2 citations · 5 across the 6 of their papers we have counts for
Showing 2026Show all
2 papers · 1 filter
cs.AI2026
Verified but Vacuous: Three Failure Surfaces of LLM-Driven Verified Code Generation
Aditya Vempaty, Himaneesh Sompalle, Manan Tayal +4
What should ``verified'' mean when evaluating LLM-generated code? Many existing pipelines report a single verification number. We argue this collapses three distinct properties mea…
cs.SE2026
Eliciting Specifications: When Does Interactive Clarification Help?
Mowafak Allaham, Aditya Vempaty, Ashish Jagmohan +2
LLMs can generate Lean specifications from task descriptions, but those descriptions are often underspecified or ambiguous and may omit verifier-relevant edge cases, yielding imper…