QLL is a novel logic for neuro-symbolic learning that uses ML-native operations (sum, log-sum-exp) on logits to embed constraints, satisfying most linear logic properties and showing stronger correlation between empirical robustness and formal verification than prior approaches.
Strong, Clark W
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
years
2026 5verdicts
UNVERDICTED 5representative citing papers
A tensor-based batch fuzzing framework with adaptive perturbation scaling from specification ranges achieves up to 40X higher throughput and 4X more detected violations than sequential baselines on DNN benchmarks.
MLSkip demonstrates that lightweight metadata enables data skipping for ReLU-based ML filters, with 27.4% average pruning using min-max and 38.31% using 2D convex hulls on TPC benchmarks, for a 1.07x end-to-end speedup.
A ReLU-catalyzed abstraction method yields tighter bounds for transformer verification by converting dot-product constraints into ReLU forms that leverage standard convex relaxations.
A workshop report catalogues challenges and solution pathways for verification, engineering, and architecting reliable autonomous systems.
citing papers explorer
-
Quantitative Linear Logic for Neuro-Symbolic Learning and Verification
QLL is a novel logic for neuro-symbolic learning that uses ML-native operations (sum, log-sum-exp) on logits to embed constraints, satisfying most linear logic properties and showing stronger correlation between empirical robustness and formal verification than prior approaches.
-
Tensor-Based Batch Fuzzing with Adaptive Perturbation Scaling for Deep Neural Networks
A tensor-based batch fuzzing framework with adaptive perturbation scaling from specification ranges achieves up to 40X higher throughput and 4X more detected violations than sequential baselines on DNN benchmarks.
-
MLSkip: Data Skipping for ML Filters via Lightweight Metadata
MLSkip demonstrates that lightweight metadata enables data skipping for ReLU-based ML filters, with 27.4% average pruning using min-max and 38.31% using 2D convex hulls on TPC benchmarks, for a 1.07x end-to-end speedup.
-
Precise Verification of Transformers through ReLU-Catalyzed Abstraction Refinement
A ReLU-catalyzed abstraction method yields tighter bounds for transformer verification by converting dot-product constraints into ReLU forms that leverage standard convex relaxations.
-
Engineering Reliable Autonomous Systems: Challenges and Solutions
A workshop report catalogues challenges and solution pathways for verification, engineering, and architecting reliable autonomous systems.