Monte Carlo study of a kinetic lattice model with random diffusion of disorder

We report Monte Carlo results for a nonequilibrium Ising-like model in two and three dimensions. Nearest-neighbor interactions J change sign randomly with time due to competing kinetics. There follows a fast and random, i.e., spin-configuration-independent diffusion of Js, of the kind that takes place in dilute metallic alloys when magnetic ions diffuse. The system exhibits steady states of the ferromagnetic (antiferromagnetic) type when the probability p that J>0 is large (small) enough. No counterpart to the freezing phenomena found in quenched spin glasses occurs. We compare our results with existing mean-field and exact ones, and obtain information about critical behavior.

Shallow-transfer rule-based machine translation for Czech to Polish

This article describes the developmentof an Open Source shallow-transfer machine translation system from Czech to Polish in theApertium platform. It gives details ofthe methods and resources used in contructingthe system. Although the resulting system has quite a high error rate, it is still competitive with other systems.

Automatic acquisition of Named Entities for Rule-Based Machine Translation

This paper proposes to enrich RBMTdictionaries with Named Entities(NEs) automatically acquired fromWikipedia. The method is appliedto the Apertium English-Spanishsystem and its performance comparedto that of Apertium with and withouthandtagged NEs. The system withautomatic NEs outperforms the onewithout NEs, while results vary whencompared to a system with handtaggedNEs (results are comparable forSpanish to English but slightly worstfor English to Spanish). Apart fromthat, adding automatic NEs contributesto decreasing the amount of unknownterms by more than 10%.

Bootstrapping a statistical speech translator from a rule-based one

We describe a series of experiments in which we start with English to French and English to Japanese versions of an Open Source rule-based speech translation system for a medical domain, and bootstrap correspondign statistical systems. Comparative evaluation reveals that the rule-based systems are still significantly better than the statistical ones, despite the fact that considerable effort has been invested in tuning both the recognition and translation components; also, a hybrid system only marginally improved recall at the cost of a los in precision. The result suggests that rule-based architectures may still be preferable to statistical ones for safety-critical speech translation tasks.

Epître dédicatoire à M. le Président du Conseil des ministres. [Signé : Henry Saint-Simon. - Première opinion politique des industriels. - Premier chant des industriels, signé Rouget de Lisle.]

Comprend : Premier chant des industriels