EMNLP'25] ExeCoder: Empowering Large Language Models with Executability Representation for Code Translation, Youwei Ran
    Abstract: Code translation is a crucial activity in the software development and maintenance process, and researchers have recently begun to focus on using pre-trained large language models (LLMs) for code translation. However, existing LLMs only learn the contextual semantics of code during pre-training, neglecting executability information closely related to the execution state of the code, which results in unguaranteed code executability and unreliable automated code translation. To address this issue, we propose ExeCoder, an LLM specifically designed for code translation, aimed at utilizing executability representations such as functional semantics, syntax structures, and variable dependencies to enhance the capabilities of LLMs in code translation. To evaluate the effectiveness of ExeCoder, we manually enhanced the widely used benchmark TransCoder-test, resulting in a benchmark called TransCoder-test-X that serves LLMs. Evaluation of TransCoder-test-X indicates that ExeCoder achieves state-of-the-art performance in code translation, surpassing existing open-source code LLMs by over 10.88% to 38.78% and over 27.44% to 42.97% on two metrics, and even outperforms the renowned closed-source LLM GPT-4o.
  2. [CoRL'24] Mobile ALOHA: Learning Bimanual Mobile Manipulation using Low-Cost Whole-Body Teleoperation, Yi Zhou
    Abstract: Imitation learning from human demonstrations has shown impressive performance in robotics. However, most results focus on table-top manipulation, lacking the mobility and dexterity necessary for generally useful tasks. In this work, we develop a system for imitating mobile manipulation tasks that are bimanual and require whole-body control. We first present Mobile ALOHA, a low-cost and whole-body teleoperation system for data collection. It augments the ALOHA system with a mobile base, and a whole-body teleoperation interface. Using data collected with Mobile ALOHA, we then perform supervised behavior cloning and find that co-training with existing static ALOHA datasets boosts performance on mobile manipulation tasks. With 50 demonstrations for each task, co-training can increase success rates by up to 90%, allowing Mobile ALOHA to autonomously complete complex mobile manipulation tasks such as sauteing and serving a piece of shrimp, opening a two-door wall cabinet to store heavy cooking pots, calling and entering an elevator, and lightly rinsing a used pan using a kitchen faucet. We will open-source all the hardware and software implementations upon publication.

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=3465"