"... In PAGE 7: ... This eliminated any unfair compiler advantages one ISA may have had over the other. Table2 lists the eight SPECint95 benchmarks used for our comparison 2, along with the input data sets, and the number of conventional ISA instructions required to run each benchmark to completion. Figure 3 compares the performance of the block- structured ISA executables to the performance of the con- ventional ISA executables when using a 64KB, four-way set associative L1 icache.... ..."

"... In PAGE 2: ..., 2005) and the eVOC anatomical ontology (Kelso et al., 2003) ( Table1 ). Relevance is gene db gene db Ontologies Gene db Input text Ranked query terms A Query results Ranked candidates Phenotype Mouse phenotypes Associations Anatomy Expression in relevant tissues Gene ontology Gene ontology similarity B MGD GAD Uni- Prot GOA Extracted genes Interaction partners Pathway members Protein domain similarity Genomic coord- inates Genes in genomic region BIND/ HPRD KEGG linkage peak Annotation dbs IPro Fig.... In PAGE 6: ... For comparison, we selected for each disease the relevant records from the OMIM database. For all tests the int4 method was used ( Table1 ). The test for candidate genes of myocardial infarction was omitted because the OMIM record for this disease is only C24100 words in length, which would be insufficient for reliably scoring a large number of ontology terms.... ..."

"... In PAGE 23: ... We then made three additional versions of these lists from which we deleted poten- tially harmful entries: one from which we removed all abbreviations that had obvious non-abbreviation homographs, one from which we removed all single character abbre- viations, and one from which we removed both. Table13 gives the number of remaining abbreviations for these different versions. We produced these additional versions to test how much care is needed when preparing abbreviation lists for a system like Punkt.... ..."

"... In PAGE 8: ...List of Tables Table 1. Software used to estimate vehicle weights 4 Table2 . Abbreviated version of the CONSIZE_MOD Text file 10 Table 3.... ..."

"... In PAGE 23: ... We then made three additional versions of these lists from which we deleted poten- tially harmful entries: one from which we removed all abbreviations that had obvious non-abbreviation homographs, one from which we removed all single character abbre- viations, and one from which we removed both abbreviations homographic to ordinary words and single character abbreviations. Table13 gives the number of remaining ab- breviations for these different versions. We produced the additional versions of the lists in order to test how much care is needed when preparing abbreviation lists for a sen- tence boundary detection system like ours.... ..."

"... In PAGE 22: ... The modifications on the test objects are based on Boundary Random Noise Model. Table1 presents the average retrieval accuracy of single object queries such that each query object (in original version) is queried with all of the objects in the noisy version. It is shown that the histogram-based approach retrieves the noisy version of the query object before any other object.... ..."

"... In PAGE 8: ... In- creasingly detailed comparisons between model and experimental data will be included along with each point. To ease comparisons, we group models into categories based on similarities in goals or implementation, Table1 . Structural and spectral models attempt to characterize map patterns using schematic drawings or concise equations.... In PAGE 26: ... The results of our comparison between model predictions and experimental data obtained from the upper layers of macaque striate cortex are summarized in table 2. References to articles on each model are given in Table1 . Many of the models are also brie y described in the appendix.... ..."

"... In PAGE 3: ... Each package has different versions. The name of the packages along with their abbreviations and versions are shown in Table1 . There are various constraints in terms of what one package depends ... ..."