TSNBench finds LLMs achieve 67-95% accuracy on TSN multiple-choice questions but produce 36-100% mean absolute percentage errors on open-ended worst-case delay calculations for Credit-Based Shaper and Cyclic Queuing and Forwarding.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.NI 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
TSNBench: Benchmarking LLM Proficiency in Time-Sensitive Networking
TSNBench finds LLMs achieve 67-95% accuracy on TSN multiple-choice questions but produce 36-100% mean absolute percentage errors on open-ended worst-case delay calculations for Credit-Based Shaper and Cyclic Queuing and Forwarding.