Poke here for What is a Word, What is a Sentence? Problems of Tokenization, by Gregory Grefenstette and Pasi Tapanainen, Rank Xerox Research Centre, Grenoble Laboratory, Meylan, France (April 22, 1994).

Poke here for Trevor Hopkins's Implementing State Machines in Smalltalk, Technical Report UMCS-93-3-1, Department of Computer Science, University of Manchester.

Poke here for Lauri Karttunen's Constructing Lexical Transducers, Rank Xerox Research Center, Grenoble, France (1994).

Poke here for Charles Crowley's A Finite State Machine for Western Swing (that's right, finite-state machines applied to country and western dancing).

Poke here for Fool's Gold: Extracting Finite State Machines From Recurrent Network Dynamics, by John F. Kolen, Laboratory for Artificial Intelligence Research, Ohio State University

Poke here for John McGregor's A Note on Inheritance and State Machines, Department of Computer Science, Clemson University.

Self Modifying Finite Automata
An SMFA is similar to an ordinary finite automaton, except that,
1.Each transition not only changes the state and reads an input symbol, it also modifies the set of transitions of the machine.
2.Each SMFA has a fixed finite set of registers, whose values are created states (as opposed to predefined states, which are part of the initial configuration of the machine). When creating a new transition, its source and destination states must be specified; the only way to specify a created state is by retrieving it from a register; and the value of a register can only be changed to a state that doesn't already exist. Thus, once a created state is no longer stored in the register that created it, there is no way to create more transitions to/from that state.

Probabilistic Analysis of Large Finite State Machines
Regarding finite state machines as Markov chains facilitates the application of probabilistic methods to very large logic synthesis and formal verification problems. In this paper we present symbolic algorithms to compute the steady-state probabilities for very large finite state machines. These algorithms, based on Algebraic Decision Diagrams (ADDs) -- an extension of BDDs that allows arbitrary values to be associated with the terminal nodes of the diagrams -- determine the steady-state probabilities by regarding finite state machines as homogeneous, discrete-parameter Markov chains with finite state spaces, and by solving the corresponding Chapman-Kolmogorov equations.

Exoterica White Paper on SGML Content Model Algebra
Discusses derivatives of finite-state machines and regular expressions.