Search

Your search keyword '"Ma, Dekun"' showing total 205 results
Start Over Author "Ma, Dekun"
205 results on '"Ma, Dekun"'

1. Into the Unknown Unknowns: Engaged Human Learning through Participation in Language Model Agent Conversations

Author

Jiang, Yucheng, Shao, Yijia, Ma, Dekun, Semnani, Sina J., and Lam, Monica S.

Subjects
Computer Science - Computation and Language, Computer Science - Artificial Intelligence, Computer Science - Information Retrieval, I.2.7, H.5.2, H.3.3
Abstract

While language model (LM)-powered chatbots and generative search engines excel at answering concrete queries, discovering information in the terrain of unknown unknowns remains challenging for users. To emulate the common educational scenario where children/students learn by listening to and participating in conversations of their parents/teachers, we create Collaborative STORM (Co-STORM). Unlike QA systems that require users to ask all the questions, Co-STORM lets users observe and occasionally steer the discourse among several LM agents. The agents ask questions on the user's behalf, allowing the user to discover unknown unknowns serendipitously. To facilitate user interaction, Co-STORM assists users in tracking the discourse by organizing the uncovered information into a dynamic mind map, ultimately generating a comprehensive report as takeaways. For automatic evaluation, we construct the WildSeek dataset by collecting real information-seeking records with user goals. Co-STORM outperforms baseline methods on both discourse trace and report quality. In a further human evaluation, 70% of participants prefer Co-STORM over a search engine, and 78% favor it over a RAG chatbot., Comment: EMNLP 2024 Main

Published
2024

4. V2X-Sim: Multi-Agent Collaborative Perception Dataset and Benchmark for Autonomous Driving

Author

Li, Yiming, Ma, Dekun, An, Ziyan, Wang, Zixun, Zhong, Yiqi, Chen, Siheng, and Feng, Chen

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Vehicle-to-everything (V2X) communication techniques enable the collaboration between vehicles and many other entities in the neighboring environment, which could fundamentally improve the perception system for autonomous driving. However, the lack of a public dataset significantly restricts the research progress of collaborative perception. To fill this gap, we present V2X-Sim, a comprehensive simulated multi-agent perception dataset for V2X-aided autonomous driving. V2X-Sim provides: (1) \hl{multi-agent} sensor recordings from the road-side unit (RSU) and multiple vehicles that enable collaborative perception, (2) multi-modality sensor streams that facilitate multi-modality perception, and (3) diverse ground truths that support various perception tasks. Meanwhile, we build an open-source testbed and provide a benchmark for the state-of-the-art collaborative perception algorithms on three tasks, including detection, tracking and segmentation. V2X-Sim seeks to stimulate collaborative perception research for autonomous driving before realistic datasets become widely available. Our dataset and code are available at \url{https://ai4ce.github.io/V2X-Sim/}., Comment: 2022 IEEE Robotics and Automation Letters (RA-L) (The extended abstract is presented at 2021 IEEE International Conference on Computer Vision (ICCV) Simulation Technology for Embodied AI Workshop)

Published
2022

15. All‐In‐One Photonic Crystals With Multi‐Stimuli‐Chromic, Color‐Recordable, Self‐Healable, and Adhesive Functions.

Author

Yu, Siyi, Ma, Dekun, Qi, Chenze, Yang, Dongpeng, and Huang, Shaoming

Subjects
*STRUCTURAL colors, *SMART materials, *PHOTONIC crystals, *LIGHT filters, *POLYMER networks, *THERMOCHROMISM
Abstract

Fabricating photonic crystals (PCs) with dynamic multi‐stimuli‐responsive structural colors, recordable colors, and self‐healable properties is significant for their emerging applications, yet remains a big challenge. Here, an all‐in‐one multifunctional PC (MFPC) film with outstanding mechanochromic, thermochromic, solvatochromic, color/shape‐recordable, self‐healable, and adhesive functions is designed and prepared by simply non‐close‐assembling silica particles into the unique 2‐[[(Butylamino)carbonyl]oxy]ethyl acrylate (BCOEA) followed by photopolymerization. The success is mainly due to the rational combination of non‐close‐packing structures and BCOEA's characteristics, including its urethane groups with numerous sacrificial hydrogen bonds, temperature‐dependent refractive index, swellable and deformable polymer network, as well as low glass‐transition temperature. The MFPC's hue and color saturation can be reversibly and dynamically modulated by strain/pressure/solvents and temperature, respectively, thereby realizing visually and spectrally sensing these stimuli. More interestingly, the color‐responsiveness combined with other functions endows MFPCs with fascinating emerging applications, including multilayer optical filters, inkless printing, reconfigurable multicolor patterns, stretching‐based anti‐counterfeiting, etc. This work offers a new perspective for designing next‐generation smart photonic materials and will facilitate their all‐round applications. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

24. Structural Color‐Based Smart Liquid Windows Address the Tradeoff Between High Optical Transparency and Brilliant Color.

Author

Hu, Yang, Yang, Dongpeng, Ma, Dekun, Qi, Chenze, and Huang, Shaoming

Subjects
ELECTROCHROMIC windows, STRUCTURAL colors, OPTICAL diffraction, DIFFRACTIVE scattering, LIGHT scattering, REFRACTIVE index
Abstract

It is challenging for traditional smart windows to possess both high optical transparency (90%, 380–780 nm) and brilliant colors since they are usually contradictory to each other. Here, novel liquid thermal‐responsive smart windows (TSWs) exhibiting high optical transparency (94%), bright iridescent forward scattering structural color, and extremely weak reflective structural colors are fabricated by constructing a refractive index (RI)‐matched non‐close‐packing ordered structure with large lattice distances and thus unique light diffraction mechanisms. The transmittance (24‐94%) and dual‐mode structural color saturation (1–55 times) can be dynamically regulated by simply altering temperatures (20–100 °C), which can be attributed to the temperature‐dependent light diffraction and scattering efficiency. A new information encryption‐decryption technique is developed by combining TSWs with outstanding shapeability, recyclability, and flowability. The information is hidden under normal conditions, while 76 different pieces of information can be decrypted in programmable ways under different stimuli, showing their potential in anti‐counterfeiting and information encryption. [ABSTRACT FROM AUTHOR]

Published
2024
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results