BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//cfp.ducttape.events//ethereum-zurich-2026//speaker//9DUD
 Q3
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-ethereum-zurich-2026-DFW9WF@cfp.ducttape.events
DTSTART;TZID=CET:20260410T154500
DTEND;TZID=CET:20260410T160500
DESCRIPTION:Many Ethereum smart contracts are deployed without verified sou
 rce code\, making auditing\, security analysis\, and research difficult in
  practice. Decompilation aims to bridge this gap by recovering readable\, 
 high-level code from EVM bytecode\, but doing so reliably remains challeng
 ing.\n\nThis talk shares our experience using large language models for sm
 art contract decompilation\, based on building and evaluating multiple app
 roaches. I will discuss different ways LLMs can be applied to decompilatio
 n and the practical trade-offs involved. A central focus of the talk is ho
 w to evaluate decompilers in a meaningful way. I will explain why traditio
 nal metrics are insufficient\, what properties are important in practice.\
 n\nThe goal of this session is to provide a clear\, experience-based view 
 of what LLM-based decompilation can achieve today\, where it still falls s
 hort\, and how careful benchmarking helps guide both research and real-wor
 ld tooling.
DTSTAMP:20261013T172843Z
LOCATION:Glatt (Y22-F-68/70)
SUMMARY:Using Large Language Models to Decompile Ethereum Smart Contracts -
  Kaihua Qin
URL:https://cfp.ducttape.events/ethereum-zurich-2026/talk/DFW9WF/
END:VEVENT
END:VCALENDAR