#programanalysis
September 19, 2026 at 11:59 PM
Datalog para LLMs: memoria que no olvida conclusiones

¿Tu agente de IA pierde el hilo después de 2 horas de investigación? Un motor Datalog mantiene la consistencia de las conclusiones mientras el LLM hace ...

#datalog #llm #lemmalog #memoria #programanalysis
Datalog para LLMs: memoria que no olvida conclusiones
Lemmalog usa Datalog para mantener la consistencia de las conclusiones de un agente de IA. Si un hecho cambia, solo se invalidan las conclusiones que dependían de él. Probado en LongMemEval con res...
blog.donweb.com
August 29, 2026 at 2:24 AM
AI ranked 3,762 JNI-reachable Zoom functions; the vulnerable annotation library was 45th. The useful move was mapping remote meeting bytes to loaded parsers before ranking sinks. Reachability before danger: https://bokonon.ai/notes/2026-08-13-rank-reachability-before-danger/ #AppSec #ProgramAnalysis
August 13, 2026 at 8:01 PM
Next came Karl et al.'s "Uncovering Bigger Truths: Deobfuscating PHP with Phoebe" on reversing PHP obfuscation with 80% similarity and no syntax errors. (www.acsac.org/2025/p...) 3/6
#CodeObfuscation #ProgramAnalysis #ReverseEngineering
June 25, 2026 at 2:00 PM
[Transactions on Artificial Intelligence] A new TAI survey reviews LLMs for code comprehension and security detection across static, dynamic, and hybrid methods.

www.sciltp.com/journals/tai...

#LargeLanguageModels #Cybersecurity #ProgramAnalysis #ArtificialIntelligence #SoftwareEngineering
June 22, 2026 at 1:26 AM
#CallforReadding This survey reviews the application of #LLMs in program analysis, categorizing existing work into static, dynamic, and hybrid approaches. Qingchuan Zhao of City University of Hong Kong co-authored this paper.

doi.org/10.53941/tai...

#programanalysis #vulnerabilitydetection
April 29, 2026 at 2:33 AM
GPU‑accelerated Loopy Belief Propagation runs data‑race detection 2.14× faster than sequential LBP and 5.56× over prior GPU methods, per a Sep 2025 study on eight Java programs. https://getnews.me/gpu-accelerated-loopy-belief-propagation-boosts-program-analysis-speed/ #gpu #programanalysis
September 29, 2025 at 1:59 PM
1960年代初頭あたりからアスキーアートでフローチャートを書くようなことはしてたんだな
blog.livedoor.jp/suzanhud/arc...
IBM DOCUMENTATION AIDS SYSTEM : suzanhudのblog
(これは以前書いたものの修正版です) IBM DOCUMENTATION AIDS SYSTEM adc00227は1401-SE-12Xで「IBM DOCUMENTATION AIDS SYSTEM」(DAと略していますね)ですね。以下の4つのプログラムで構成されています。Update ProgramAnalysis ProgramFlowchart Program (Flowcharter)V...
blog.livedoor.jp
March 31, 2025 at 1:54 PM
The Codellm-Devkit: A Framework To Contextualize Code #LLMs with #ProgramAnalysis Insights Tool Demo #TechnicalBriefing at #icse2025 will present an open-source library that simplifies performing program analysis at various levels of granularity.
See more: conf.researchr.org/info/icse-20...
Technical Briefings - ICSE 2025
ICSE, the IEEE/ACM International Conference on Software Engineering, is the premier software engineering conference. It will be held April 27-May 3 2025 in Ottawa. Core conference days will be Wednesd...
conf.researchr.org
December 10, 2024 at 2:41 PM
For #ACSAC2023 we have seven different pre-conference workshop with topics ranging from #IndustrialControlSystems, IoT, #ThreatIntelligence, #ProgramAnalysis, # ProgramVerification, #Forensics, #IncidentResponse, #BlockchainSecurity, to #InsiderThreats. www.acsac.org/2023/w...
Workshops - Annual Computer Security Applications Conference (ACSAC)
The Annual Computer Security Applications Conference (ACSAC) brings together cutting-edge researchers, with a broad cross-section of security professionals drawn from academia, industry, and government, gathered to present and discuss the latest security results and topics. With peer reviewed technical papers, invited talks, panels, national interest discussions, and workshops, ACSAC continues its core mission of investigating practical solutions for computer and network security technology.
www.acsac.org
November 9, 2023 at 3:00 PM