Ask Me Anything with Işıl Dillig
summary
Covers practical advising tips — co-design problems with students, balance hands-off and hands-on mentorship, and foster group social time — plus concrete research directions: couple pruning the search space via program analysis and SMT with learned prioritization, combine multiple specification modalities (tests, language, logic), and shift from one-shot synthesis toward iterative human–tool conversations. Mentions influential techniques like CEGAR/interpolants and an ILP breakthrough (“cuts from proofs”), reports successful use of reinforcement learning for relational verification, and praises teaching discrete math.
AI-written summary · about
chapters (7)
- 0:00Start
- 1:12What Would You Be if You Were Not a Computer Scientist
- 3:25Three Things That I Try To Do Uh for Being a Good Advisor
- 7:23What Suggestions Do You Have on Doing Research in this Area for the Next 10 Years
- 11:30What Is One of Your Favorite Research Papers by Someone Else and Why It's a Good
- 12:55What Is the Story behind that Paper
- 19:23How Do You Manage Your Personal and Professional Life What Kind of Choices or Sacrifices Have You Made in Your Life To Be Successful in Research and Academia
related talks
32mStudent research contest talks
30mAsk the expert - Thinking Mathematically Above The Code Level
40mReasoning Under Uncertainty in SMT Solving, Research, and Life
31mLetting users do the undecidable: lessons from program synthesis
01h 01mChallenges in the Design and Implementation of Teaching Languages for EDSLs
53mSemantics-based program manipulation in the age of LLMs
01h 05mFrom meta frameworks and transformations to distributed computing and more
32mDon’t Call Us, We’ll Call You