Handling Exceptions and Effects with Automatic Resource Analysis
2026-03-04T19:53:49Z•45527ac6cbbefd660ad56c281728c5f04db0442961175cdbbee0299d5b162020
AARACUDACUTLASSCuTeGPURTL-to-specificationStitchCUDAagentic-RLconcurrencyeffectsexceptionsformal-verificationhappens-beforekernel-generationmulti-agent-systemsperformance-engineeringprobabilistic-programsprogramming-languagesrace-detectionresource-analysisreward-transformationstatic-analysissync-preserving-racestensor-layoutwp-reasoning
What happened
This silver document aggregates recent PL/SE research (arXiv) covering: (1) a novel automatic amortized resource analysis (AARA) that supports non-local control (exceptions/effects) with type-soundness against a stack-based abstract machine; (2) a programmatic reward-transformation enabling many probabilistic-program objectives to be analyzed via probabilistic weakest-precondition reasoning; (3) efficient, low-space dynamic algorithms for predicting short data races (happens-before and sync-preserving variants); (4) CuTe, a hierarchical tensor layout representation and algebra used in NVIDIA’s
Why it matters
A reviewed impact interpretation has not been published for this record.
Evidence and limitations
- Source ID
- arxiv_cs_pl
- Record identifier
- 45527ac6cbbefd660ad56c281728c5f04db0442961175cdbbee0299d5b162020
- Enrichment time
- 2026-03-04T19:53:49Z
- AI-assisted enrichment
- Yes
This record may overlap with other records. Its enrichment can be incomplete or wrong, and machine assistance was used. Validate consequential decisions against the linked source and your own environment.