![\"\"](\"https:\/\/www.inforsec.org\/wp\/wp-content\/uploads\/2020\/07\/56-2.png\")
<\/figure><\/div>\n\n\n\n![\"\"](\"https:\/\/www.inforsec.org\/wp\/wp-content\/uploads\/2020\/07\/2-2.png\")
<\/figure><\/div>\n\n\n\n![\"\"](\"https:\/\/www.inforsec.org\/wp\/wp-content\/uploads\/2020\/07\/15948231331-4.jpg\")
<\/figure><\/div>\n\n\n\n\u8bf7\u6709\u610f\u5411\u53c2\u4e0e\u590f\u4ee4\u8425\u6d3b\u52a8\u5e76\u8054\u7cfb\u76f8\u5e94\u7684\u5bfc\u5e08\u7684\u540c\u5b66\u8bbf\u95eehttps:\/\/www.inforsec.org\/wp\/?p=3966<\/a>\uff0c\u4e0b\u8f7d\u5e76\u63d0\u4ea4\u76f8\u5173\u8d44\u6599\u3002\u6d3b\u52a8\u5feb\u6377\u62a5\u540d\uff0c\u8bf7\u70b9\u51fb\u5e76\u8bc6\u522b\u6b63\u6587\u4e0b\u65b9\u4e8c\u7ef4\u7801<\/strong>\u3002\u62a5\u540d\u622a\u6b62\u65e5\u671f\uff1a2020\u5e747\u67085\u65e5<\/strong>\u3002 \u6f14\u8bb2\u5609\u5bbe<\/strong> \u7eaa\u5b88\u9886<\/strong><\/p>\n\n\n\n \u6d59\u6c5f\u5927\u5b66\u7814\u7a76\u5458<\/strong><\/p>\n\n\n\n \u6f14\u8bb2\u4e3b\u9898\uff1aRevisited Deep Learning Security: Empirical Analysis and Robustness Quantification<\/strong><\/p>\n\n\n\n \u5bfc\u5e08\u4ecb\u7ecd\uff1a<\/strong><\/p>\n\n\n\n \u7eaa\u5b88\u9886\u83b7\u4f50\u6cbb\u4e9a\u7406\u5de5\u5b66\u9662\u7535\u5b50\u4e0e\u8ba1\u7b97\u673a\u5de5\u7a0b\u535a\u58eb\u5b66\u4f4d\u3001\u4f50\u6cbb\u4e9a\u5dde\u7acb\u5927\u5b66\u8ba1\u7b97\u673a\u79d1\u5b66\u535a\u58eb\u5b66\u4f4d\uff0c\u73b0\u4efb\u6d59\u6c5f\u5927\u5b66\u201c\u767e\u4eba\u8ba1\u5212\u201d\u7814\u7a76\u5458\u3001\u535a\u58eb\u751f\u5bfc\u5e08\uff0c\u517c\u4efb\u4f50\u6cbb\u4e9a\u7406\u5de5\u5b66\u9662Research Faculty\u3002\u4e3b\u8981\u7814\u7a76\u65b9\u5411\u4e3a\u4eba\u5de5\u667a\u80fd\u5b89\u5168\u3001\u6570\u636e\u9a71\u52a8\u5b89\u5168\u3001\u8f6f\u4ef6\u4e0e\u7cfb\u7edf\u5b89\u5168\u3001\u5927\u6570\u636e\u5206\u6790\uff0c\u53d1\u8868\u8bba\u6587100\u4f59\u7bc7\u3002\u66fe\u83b7\u4e2d\u56fd\u56fd\u5bb6\u4f18\u79c0\u7559\u5b66\u751f\u5956\u3001\u4e03\u9879\u6700\u4f73\u8bba\u6587\u5956\u3001GSU\u6770\u51fa\u7814\u7a76\u5956\u7b49\u3002<\/p>\n\n\n\n \u5185\u5bb9\u6458\u8981\uff1a<\/strong><\/p>\n\n\n\n Deep learning security has drawn significant attention from the academia. Various kinds of attacks and defenses spring up like bamboo shoots. Given this, two interesting questions have raised: (1) what is the performance of the attacks against real world systems? and (2) how to understand\/quantify the security space of deep models? Towards this, we recently revisited the transferability of adversarial examples against real world systems and studied the robustness for DNNs. In this talk, I will introduce some interesting results from these two perspectives.<\/p>\n\n\n\n \u6797\u5fd7\u5f3a<\/strong><\/p>\n\n\n\n \u4fc4\u4ea5\u4fc4\u5dde\u7acb\u5927\u5b66\u526f\u6559\u6388<\/strong><\/p>\n\n\n\n \u6f14\u8bb2\u4e3b\u9898\uff1aUnveiling IoT Security from Bluetooth Low Energy<\/strong><\/p>\n\n\n\n \u5bfc\u5e08\u4ecb\u7ecd:<\/strong><\/p>\n\n\n\n Dr. Zhiqiang Lin is an Associate Professor of Computer Science and Engineering at The Ohio State University (OSU). He is also a faculty member at Translational Data Analytics Institute (TDAI), Center for Automotive Research (CAR), and the recently launched Institute for Cybersecurity and Digital Trust (ICDT) at OSU. His primary research interests are trusted computing, systems and software security, with an emphasis on developing automated program analysis and reverse engineering techniques and applying them to secure both the application programs including mobile apps and the underlying system software such as OS kernels and hypervisors. Dr. Lin is a recipient of both NSF CAREER award and AFOSR Young Investigator award.<\/p>\n\n\n\n \u5185\u5bb9\u6458\u8981\uff1a<\/strong><\/p>\n\n\n\n Being an easy-to-deploy and cost-effective low power wireless solution, Bluetooth Low Energy (BLE) has been used by billions of Internet-of-Things (IoT) devices for various applications in many areas such as transportation, healthcare (e.g., recently on COVID19 contact tracing), wearables, retail, and smart homes. However, numerous vulnerabilities have also been identified in BLE IoT. In this talk, Dr. Lin will talk about a line of recent research from his group on BLE IoT security. In particular, he will first about the internals of how BLE IoT works including the corresponding protocols, then talk about the various flaws in BLE stacks from the protocol specification, to the design and implementation at the operating systems and applications, and finally he will talk about future works in this direction.<\/p>\n\n\n\n \u5218\u70c3<\/strong><\/p>\n\n\n\n \u897f\u5b89\u4ea4\u901a\u5927\u5b66\u7f51\u7edc\u7a7a\u95f4\u5b89\u5168\u5b66\u9662\u526f\u9662\u957f\u3001\u6559\u6388<\/strong><\/p>\n\n\n\n \u6f14\u8bb2\u4e3b\u9898\uff1a\u667a\u80fd\u8f6f\u4ef6\u5de5\u7a0b<\/strong><\/p>\n\n\n\n \u5bfc\u5e08\u4ecb\u7ecd\uff1a<\/strong><\/p>\n\n\n\n \u5218\u70c3\uff0c\u535a\u58eb\uff0c\u6559\u6388\uff0c\u897f\u5b89\u4ea4\u901a\u5927\u5b66\u7f51\u7edc\u7a7a\u95f4\u5b89\u5168\u5b66\u9662\u526f\u9662\u957f\u3002\u4e3b\u8981\u7814\u7a76\u65b9\u5411\u5305\u62ec\u8f6f\u4ef6\u5de5\u7a0b\u548c\u4fe1\u606f\u7269\u7406\u878d\u5408\u7cfb\u7edf\u30022003\u5e74\u548c2010\u5e74\u5206\u522b\u83b7\u897f\u5b89\u4ea4\u901a\u5927\u5b66\u5b66\u58eb\u548c\u535a\u58eb\u5b66\u4f4d\uff1b2016-2017\u5e74\u7f8e\u56fd\u5eb7\u5948\u5c14\u5927\u5b66\u8bbf\u95ee\u6559\u6388\u3002\u83b7\u5f972017\u5e74\u56fd\u5bb6\u79d1\u6280\u8fdb\u6b65\u4e8c\u7b49\u5956\u3001\u4ee5\u53ca2015\u5e74\u6559\u80b2\u90e8\u79d1\u6280\u8fdb\u6b65\u4e00\u7b49\u5956\u7b49\u7701\u90e8\u7ea7\u79d1\u6280\u5956\u52b15\u9879\u3002\u5728IEEE TIFS\u3001IEEE TSG\u3001IEEE TSE\u7b49\u56fd\u9645\u6743\u5a01\u671f\u520a\u548cICSE\u3001FSE\u3001ASE\u3001ISSTA\u3001INFOCOM\u7b49\u56fd\u9645\u9876\u7ea7\u4f1a\u8bae\u4e0a\u53d1\u8868\u8bba\u658770\u4f59\u7bc7\uff0c\u83b7\u5f972019 IEEE INFOCOM\u30012016 IEEE ISSRE\u7b49\u4f1a\u8bae\u6700\u4f73\u8bba\u6587\u5956\u3002<\/p>\n\n\n\n \u5185\u5bb9\u6458\u8981\uff1a<\/strong><\/p>\n\n\n\n \u4ee3\u7801\u81ea\u52a8\u751f\u6210\u548c\u7ef4\u62a4\u662f\u8f6f\u4ef6\u5de5\u7a0b\u9886\u57df\u5386\u53f2\u6700\u60a0\u4e45\u548c\u6700\u53d7\u5173\u6ce8\u7684\u65b9\u5411\u4e4b\u4e00\u3002\u8fd1\u5e74\u6765\uff0c\u5feb\u901f\u53d1\u5c55\u7684\u4eba\u5de5\u667a\u80fd\u6280\u672f\u4f7f\u5f97\u673a\u5668\u5bf9\u4ee3\u7801\u548c\u4eba\u7c7b\u9700\u6c42\u7684\u7406\u89e3\u80fd\u529b\u5f97\u5230\u663e\u8457\u63d0\u5347\uff0c\u4e3a\u4ee3\u7801\u81ea\u52a8\u751f\u6210\u548c\u7ef4\u62a4\u63d0\u4f9b\u4e86\u65b0\u7684\u601d\u8def\u3002\u672c\u62a5\u544a\u4ee5\u81ea\u7136\u8bed\u8a00\u7406\u89e3\u9a71\u52a8\u7684\u8f6f\u4ef6\u81ea\u52a8\u751f\u6210\u548c\u7ef4\u62a4\u4e3a\u4f8b\uff0c\u63a2\u8ba8\u4eba\u5de5\u667a\u80fd\u6280\u672f\u5728\u8f6f\u4ef6\u5de5\u7a0b\u9886\u57df\u7684\u5e94\u7528\u3002<\/p>\n\n\n\n \u6881\u632f\u51ef<\/strong><\/p>\n\n\n\n \u65b0\u52a0\u5761\u56fd\u7acb\u5927\u5b66\u526f\u6559\u6388<\/strong><\/p>\n\n\n\n \u6f14\u8bb2\u9898\u76ee\uff1a\u7406\u89e3\u7cfb\u7edf\u6267\u884c\u73af\u5883\u4e2d\u7684\u7075\u6d3b\u5ea6\u4e0e\u5b89\u5168\u95ee\u9898<\/strong><\/p>\n\n\n\n \u5bfc\u5e08\u4ecb\u7ecd\uff1a<\/strong><\/p>\n\n\n\n Dr. LIANG Zhenkai is an Associate Professor in the Department of Computer Science at National University of Singapore. He is also a co-Lead Principal Investigator of National Security R&D Lab of Singapore. His research interests are in system and software security, such as binary program analysis, security in Web, mobile, and Internet-of-things (IoT) platforms. He has been publishing high-impact papers in top security and software engineering conferences, and has won seven best paper awards, including Annual Computer Security Applications Conference (ACSAC), USENIX Security Symposium, and ACM SIGSOFT Symposium on the Foundations of Software Engineering (FSE). He has been actively served as technical committee members and editorial board members of top security conferences and journals, including ACM Conference on Computer and Communications Security (CCS), USENIX Security Symposium, Network and Distributed System Security Symposium (NDSS), and IEEE Transactions on Dependable and Secure Computing (TDSC). He is also a member of the Steering Group of NDSS. He has won the Annual Teaching Excellence Award of NUS in 2014 and 2015. He received his Ph.D. degree in Computer Science from Stony Brook University in 2006, and B.S. degrees in Computer Science and Economics from Peking University in 1999.<\/p>\n\n\n\n \u5f20\u950b\u5dcd<\/strong><\/p>\n\n\n\n \u5357\u65b9\u79d1\u6280\u5927\u5b66\u526f\u6559\u6388<\/strong><\/p>\n\n\n\n \u6f14\u8bb2\u4e3b\u9898\uff1aUsing Hardware Features for (In)Security on Arm<\/strong><\/p>\n\n\n\n \u5bfc\u5e08\u4ecb\u7ecd\uff1a<\/strong> \u5f20\u950b\u5dcd\u8001\u5e08\u662f\u5357\u65b9\u79d1\u6280\u5b66\u8ba1\u7b97\u673a\u79d1\u5b66\u4e0e\u5de5\u7a0b\u7cfb\u526f\u6559\u6388\uff0c\u7814\u7a76\u5458\u3002\u5728\u52a0\u5165\u5357\u79d1\u5927\u4e4b\u524d\uff0c\u4ed6\u662f\u7f8e\u56fd\u97e6\u6069\u5dde\u7acb\u5927\u5b66\u8ba1\u7b97\u673a\u7cfb\u52a9\u7406\u6559\u6388\uff0c\u8ba1\u7b97\u673a\u548c\u7cfb\u7edf\u5b89\u5168\u5b9e\u9a8c\u5ba4\u7684\u4e3b\u4efb\uff082015-2019\uff09\u3002\u4ed6\u4e8e2015\u5e74\u83b7\u5f97\u7f8e\u56fd\u4e54\u6cbb\u6885\u68ee\u5927\u5b66\u8ba1\u7b97\u673a\u4e13\u4e1a\u7684\u535a\u58eb\u5b66\u4f4d\u3002\u4e3b\u8981\u7814\u7a76\u9886\u57df\u662f\u7cfb\u7edf\u5b89\u5168\uff0c\u5305\u62ec\u53ef\u4fe1\u6267\u884c\uff08trustworthy execution\uff09\u3001\u786c\u4ef6\u8f85\u52a9\u5b89\u5168\uff08hardware-assisted security\uff09\u3001\u6076\u610f\u8f6f\u4ef6\u900f\u660e\u5206\u6790\uff08transparent malware debugging\uff09\u3001\u4ea4\u901a\u5b89\u5168\uff08transportation security\uff09\uff0c\u4ee5\u53ca\u53ef\u5426\u8ba4\u52a0\u5bc6\uff08plausible deniability encryption\uff09\u3002\u5f20\u8001\u5e08\u5df2\u7ecf\u53d1\u8868\u4e8640\u4f59\u7bc7\u56fd\u9645\u4f1a\u8bae\u548c\u671f\u520a\u8bba\u6587\uff0c\u5305\u62ecIEEE S&P, USENIX Security, NDSS, IEEE TIFS\u548cIEEE TDSC\u3002\u5f20\u8001\u5e08\u662f\u591a\u4e2a\u56fd\u9645\u9876\u7ea7\u4f1a\u8bae\u7a0b\u5e8f\u59d4\u5458\u4f1a\u7684\u6210\u5458\uff0c\u5305\u62ecIEEE S&P\uff0cACM CCS\uff0cACSAC\uff0cIEEE\/IFIP DSN\u7b49\u3002\u5f20\u8001\u5e08\u662fACSAC 2017\u5e74\u6770\u51fa\u8bba\u6587\u5956\u7684\u83b7\u5f97\u8005\u548c\u5165\u9009IEEE\/IFIP DSN 2020\u6700\u4f73\u8bba\u6587\u63d0\u540d\u3002\u4ed6\u7684\u7814\u7a76\u5de5\u4f5c\u66fe\u83b7\u5f97\u4e86\u591a\u9879\u7f8e\u56fd\u81ea\u7136\u57fa\u91d1\u9879\u76ee\u7684\u652f\u6301\u3002<\/p>\n\n\n\n \u5185\u5bb9\u6458\u8981\uff1a<\/strong><\/p>\n\n\n\n Processors nowadays are consistently equipped with debugging features to facilitate the program debugging and analysis. Although the debugging architecture has been presented for years, the security of the debugging features is under-examined since it normally requires physical access to use these features in the traditional debugging model. Arm introduces a new debugging model that requires no physical access since Armv7. In this new debugging model, a host processor is able to pause and debug another target processor on the same chip (inter-processor debugging). The idea of Nailgun attack is to misuse the debugging architecture with the inter-processor debugging since recent Arm systems allow the debug host to pause and debug the target even when the target owns a higher privilege. Our experiments discover a number of vulnerable devices including IoT devices like Raspberry PI, all commercial Arm-based cloud platforms, and mobile phones from Huawei, Motorola, and Xiaomi. For further verification, we show that Nailgun attack can be used to access the Secure Configuration Register (which is only accessible in the secure state) on Raspberry PI and extract the fingerprint image stored in the secure memory of a mobile phone with a non-secure kernel module.<\/p>\n\n\n\n \u94b1\u5fd7\u4e91<\/strong><\/p>\n\n\n\n \u52a0\u5dde\u5927\u5b66\u6cb3\u6ee8\u5206\u6821\u526f\u6559\u6388<\/strong><\/p>\n\n\n\n \u6f14\u8bb2\u4e3b\u9898\uff1aSymTCP: Eluding Stateful Deep Packet Inspection with Automated Discrepancy Discovery<\/strong><\/p>\n\n\n\n \u5bfc\u5e08\u4ecb\u7ecd\uff1a<\/strong><\/p>\n\n\n\n Zhiyun Qian is the Everett and Imogene Ross associate professor in the CSE department at the University of California Riverside. He is a recipient of the NSF CAREER Award for 2017. His main research interests are in the area of system and network security, including vulnerability discovery, side channel analysis, applied program analysis, system building, and measurement of real-world security problems. He is a recipient of the NSF CAREER Award in 2017, Applied Networking Research Prize from IRTF in 2019, Facebook Internet<\/p>\n\n\n\n Defense Prize Finalist in 2016, and the most creative idea award from Geekpwn 2016.<\/p>\n\n\n\n \u5185\u5bb9\u6458\u8981: <\/strong><\/p>\n\n\n\n A key characteristic of commonly deployed deep packet inspection (DPI) systems is that they implement a simplified state machine of the network stack that often differs from that of endhosts. The discrepancies between the two state machines have been exploited to bypass such DPI based middleboxes. However, most prior approaches to do so rely on manually crafted adversarial packets, which not only are labor-intensive but may not work well across a plurality of DPI-based middleboxes. Our goal in this work is to develop an automated way to craft candidate adversarial packets, targeting TCP implementations in particular. Our approach to achieving this goal hinges on the key insight that while the TCP state machines of DPI implementations are obscure, those of the endhosts are well established. To this end, we develop a system called SYMTCP to systematically explore the TCP implementation of an endhost, identifying candidate packets that can reach critical points in the code (e.g., which causes the packets to be accepted or dropped\/ignored); such automatically identified packets are then fed through the DPI middlebox to determine if a discrepancy is induced and the middlebox can be eluded. The results are promising as it discovered novel evasion strategies that are not previously reported (e.g., involving the urgent pointer).<\/p>\n\n\n\n \u6768\u73c9<\/strong><\/p>\n\n\n\n \u590d\u65e6\u5927\u5b66\u8ba1\u7b97\u673a\u5b66\u9662\u526f\u9662\u957f\uff0c\u6559\u6388<\/strong><\/p>\n\n\n\n \u6f14\u8bb2\u4e3b\u9898\uff1a\u9762\u5411AI\u7cfb\u7edf\u7684\u5b89\u5168\u95ee\u9898\u7814\u7a76<\/strong><\/p>\n\n\n\n \u5bfc\u5e08\u4ecb\u7ecd\uff1a<\/strong> \u6768\u73c9\uff0c\u56fd\u5bb6973\u9879\u76ee\u9996\u5e2d\u79d1\u5b66\u5bb6\uff0c\u6559\u80b2\u90e8\u9752\u5e74\u957f\u6c5f\u7279\u8058\u6559\u6388\uff0c\u590d\u65e6\u5927\u5b66\u8ba1\u7b97\u673a\u5b66\u9662\u6559\u6388\u3001\u535a\u58eb\u751f\u5bfc\u5e08\u3002\u73b0\u4efb\u4e2d\u56fd\u7f51\u7edc\u7a7a\u95f4\u5b89\u5168\u534f\u4f1a\u7406\u4e8b\uff0c\u56fd\u5bb6\u7f51\u7edc\u7a7a\u95f4\u5b89\u5168\u91cd\u70b9\u4e13\u9879\u7ec8\u8bc4\u4e13\u5bb6\u3001\u56fd\u5bb6\u81ea\u7136\u79d1\u5b66\u57fa\u91d1\u7ec8\u8bc4\u4e13\u5bb6\u3001\u6559\u80b2\u90e8\u91cd\u70b9\u5b9e\u9a8c\u5ba4\u8bc4\u4f30\u7ec4\u4e13\u5bb6\u7b49\u5b66\u672f\u517c\u804c\u3002<\/p>\n\n\n\n \u4e3b\u7814\u9886\u57df\u4e3a\u7f51\u7edc\u7a7a\u95f4\u5b89\u5168\uff0c\u5728\u56fd\u5185\u7387\u5148\u5f00\u5c55\u79fb\u52a8\u5b89\u5168\u95ee\u9898\u7814\u7a76\uff0c\u5f62\u6210\u8f83\u5927\u7684\u793e\u4f1a\u548c\u5b66\u672f\u5f71\u54cd\u529b\uff0c\u7814\u7a76\u53d1\u73b0\u66fe\u5f97\u5230\u591a\u4f4d\u56fd\u5bb6\u9886\u5bfc\u4eba\u7684\u591a\u6b21\u6279\u793a\u3002\u66fe\u8363\u83b7\u4e0a\u6d77\u5e02IT\u9752\u5e74\u5341\u5927\u65b0\u9510\u3001\u4e0a\u6d77\u5e02\u9752\u5e74\u4e94\u56db\u5956\u7ae0\u3001\u4e0a\u6d77\u5e02\u5341\u5927\u6770\u51fa\u9752\u5e74\u63d0\u540d\u5956\u7b49\u8363\u8a89\u79f0\u53f7\u3002<\/p>\n\n\n\n \u7a0b\u5149<\/strong><\/p>\n\n\n\n \u4e1c\u5357\u5927\u5b66\u7f51\u7edc\u7a7a\u95f4\u5b89\u5168\u5b66\u9662\u9662\u957f\u3001\u6559\u6388<\/strong><\/p>\n\n\n\n \u6f14\u8bb2\u4e3b\u9898\uff1a\u52a0\u5bc6\u7f51\u7edc\u6d41\u91cf\u6d4b\u91cf\u548c\u5206\u6790<\/strong><\/p>\n\n\n\n \u5bfc\u5e08\u4ecb\u7ecd\uff1a<\/strong> \u7a0b\u5149\uff0c\u4e1c\u5357\u5927\u5b66\u6559\u6388\u3001\u535a\u5bfc\uff0c\u73b0\u4efb\u4e1c\u5357\u5927\u5b66\u7f51\u7edc\u7a7a\u95f4\u5b89\u5168\u5b66\u9662\u6267\u884c\u9662\u957f\uff0c\u8ba1\u7b97\u673a\u7f51\u7edc\u548c\u4fe1\u606f\u96c6\u6210\u6559\u80b2\u90e8\u91cd\u70b9\u5b9e\u9a8c\u5ba4\u4e3b\u4efb\u3001\u56fd\u5bb6\u7f51\u7edc\u7a7a\u95f4\u56fd\u9645\u6cbb\u7406\u7814\u7a76\u57fa\u5730(\u4e1c\u5357\u5927\u5b66)\u6267\u884c\u4e3b\u4efb\u3001\u6c5f\u82cf\u7701\u7f51\u7edc\u7a7a\u95f4\u5b89\u5168\u9ad8\u6821\u8054\u76df\u7406\u4e8b\u957f\u3001\u4e2d\u56fd\u6307\u6325\u63a7\u5236\u5b66\u4f1a\u7f51\u7edc\u7a7a\u95f4\u5b89\u5168\u4e13\u59d4\u4f1a\u526f\u4e3b\u4efb\u3001\u4fe1\u606f\u7cfb\u7edf\u5b89\u5168\u56fd\u9632\u5b9e\u9a8c\u5ba4\u5b66\u672f\u59d4\u5458\u4f1a\u59d4\u5458\u7b49\u793e\u4f1a\u517c\u804c\u3002<\/p>\n\n\n\n 2003\u5e74\u83b7\u4e1c\u5357\u5927\u5b66\u5de5\u5b66\u535a\u58eb\uff0c2006\u5e74-2007\u5e74\u5728\u7f8e\u56fd\u4f50\u6cbb\u4e9a\u7406\u5de5\u5927\u5b66\u4ece\u4e8b\u535a\u58eb\u540e\u7814\u7a76,2012\u5e74\u548c2016\u5e74\u671f\u95f4\u5206\u522b\u5728\u7f8e\u56fd\u4e9a\u5229\u6851\u90a3\u5927\u5b66\u548c\u7f8e\u56fd\u5fb7\u5dde\u5965\u65af\u6c40\u5206\u6821\u4ece\u4e8b\u9ad8\u7ea7\u8bbf\u95ee\u5b66\u8005\u7814\u7a76\u5de5\u4f5c\u30022017\u5e74\u88ab\u4e2d\u592e\u7f51\u4fe1\u529e\u8bc4\u4e3a \u201c\u7f51\u7edc\u5b89\u5168\u4f18\u79c0\u6559\u5e08\u201d\uff0c\u5165\u9009\u6c5f\u82cf\u7701\u201c\u9752\u84dd\u5de5\u7a0b\u201d\u3001\u201c333\u5de5\u7a0b\u201d\u3001\u201c\u516d\u5927\u4eba\u624d\u9ad8\u5cf0\u201d\u3001\u201c\u534e\u82f1\u9752\u5e74\u5b66\u8005\u201d\u3001\u201c\u534e\u82f1\u5b66\u8005\u201d\u3002\u4e3b\u6301\u4e2d\u592e\u7f51\u4fe1\u529e\u3001\u6559\u80b2\u90e8\u7684\u201c\u4e00\u6d41\u7f51\u7edc\u5b89\u5168\u5b66\u9662\u5efa\u8bbe\u793a\u8303\u9879\u76ee\u201d\u3001\u201c\u7f51\u7edc\u7a7a\u95f4\u56fd\u9645\u6cbb\u7406\u7814\u7a76\u57fa\u5730\u201d\u3001\u6559\u80b2\u90e8\u7f51\u7edc\u7a7a\u95f4\u5b89\u5168\u65b0\u5de5\u79d1\u7b49\u9879\u76ee\u30022018\u5e74\u83b7\u5f97\u56fd\u5bb6\u6559\u5b66\u6210\u679c\u4e00\u7b49\u5956\uff0c2014\u5e74\u83b7\u5f97\u6c5f\u82cf\u7701\u79d1\u5b66\u6280\u672f\u5956\u4e8c\u7b49\u5956\u7b49\u3002<\/p>\n\n\n\n \u7814\u7a76\u65b9\u5411\u4e3a\u5728\u7f51\u7edc\u7a7a\u95f4\u5b89\u5168\u3001\u6d41\u91cf\u5927\u6570\u636e\u3001\u672a\u6765\u7f51\u7edc\u7406\u8bba\u4e0e\u65b9\u6cd5\u7b49\u7814\u7a76\u9886\u57df\u627f\u62c5\u56fd\u5bb6\u7701\u90e8\u7ea7\u79d1\u7814\u9879\u76ee30\u9879\u3002\u5728\u56fd\u5185\u5916\u9876\u7ea7\u4f1a\u8bae\u548c\u671f\u520a\u5982IEEE\/ACM Transactions on Networking\u3001\u8ba1\u7b97\u673a\u5b66\u62a5\u3001\u8f6f\u4ef6\u5b66\u62a5\u7b49\u53d1\u8868\u5b66\u672f\u8bba\u6587100\u4f59\u7bc7\uff0c\u53d1\u8868SCI\u7d22\u5f15\u8bba\u6587\u8fd120\u7bc7\uff0c\u51fa\u7248\u4e13\u84576\u90e8\u3002\u83b7\u5f97\u56fd\u5bb6\u3001\u7701\u90e8\u7ea7\u548c\u4f01\u4e1a\u79d1\u7814\u9879\u76ee\u8d44\u52a930\u4f59\u9879\uff0c\u53d6\u5f97\u4e86\u521b\u65b0\u6027\u7814\u7a76\u6210\u679c\uff1b\u83b7\u5f97\u6388\u6743\u53d1\u660e\u4e13\u522922\u9879\uff0c\u5df2\u7ecf\u6210\u529f\u8f6c\u53165\u9879\u4e13\u5229\u3002\u51fa\u7248\u4e13\u84577\u90e8\uff0c\u57f9\u517b\u7814\u7a76\u751f100\u4f59\u4eba\u3002<\/p>\n\n\n\n \u6b66\u6210\u5c97<\/strong><\/p>\n\n\n\n \u4e2d\u79d1\u9662\u8ba1\u7b97\u6240\u6559\u6388\u7ea7\u9ad8\u5de5<\/strong><\/p>\n\n\n\n \u6f14\u8bb2\u4e3b\u9898\uff1a\u4e00\u79cd\u57fa\u4e8eSMAP\u7684\u5b89\u5168\u9ad8\u6548\u7684\u8fdb\u7a0b\u5185\u9694\u79bb\u65b9\u6cd5<\/strong><\/p>\n\n\n\n \u5bfc\u5e08\u4ecb\u7ecd\uff1a<\/strong><\/p>\n\n\n\n \u6b66\u6210\u5c97\uff0c\u4e2d\u56fd\u79d1\u5b66\u9662\u8ba1\u7b97\u6280\u672f\u7814\u7a76\u6240\u7814\u7a76\u5458\u7ea7\u9ad8\u5de5\uff0c\u535a\u58eb\u751f\u5bfc\u5e08\u30022001\u5e74\u4e8e\u4e2d\u56fd\u79d1\u5b66\u9662\u8ba1\u7b97\u6280\u672f\u7814\u7a76\u6240\u83b7\u535a\u58eb\u5b66\u4f4d\u3002\u957f\u671f\u4ece\u4e8b\u57fa\u4e8e\u7f16\u8bd1\u6280\u672f\u7684\u8f6f\u4ef6\u5b89\u5168\u6280\u672f\u7684\u7814\u7a76\uff0c\u65e8\u5728\u901a\u8fc7\u7a0b\u5e8f\u5206\u6790\u6280\u672f\uff0c\u589e\u5f3a\u8f6f\u4ef6\u7684\u5b89\u5168\u6027\u3001\u53ef\u9760\u6027\u548c\u9ad8\u6548\u6027\u3002\u5728\u7a0b\u5e8f\u7f3a\u9677\u5b9a\u4f4d\u3001\u4e8c\u8fdb\u5236\u4ee3\u7801\u7684\u5206\u6790\u53d8\u6362\u53ca\u4f18\u5316\u3001\u8fd0\u884c\u65f6\u7a0b\u5e8f\u884c\u4e3a\u76d1\u6d4b\u3001\u5b89\u5168\u9632\u5fa1\u7b49\u65b9\u9762\uff0c\u6709\u7740\u8f83\u6df1\u7684\u6280\u672f\u79ef\u7d2f\u3002\u53d1\u8868\u8bba\u658730\u4f59\u7bc7\uff0c\u90e8\u5206\u53d1\u8868\u5728TSE\u3001ICSE\u3001ASE\u3001TPDS\u3001TACO\u3001SIGMETRICS\u3001PACT\u3001CGO\u3001VEE\u3001DATE\u7b49\u5b66\u672f\u4f1a\u8bae\u548c\u671f\u520a\u4e0a\uff0c\u83b7\u5f97\u6388\u6743\u4e13\u522916\u9879\uff0c\u8f6f\u4ef6\u8457\u4f5c\u67435\u9879\uff0c\u5e76\u4e8e2012\u5e74\u8363\u83b7\u5317\u4eac\u5e02\u79d1\u5b66\u6280\u672f\u4e8c\u7b49\u5956\u3002\u62c5\u4efb\u56fd\u9645\u5b66\u672f\u4f1a\u8baeCGO 2013\u5927\u4f1a\u4e3b\u5e2d\u3001APPT 2013\u7a0b\u5e8f\u59d4\u5458\u4f1a\u4e3b\u5e2d\uff0c\u8fd8\u62c5\u4efbCGO2015-2018\u3001PPoPP2017-2018\u3001PLDI2012\u3001CCGrid 2015\u3001ICPADS 2014\u3001PPPJ2014\u7b49\u4f1a\u8bae\u7684\u7a0b\u5e8f\u59d4\u5458\u4f1a\u59d4\u5458\u3002<\/p>\n\n\n\n \u5f20\u9633<\/strong><\/p>\n\n\n\n \u5fb7\u56fdCISPA\u7814\u7a76\u4e2d\u5fc3\u7814\u7a76\u5458\u3001\u535a\u58eb<\/strong><\/p>\n\n\n\n \u6f14\u8bb2\u4e3b\u9898\uff1aPrivacy in the Era of Machine Learning<\/strong><\/p>\n\n\n\n \u5bfc\u5e08\u4ecb\u7ecd\uff1a<\/strong> Yang Zhang is a faculty member at CISPA Helmholtz Center for Information Security, Germany. Previously, he was a group leader at CISPA. He obtained his Ph.D. degree from University of Luxembourg in November 2016 under the supervision of Prof. Sjouke Mauw and Dr. Jun Pang. Yang’s research interests lie at the intersection of privacy and machine learning. Over the years, he has published multiple papers at top venues in computer science, including WWW, CCS, NDSS, USENIX Security, and IJCAI. His work has received NDSS 2019 distinguished paper award. Yang has served in the technical program committee of ACM CCS 2020 2019, ISMB 2019, WWW 2020, RAID 2020, ICWSM 2020, and PETS 2021 2020.<\/p>\n\n\n\n \u5185\u5bb9\u6458\u8981\uff1a<\/strong><\/p>\n\n\n\n The past decade has witnessed the fast development of machine learning techniques, and the key factor that drives the current progress is the unprecedented large-scale data. On the one hand, machine learning and big data can help improve various domains of people’s life quality. On the other hand, they can also cause severe risks to people’s privacy. In this talk, I will present our research on the intersection of data privacy and machine learning. First, I will show how to use machine learning techniques to assess and mitigate privacy risks for various types of data, including social network data, location data, and biomedical data. Then, I will present our research on quantifying privacy risks caused by machine learning models. In particular, I will discuss our newest results on membership inference and data reconstruction.<\/p>\n\n\n\n \u6bb5\u6d77\u65b0<\/strong><\/p>\n\n\n\n \u6e05\u534e\u5927\u5b66\u7f51\u7814\u9662\u6559\u6388<\/strong><\/p>\n\n\n\n \u6f14\u8bb2\u4e3b\u9898\uff1a\u4e92\u8054\u7f51\u57fa\u7840\u8bbe\u65bd\u5b89\u5168\u7684\u4f9d\u8d56\u6027\u5206\u6790<\/strong><\/p>\n\n\n\n \u5bfc\u5e08\u4ecb\u7ecd\uff1a<\/strong> \u6bb5\u6d77\u65b0\uff0c\u6e05\u534e\u5927\u5b66\u7f51\u7edc\u79d1\u5b66\u4e0e\u7f51\u7edc\u7a7a\u95f4\u7814\u7a76\u9662\u6559\u6388\uff0c\u7f51\u7edc\u7a7a\u95f4\u5b89\u5168\u5b9e\u9a8c\u5ba4\u4e3b\u4efb\uff0c\u535a\u58eb\u751f\u5bfc\u5e08\u3002\u8fd120\u5e74\u6765\u4e00\u76f4\u5de5\u4f5c\u5728\u7f51\u7edc\u5b89\u5168\u79d1\u7814\u548c\u6559\u5b66\u7b2c\u4e00\u7ebf\u3002\u4f5c\u4e3a\u8bfe\u9898\u8d1f\u8d23\u4eba\u627f\u62c5\u56fd\u5bb6\u81ea\u7136\u79d1\u5b66\u57fa\u91d1\u3001973\u3001863\u7b49\u591a\u9879\u8bfe\u9898\uff0c\u5728\u7f51\u7edc\u548c\u7cfb\u7edf\u5b89\u5168\u56fd\u9645\u9876\u7ea7\u5b66\u672f\u4f1a\u8bae\uff08Security&Privacy\u3001USENIX Security\u3001CCS\u3001NDSS\u7b49\uff09\u4e0a\u53d1\u8868\u591a\u7bc7\u8bba\u6587\uff0c\u5e76\u83b7NDSS 2016\u5e74\u201c\u6770\u51fa\u8bba\u6587\u5956\u201d\u3002\u517c\u4efb\u4e2d\u56fd\u7f51\u7edc\u7a7a\u95f4\u5b89\u5168\u534f\u4f1a\u7406\u4e8b\u3001\u5bc6\u7801\u5b66\u4f1a\u534f\u8bae\u5b89\u5168\u4e13\u59d4\u4f1a\u59d4\u5458\u3001\u4e92\u8054\u7f51\u534f\u4f1a\u5b89\u5168\u5de5\u4f5c\u59d4\u5458\u4f1a\u59d4\u5458\u7b49\u804c\u52a1\uff0c\u662f\u591a\u4e2a\u56fd\u9645\u4f1a\u8bae\u7a0b\u5e8f\u59d4\u5458\u548c\u56fd\u9645\u671f\u520a\u5ba1\u7a3f\u4eba\u3002<\/p>\n\n\n\n \u6bb5\u6d77\u65b0\u5e26\u9886\u56e2\u961f\u53d1\u73b0\u4e86\u4e92\u8054\u7f51\u534f\u8bae\u4e2d\u57df\u540d\u7cfb\u7edf\u3001\u5185\u5bb9\u5206\u53d1\u7f51\u7edc\u548cWeb\u534f\u8bae\u7b49\u82e5\u5e72\u91cd\u5927\u5b89\u5168\u6f0f\u6d1e\u5e76\u63d0\u51fa\u4e86\u89e3\u51b3\u65b9\u6848\uff0c\u4fc3\u8fdb\u4e86\u4ea7\u54c1\u548c\u6807\u51c6\u7684\u66f4\u65b0\uff0c\u4ea7\u751f\u4e86\u91cd\u8981\u7684\u56fd\u9645\u5f71\u54cd\uff1b\u4ed6\u957f\u671f\u4ece\u4e8b\u4e92\u8054\u7f51\u5b89\u5168\u68c0\u6d4b\u4e0e\u5e94\u6025\u54cd\u5e94\uff0c\u68c0\u6d4b\u9ed1\u8272\u4ea7\u4e1a\u3001\u6253\u51fb\u7f51\u7edc\u72af\u7f6a\uff0c\u4e3a\u56fd\u5bb6\u7f51\u7edc\u5b89\u5168\u505a\u51fa\u4e86\u91cd\u8981\u8d21\u732e\uff1b\u4ed6\u4ece\u4e8b\u7f51\u7edc\u5b89\u5168\u6559\u5b66\u5341\u591a\u5e74\uff0c\u6ce8\u91cd\u57f9\u517b\u5b66\u751f\u7684\u5b9e\u8df5\u548c\u521b\u65b0\u80fd\u529b\uff0c\u4f5c\u4e3a\u8054\u5408\u521b\u59cb\u4eba\u7684\u201c\u84dd\u83b2\u82b1\u6218\u961f\u201d\u591a\u6b21\u5728\u56fd\u9645\u8457\u540d\u7684\u7f51\u7edc\u5b89\u5168\u7ade\u8d5b\u4e2d\u540d\u5217\u524d\u8305\uff0c\u4f5c\u4e3a\u8054\u5408\u53d1\u8d77\u4eba\u7ec4\u7ec7\u7f51\u7edc\u5b89\u5168\u7814\u7a76\u56fd\u9645\u5b66\u672f\u8bba\u575bInForSec\u548c\u5168\u56fd\u7f51\u7edc\u5b89\u5168\u5bf9\u6297\u8054\u8d5bXCTF\uff0c\u5bf9\u6211\u56fd\u7f51\u7edc\u5b89\u5168\u4eba\u624d\u57f9\u517b\u53d1\u6325\u4e86\u91cd\u8981\u4f5c\u7528\u30022016\u5e74\u83b7\u5f97\u4e2d\u592e\u7f51\u4fe1\u529e\u9881\u53d1\u7684\u201c\u7f51\u7edc\u5b89\u5168\u4f18\u79c0\u4eba\u624d\u201d\u5956\u3002<\/p>\n\n\n\n \u90a2\u65b0\u5b87<\/strong><\/p>\n\n\n\n \u5bbe\u5915\u6cd5\u5c3c\u4e9a\u5dde\u7acb\u5927\u5b66\u52a9\u7406\u6559\u6388<\/strong><\/p>\n\n\n\n \u6f14\u8bb2\u4e3b\u9898\uff1aPostmortem Program Analysis<\/strong><\/p>\n\n\n\n \u5bfc\u5e08\u4ecb\u7ecd\uff1a<\/strong><\/p>\n\n\n\n Dr. Xinyu Xing is an Assistant Professor at Pennsylvania State University. His research interests include exploring, designing and developing new techniques to automate vulnerability discovery, failure reproduction, vulnerability diagnosis (and triage), exploit and security patch generation. His past research has been featured by many mainstream media and received the best paper awards from ACM CCS and ACSAC. Going beyond academic research, he also actively participates and hosts many world-class cybersecurity competitions (such as HITB and XCTF) and frequently shared thoughts at prestigious industrial conferences (such as Black Hat and DEFCON). \u5185\u5bb9\u6458\u8981\uff1a<\/strong><\/p>\n\n\n\n Despite the best efforts of developers, software inevitably contains flaws that may be leveraged as security vulnerabilities. Modern operating systems integrate various security mechanisms to prevent software faults from being exploited. To bypass these defenses and hijack program execution, an attacker needs to constantly mutate an exploit and make many attempts. While in their attempts, the exploit triggers a security vulnerability and makes the running process abnormally terminate.<\/p>\n\n\n\n After a program has crashed and abnormally terminated, it typically leaves behind a snapshot of its crashing state in the form of a core dump. While a core dump carries a large amount of information, which has long been used for software debugging, it barely serves as informative debugging aids in locating software faults, particularly memory corruption vulnerabilities. As such, previous research mainly seeks fully reproducible execution tracing to identify software vulnerabilities in crashes. However, such techniques are usually impractical for complex programs. Even for simple programs, the overhead of fully reproducible tracing may only be acceptable at the time of in-house testing.<\/p>\n\n\n\n In this talk, I will discuss how we tackle this issue by bridging program analysis with artificial intelligence (AI). More specifically, I will first talk about the history of postmortem program analysis, characterizing and disclosing their limitations. Second, I will introduce how we design a new reverse-execution approach for postmortem program analysis. Third, I will discuss how we integrate AI into our reverse-execution method to escalate its analysis efficiency and accuracy. Last but not least, as part of this talk, I will demonstrate the effectiveness of this AI-assisted postmortem program analysis framework by using massive amounts of real-world programs.<\/p>\n\n\n\n \u5468\u6587\u67cf<\/strong><\/p>\n\n\n\n \u4e2d\u56fd\u79d1\u5b66\u6280\u672f\u5927\u5b66\u535a\u58eb\u540e<\/strong><\/p>\n\n\n\n \u6f14\u8bb2\u4e3b\u9898\uff1aDeepfake\u53ca\u5176\u68c0\u6d4b\u6280\u672f<\/strong><\/p>\n\n\n\n \u5609\u5bbe\u4ecb\u7ecd\uff1a <\/strong> \u5468\u6587\u67cf\uff0c\u4e2d\u56fd\u79d1\u5b66\u6280\u672f\u5927\u5b66\u7f51\u7edc\u7a7a\u95f4\u5b89\u5168\u5b66\u9662\uff0c\u535a\u58eb\u540e\u7814\u7a76\u5458\uff0c\u56fe\u8c61\u56fe\u5f62\u5b66\u5b66\u4f1a\u591a\u5a92\u4f53\u53d6\u8bc1\u4e0e\u5b89\u5168\u4e13\u5bb6\u59d4\u5458\u4f1a\u59d4\u5458\uff0c\u4e2d\u56fd\u79d1\u5b66\u6280\u672f\u5927\u5b66\u201c\u58a8\u5b50\u6770\u51fa\u9752\u5e74\u201d\u83b7\u5f97\u8005\u30022019\u5e74\u535a\u58eb\u6bd5\u4e1a\u4e8e\u4e2d\u56fd\u79d1\u5b66\u6280\u672f\u5927\u5b66\uff0c\u4e3b\u8981\u7814\u7a76\u5174\u8da3\u5305\u62ec\u4fe1\u606f\u9690\u85cf\u4e0e\u4eba\u5de5\u667a\u80fd\u5b89\u5168\u3002\u5728TIFS\u3001TCSVT\u3001ICCV\u3001AAAI\u7b49\u56fd\u9645\u77e5\u540d\u671f\u520a\u548c\u4f1a\u8bae\u4e0a\u53d1\u8868\u9ad8\u6c34\u5e73\u5b66\u672f\u8bba\u6587\u5341\u4f59\u7bc7\uff0c\u4e2d\u56fd\u79d1\u6280\u5927\u5b66\u7f51IPC\u4eba\u5de5\u667a\u80fd\u5b89\u5168\u6218\u961f\u961f\u957f\uff0c\u66fe\u83b7Facebook Deepfake Detection Challenge\uff08DFDC\uff09\u5168\u7403\u7b2c\u4e8c\uff0cIJCAI-19-\u963f\u91cc\u5df4\u5df4\u4eba\u5de5\u667a\u80fd\u5bf9\u6297\u7b97\u6cd5\u7ade\u8d5b\u7b2c\u4e00\uff0c2018 CAAD\u5bf9\u6297\u6837\u672c\u653b\u9632\u8d5b\u516d\u5f3a\u3002\u4e3b\u6301\u5b89\u5fbd\u7701\u81ea\u7136\u79d1\u5b66\u57fa\u91d11\u9879\u3002<\/p>\n\n\n\n \u5185\u5bb9\u6458\u8981\uff1a<\/strong><\/p>\n\n\n\n Deepfake\u201c\u6df1\u5ea6\u6362\u8138\u201d\u662f\u8fd1\u671f\u65b0\u5174\u7684\u70ed\u70b9\u95ee\u9898\uff0c\u901a\u8fc7\u4eba\u5de5\u667a\u80fd\u6280\u672f\uff0c\u5b9e\u73b0\u4eba\u8138\u89c6\u9891\u7684\u6a21\u62df\u4e0e\u4f2a\u9020\u3002\u4e0e\u4f20\u7edf\u7684\u57fa\u4e8e\u8ba1\u7b97\u673a\u56fe\u5f62\u5b66\u7684\u9ad8\u95e8\u69db\u6362\u8138\u6280\u672f\u4e0d\u540c\uff0cDeepfake\u5f97\u76ca\u4e8eAI\u6280\u672f\u7684\u98de\u901f\u53d1\u5c55\uff0c\u5927\u5927\u964d\u4f4e\u4e86\u6362\u8138\u624b\u6bb5\u7684\u6280\u672f\u95e8\u69db\u3002\u8fd9\u7c7b\u6280\u672f\u76ee\u524d\u5df2\u88ab\u5e7f\u6cdb\u7528\u4e8e\u5a31\u4e50\u5e94\u7528\u4e2d\uff0c\u5e76\u5f00\u53d1\u4e86\u5982ZAO\u3001Impressions\u7b49\u6362\u8138\u8f6f\u4ef6\u3002\u800cDeepfake\u672c\u8eab\u4ea6\u662f\u4e00\u628a\u201c\u53cc\u5203\u5251\u201d\uff0c\u8fd9\u7c7b\u6280\u672f\u7684\u76db\u884c\u5e26\u6765\u4e86\u6781\u5927\u7684\u6f5c\u5728\u98ce\u9669\uff0c\u5c24\u5176\u662f\u5728\u653f\u6cbb\u3001\u91d1\u878d\u3001\u516c\u5171\u5b89\u5168\u7b49\u9886\u57df\uff0c\u4e00\u65e6\u51fa\u73b0\u6076\u610f\u4f2a\u9020\uff0c\u5982\u8eab\u4efd\u4f2a\u9020\u3001\u8206\u60c5\u717d\u52a8\u7b49\uff0c\u90fd\u5c06\u5bf9\u56fd\u5bb6\u548c\u793e\u4f1a\u5e26\u6765\u5de8\u5927\u5371\u5bb3\uff0c\u56e0\u6b64\u5404\u56fd\u653f\u5e9c\u4e0e\u8bf8\u591a\u56fd\u9645\u673a\u6784\u9ad8\u5ea6\u5173\u6ce8Deepfake\u7684\u9632\u6cbb\u3002<\/p>\n\n\n\n Deepfake\u9700\u8981\u7ecf\u5386\u5a92\u4f53\u6570\u636e\u83b7\u53d6\u3001\u4eba\u8138\u5206\u5272\u3001\u9762\u90e8\u4f2a\u9020\u3001\u66ff\u6362\u878d\u5408\u3001\u4f2a\u9020\u75d5\u8ff9\u4fee\u590d\u7b49\u4e3b\u8981\u8fc7\u7a0b\uff0c\u6700\u7ec8\u5b9e\u73b0\u6df1\u5ea6\u6362\u8138\uff0c\u8fd9\u4e5f\u610f\u5473\u7740Deepfake\u7684\u68c0\u6d4b\u5c06\u662f\u590d\u6742\u800c\u5168\u65b9\u4f4d\u7684\u5de5\u4f5c\u3002\u672c\u62a5\u544a\u5c06\u4ece\u73b0\u6709\u6280\u672f\u4e0e\u5b9e\u9645\u5e94\u7528\u573a\u666f\u51fa\u53d1\uff0c\u56f4\u7ed5\u56e2\u961f\u5728\u751f\u6210\u65b9\u9762\u7684\u76f8\u5173\u5de5\u4f5c\u548cDFDC\u6bd4\u8d5b\u4e2d\u7684\u89e3\u51b3\u65b9\u6848\uff0c\u7ed9\u51fa\u5bf9\u4e8eDeepfake\u53ca\u5176\u68c0\u6d4b\u6280\u672f\u7684\u4e00\u4e9b\u63a2\u7a76\u4e0e\u601d\u8003\u3002<\/p>\n\n\n\n \u90d1\u6653\u5cf0<\/strong><\/p>\n\n\n\n \u6e05\u534e-\u5947\u5b89\u4fe1\u8054\u5408\u7814\u7a76\u4e2d\u5fc3\u7814\u7a76\u5458<\/strong><\/p>\n\n\n\n \u6f14\u8bb2\u4e3b\u9898\uff1a\u5de5\u4e1a\u754c\u6570\u636e\u89c6\u89d2\u4e0b\u7684\u5b89\u5168\u7814\u7a76\uff08\u53caDataCon\u7ade\u8d5b\u4ecb\u7ecd\uff09<\/strong><\/p>\n\n\n\n \u5609\u5bbe\u4ecb\u7ecd\uff1a<\/strong> \u90d1\u6653\u5cf0 \u5947\u5b89\u4fe1\u96c6\u56e2\u6280\u672f\u7814\u7a76\u9662\u7814\u7a76\u5458\uff0c\u6e05\u534e\u5927\u5b66-\u5947\u5b89\u4fe1\u8054\u5408\u7814\u7a76\u4e2d\u5fc3\u7814\u7a76\u5458\uff0c\u7fb2\u548c\u7f51\u7edc\u5b89\u5168\u5b9e\u9a8c\u5ba4\u4e3b\u4efb\u3002\u4ece\u4e8b\u7f51\u7edc\u57fa\u7840\u8bbe\u65bd\u3001\u57fa\u7840\u534f\u8bae\u5b89\u5168\u76f8\u5173\u7814\u7a76\uff0c\u7814\u7a76\u6210\u679c\u53d1\u8868\u5728\u7f51\u7edc\u5b89\u5168\u56fd\u9645\u9876\u7ea7\u5b66\u672f\u4f1a\u8baeUSENIX Security\u3001NDSS\uff0c\u5e76\u83b7DNSS 2016\u5e74\u201c\u6770\u51fa\u8bba\u6587\u5956\u201d\u3002\u66fe\u83b7Google\u6d4f\u89c8\u5668\u5b89\u5168\u8ba1\u5212\u5956\u52b1\u3001GeekPwn\uff082015\uff0946\u4e07\u5143\u5956\u52b1\u3001GeekPwn\uff082018\uff0925\u4e07\u5143\u5956\u52b1\u3001GeekPwn\uff082019\uff0915\u4e07\u5143\u5956\u52b1\u3001TSRC\u901a\u7528\u8f6f\u4ef6\u6f0f\u6d1e10\u4e07\u5143\u5956\u52b1\u7b49\u8363\u8a89\u3002\u7814\u7a76\u6210\u679c\u5e2e\u52a9\u8bb8\u591a\u4e16\u754c\u77e5\u540d\u7684\u4e92\u8054\u7f51\u4f01\u4e1a\u4fee\u590d\u4e86\u5b89\u5168\u6f0f\u6d1e\uff0c\u5982Bank of America\u3001Google\u3001Facebook\u3001Mozilla\u3001Apple\u3001Amazon\u3001\u4e2d\u56fd\u5efa\u884c\u3001\u4e2d\u56fd\u94f6\u8054\u3001\u652f\u4ed8\u5b9d\u7b49\uff0c\u63a8\u52a8\u4e86Google\u516c\u53f8\u4fee\u6b63Chrome\uff0c\u4fc3\u8fdb\u4e86\u56fd\u9645\u4e92\u8054\u7f51\u6807\u51c6\u7ec4\u7ec7IETF\u4fee\u6539\u76f8\u5173\u56fd\u9645\u6807\u51c6\u3002<\/p>\n\n\n\n \u5e94\u51cc\u4e91<\/strong><\/p>\n\n\n\n \u6e05\u534e-\u5947\u5b89\u4fe1\u8054\u5408\u7814\u7a76\u4e2d\u5fc3\u7814\u7a76\u5458<\/strong><\/p>\n\n\n\n \u6f14\u8bb2\u4e3b\u9898\uff1a\u6c99\u7bb1\u7cfb\u7edf\u7684\u6076\u610f\u4ee3\u7801\u5206\u6790\u7814\u7a76\u53ca\u5b9e\u8df5<\/strong><\/p>\n\n\n\n \u5609\u5bbe\u4ecb\u7ecd\uff1a<\/strong> \u5e94\u51cc\u4e91\uff0c\u5947\u5b89\u4fe1\u96c6\u56e2\u6280\u672f\u7814\u7a76\u9662\u7814\u7a76\u5458\uff0c\u6e05\u534e\u5927\u5b66-\u5947\u5b89\u4fe1\u8054\u5408\u7814\u7a76\u4e2d\u5fc3\u7814\u7a76\u5458\uff0c\u4e2d\u56fd\u79d1\u5b66\u9662\u5927\u5b66\u5c97\u4f4d\u6559\u6388\uff0c\u7855\u58eb\u751f\u5bfc\u5e08\u3002\u957f\u671f\u4ece\u4e8b\u7f51\u7edc\u5b89\u5168\u76f8\u5173\u6280\u672f\u7814\u7a76\u5de5\u4f5c\uff0c\u5df2\u53d1\u8868\u5b66\u672f\u8bba\u658729\u7bc7\uff0c\u51fa\u7248\u5408\u84571\u90e8\uff0c\u4e3b\u6301\u56fd\u5bb6\u81ea\u7136\u79d1\u5b66\u57fa\u91d1\u3001\u56fd\u5bb6\u53d1\u6539\u59d4\u4fe1\u606f\u5b89\u5168\u4e13\u9879\u3001\u5317\u4eac\u5e02\u81ea\u7136\u79d1\u5b66\u57fa\u91d1\u7b49\u9879\u76ee15\u9879\uff0c\u7533\u8bf7\u6280\u672f\u53d1\u660e\u4e13\u522915\u9879\uff0c\u53c2\u4e0e\u56fd\u9645\u6807\u51c6\u63d0\u68482\u9879\u3001\u56fd\u5bb6\u884c\u4e1a\u6807\u51c6\u7f16\u52364\u9879\uff0c\u5148\u540e\u8363\u83b7\u4e2d\u56fd\u901a\u4fe1\u5b66\u4f1a\u79d1\u5b66\u6280\u672f\u5956\uff08\u79d1\u6280\u8fdb\u6b65\u7c7b\uff09\u4e00\u7b49\u5956\uff082018\uff09\u3001\u5317\u4eac\u5e02\u79d1\u5b66\u6280\u672f\u5956\u4e8c\u7b49\u5956\uff082017\uff09\u3001\u4e2d\u56fd\u7535\u529b\u6280\u672f\u53d1\u660e\u5956\u4e09\u7b49\u5956\uff082017\uff09\u7b49\u5956\u9879\u8363\u8a89\u3002<\/p>\n\n\n\n \u5218\u4fdd\u541b<\/strong><\/p>\n\n\n\n \u6e05\u534e\u5927\u5b66\u535a\u58eb\u540e<\/strong><\/p>\n\n\n\n \u6f14\u8bb2\u4e3b\u9898\uff1a\u7ba1\u4e2d\u7aa5\u8c79\uff0c\u4eceDNS\u8c08\u6570\u636e\u5b89\u5168\u5206\u6790<\/strong><\/p>\n\n\n\n \u5609\u5bbe\u4ecb\u7ecd\uff1a<\/strong> \u5218\u4fdd\u541b\u662f\u6e05\u534e\u5927\u5b66\u7f51\u7edc\u7a7a\u95f4\u4e0e\u7f51\u7edc\u79d1\u5b66\u7814\u7a76\u9662\u535a\u58eb\u540e\uff0c2018\u5e74\u4e8e\u4f2f\u514b\u5229\u56fd\u9645\u8ba1\u7b97\u673a\u79d1\u5b66\u7814\u7a76\u6240\u8bbf\u95ee\u3002\u4e3b\u8981\u7814\u7a76\u65b9\u5411\u4e3a\u7f51\u7edc\u57fa\u7840\u8bbe\u65bd\u5b89\u5168\u3002\u535a\u58eb\u671f\u95f4\u5b66\u672f\u8bba\u6587\u53d1\u8868\u4e8eIEEE S&P\u3001USENIX Security\u3001CCS\u3001NDSS\u3001IMC\u7b49\u56fd\u9645\u7f51\u7edc\u5b89\u5168\u9886\u57df\u91cd\u8981\u4f1a\u8bae\uff0c\u83b7\u5f97\u6e05\u534e\u5927\u5b66\u535a\u58eb\u751f\u56fd\u5bb6\u5956\u5b66\u91d1\u3001NDSS 2019\u6700\u4f73\u8bba\u6587\u5956\uff0cIMC 2019\u6700\u4f73\u8bba\u6587\u5956\u4e0e\u5b89\u5168\u793e\u533a\u8d21\u732e\u5956\u63d0\u540d\u3002<\/p>\n","protected":false},"excerpt":{"rendered":" 2020\u5e74InForSec\u201c\u8d70\u8fdb\u8682\u8681\u96c6\u56e2\u201d\u7f51\u7edc\u7a7a\u95f4\u5b89\u5168\u5927\u5b66\u751f\u590f\u4ee4\u8425\u6d3b\u52a8\u8ba1\u5212\u4e8e7\u670818\u65e5\u81f319\u65e5\u5728\u676d\u5dde\u4e3e\u884c\uff1b\u6d3b\u52a8\u9762\u5411\u5168\u56fd\u8ba1\u7b97\u673a\u3001\u7f51\u7edc\u7a7a\u95f4\u5b89\u5168\u76f8\u5173\u4e13\u4e1a\u7684\u672c\u79d1\u751f\u548c\u5728\u8bfb\u7814\u7a76\u751f\uff0c\u65e8\u5728\u4fc3\u8fdb\u7f51\u7edc\u5b89\u5168\u76f8\u5173\u65b9\u5411\u8001\u5e08\u3001\u5b66\u751f\u4e4b\u95f4\u7684\u4e86\u89e3\u548c\u6c9f\u901a\uff0c\u5e2e\u52a9\u5b66\u751f\u4e86\u89e3\u7f51\u7edc\u5b89\u5168\u76f8\u5173\u7684\u7814\u7a76\u65b9\u5411\u3002 Continue reading
<\/p>\n\n\n\n![\"\"](\"https:\/\/www.inforsec.org\/wp\/wp-content\/uploads\/2020\/06\/\u5fae\u4fe1\u56fe\u7247_20200629222955.jpg\")
<\/figure><\/div>\n\n\n\n
<\/p>\n\n\n\n![\"\"](\"https:\/\/www.inforsec.org\/wp\/wp-content\/uploads\/2020\/06\/15934398891.png\")
<\/figure><\/div>\n\n\n\n![\"\"](\"https:\/\/www.inforsec.org\/wp\/wp-content\/uploads\/2020\/06\/15934399111.png\")
<\/figure><\/div>\n\n\n\n![\"\"](\"https:\/\/www.inforsec.org\/wp\/wp-content\/uploads\/2020\/06\/15934399341.png\")
<\/figure><\/div>\n\n\n\n![\"\"](\"https:\/\/www.inforsec.org\/wp\/wp-content\/uploads\/2020\/06\/15934399551.png\")
<\/figure><\/div>\n\n\n\n![\"\"](\"https:\/\/www.inforsec.org\/wp\/wp-content\/uploads\/2020\/06\/15934400821.png\")
<\/figure><\/div>\n\n\n\n
<\/p>\n\n\n\n![\"\"](\"https:\/\/www.inforsec.org\/wp\/wp-content\/uploads\/2020\/06\/15934401031.png\")
<\/figure><\/div>\n\n\n\n![\"\"](\"https:\/\/www.inforsec.org\/wp\/wp-content\/uploads\/2020\/06\/15934401311.png\")
<\/figure><\/div>\n\n\n\n
<\/p>\n\n\n\n![\"\"](\"https:\/\/www.inforsec.org\/wp\/wp-content\/uploads\/2020\/06\/15934401511.png\")
<\/figure><\/div>\n\n\n\n
<\/p>\n\n\n\n![\"\"](\"https:\/\/www.inforsec.org\/wp\/wp-content\/uploads\/2020\/06\/15934401751.png\")
<\/figure><\/div>\n\n\n\n![\"\"](\"https:\/\/www.inforsec.org\/wp\/wp-content\/uploads\/2020\/06\/15934401951.png\")
<\/figure><\/div>\n\n\n\n
<\/p>\n\n\n\n![\"\"](\"https:\/\/www.inforsec.org\/wp\/wp-content\/uploads\/2020\/06\/15934402181.png\")
<\/figure><\/div>\n\n\n\n
<\/p>\n\n\n\n![\"\"](\"https:\/\/www.inforsec.org\/wp\/wp-content\/uploads\/2020\/06\/15934402411.png\")
<\/figure><\/div>\n\n\n\n
<\/p>\n\n\n\n![\"\"](\"https:\/\/www.inforsec.org\/wp\/wp-content\/uploads\/2020\/06\/15934402641.png\")
<\/figure><\/div>\n\n\n\n
<\/p>\n\n\n\n![\"\"](\"https:\/\/www.inforsec.org\/wp\/wp-content\/uploads\/2020\/06\/15934402821.png\")
<\/figure><\/div>\n\n\n\n
<\/p>\n\n\n\n![\"\"](\"https:\/\/www.inforsec.org\/wp\/wp-content\/uploads\/2020\/06\/15934403071.png\")
<\/figure><\/div>\n\n\n\n
<\/p>\n\n\n\n![\"\"](\"https:\/\/www.inforsec.org\/wp\/wp-content\/uploads\/2020\/06\/15934403271.png\")
<\/figure><\/div>\n\n\n\n
<\/p>\n\n\n\n