An L^* Algorithm for Deterministic Weighted Regular Languages
Not on the map. Why?
Abstract (arXiv)
Extracting finite state automata (FSAs) from black-box models offers a powerful approach to gaining interpretable insights into complex model behaviors. To support this pursuit, we present a weighted variant of Angluin's (1987) L^* algorithm for learning FSAs. We stay faithful to the original algorithm, devising a way to exactly learn deterministic weighted FSAs whose weights support division. Furthermore, we formulate the learning process in a manner that highlights the connection with FSA minimization, showing how L^* directly learns a minimal automaton for the target language.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 6 of its 27 references that are in the map. First posted on arXiv 9 Nov 2024.
Builds on
6 of 27 references.
References · 6
- 2017Extracting Automata from Recurrent Neural Networks Using Queries and Counterexamplescited by 66
- 2019Sequential Neural Networks as Automatacited by 54
- 2024On the Representational Capacity of Neural Language Models with Chain-of-Thought Reasoningcited by 26
- 2019Learning Deterministic Weighted Automata with Queries and Counterexamplescited by 22
- 2019Weighted Automata Extraction from Recurrent Neural Networks via Regression on State Spacescited by 21
- 2023Transparency at the Source: Evaluating and Interpreting Language Models With Access to the True Distributioncited by 10
Built on by
Citing papers per year: 2026: 1.
1 paper.
- 2026SMT-Based Active Learning of Weighted Automatacited by 0
- 2026SMT-Based Active Learning of Weighted Automata8 May 2026