Embedded Files
λambeq
λambeq
λambeq is a novel software package we developed at the Oxford office of Quantinuum, a high-level Python toolkit for experimental quantum natural language processing. The open-source toolkit offers a detailed hierarchy of modules and classes implementing all stages of a pipeline for converting sentences to string diagrams, tensor networks, and quantum circuits ready to be used on a quantum computer. λambeq supports syntactic parsing, rewriting and simplification of string diagrams, ansatz creation and manipulation, as well as a number of compositional models for preparing quantum-friendly representations of sentences, paragraphs, or even entire documents employing various degrees of syntax sensitivity.
For more information, have a look at the documentation.
The github repository can be found here.
Datasets
Datasets
A number of datasets created with Mehrnoosh Sadrzadeh on term-definition classification, sentence similarity, disambiguation, and textual entailment can be found at this QMUL's webpage.
DisCoCat
DisCoCat
A tool for converting any sentence to a DisCoCat diagram and quantum circuit, from joint work with Richie Yeung at Quantinuum [link]
The quantum circuit for a standard transitive sentence of the form subject verb object. Qubits q0, q4 correspond to subject and object, respectively, while q1-q3 represent the verb. Qubit q2 carries the result.
"Alice in Wonderland" in DisCoCat form [link]
Media
Media
Some of the various articles about λambeq's release [AI Business] [Forbes] [PR Newswire]
Some publicity on our paper at Quantinuum with Robin Lorenz, Anna Pearson, Konstantinos Meichanetzidis, and Bob Coecke, detailing the first NLP experiments on quantum hardware with datasets of more than 100 sentences [The Quantum Daily]
A nice series of popular science articles on previous work with Sanjaye Ramgoolam and Mehrnoosh Sadrzadeh [plus magazine]
Page updated
Google Sites
Report abuse