The Paralax Infrastructure: Automatic Parallelization With a Helping Hand

Speeding up sequential programs on multicores is a challenging problem that is in urgent need of a solution. Automatic parallelization of irregular pointer-intensive codes, exempli?ed by the SPECint codes, is a very hard problem. This paper shows that, with a helping hand, such auto-parallelization is possible and fruitful. This paper makes the following contributions: (i) A compiler framework for extracting pipeline-like parallelism from outer program loops is presented. (ii) Using a light-weight programming model based on annotations, the programmer helps the compiler to ?nd thread-level parallelism. Each of the annotations speci?es only a small piece of semantic information that compiler analysis misses, e.g. stating that a variable is dead at a certain program point. The annotations are designed such that correctness is easily veri?ed. Furthermore, we present a tool for suggesting annotations to the programmer. (iii) The methodology is applied to autoparallelize several SPECint benchmarks. For the benchmark with most parallelism (hmmer), we obtain a scalable 7-fold speedup on an AMD quad-core dual processor. The annotations constitute a parallel programming model that relies extensively on a sequential program representation. Hereby, the complexity of debugging is not increased and it does not obscure the source code. These properties could prove valuable to increase the ef?ciency of parallel programming.

Model of access to natural language sources in electronic government

For the actual existence of e-government it is necessary and crucial to provide public information and documentation, making its access simple to citizens. A portion, not necessarily small, of these documents is in an unstructured form and in natural language, and consequently outside of which the current search systems are generally able to cope and effectively handle. Thus, in thesis, it is possible to improve access to these contents using systems that process natural language and create structured information, particularly if supported in semantics. In order to put this thesis to test, this work was developed in three major phases: (1) design of a conceptual model integrating the creation of structured information and making it available to various actors, in line with the vision of e-government 2.0; (2) definition and development of a prototype instantiating the key modules of this conceptual model, including ontology based information extraction supported by examples of relevant information, knowledge management and access based on natural language; (3) assessment of the usability and acceptability of querying information as made possible by the prototype - and in consequence of the conceptual model - by users in a realistic scenario, that included comparison with existing forms of access. In addition to this evaluation, at another level more related to technology assessment and not to the model, evaluations were made on the performance of the subsystem responsible for information extraction. The evaluation results show that the proposed model was perceived as more effective and useful than the alternatives. Associated with the performance of the prototype to extract information from documents, comparable to the state of the art, results demonstrate the feasibility and advantages, with current technology, of using natural language processing and integration of semantic information to improve access to unstructured contents in natural language. The conceptual model and the prototype demonstrator intend to contribute to the future existence of more sophisticated search systems that are also more suitable for e-government. To have transparency in governance, active citizenship, greater agility in the interaction with the public administration, among others, it is necessary that citizens and businesses have quick and easy access to official information, even if it was originally created in natural language.

Can virtual environments enhance the learning of historical chronology?

Historical time and chronological sequence are usually conveyed to pupils via the presentation of semantic information on printed worksheets, events being rote-memorised according to date. We explored the use of virtual environments in which successive historical events were depicted as “places” in time–space, encountered sequentially in a fly-through. Testing was via “Which came first, X or Y?” questions and picture-ordering. University undergraduates experiencing the history of an imaginary planet performed better after a VE than after viewing a “washing line” of sequential images, or captions alone, especially for items in intermediate list positions. However, secondary children 11–14 years remembered no more about successive events in feudal England when they were presented virtually compared with either paper picture or 2-D computer graphic conditions. Primary children 7–9 years learned more about historical sequence after studying a series of paper images, compared with either VE or computer graphic conditions, remembering more in early/intermediate list positions. Reasons for the discrepant results are discussed and future possible uses of VEs in the teaching of chronology assessed. Keywords: timeline, chronographics

The generation effect and word learning: a test of the effect of the language experience approach versus the text approach in the acquisition of new reading vocabulary

A class of twenty-two grade one children was tested to determine their reading levels using the Stanford Diagnostic Reading Achievement Test. Based on these results and teacher input the students were paired according to reading ability. The students ages ranged from six years four months to seven years four months at the commencement of the study. Eleven children were assigned to the language experience group and their partners became the text group. Each member of the language experience group generated a list of eight to be learned words. The treatment consisted of exposing the student to a given word three times per session for ten sessions, over a period of five days. The dependent variables consisted of word identification speed, word identification accuracy, and word recognition accuracy. Each member of the text group followed the same procedure using his/her partner's list of words. Upon completion of this training, the entire process was repeated with members of the text group from the first part becoming members of the language experience group and vice versa. The results suggest that generally speaking language experience words are identified faster than text words but that there is no difference in the rate at which these words are learned. Language experience words may be identified faster because the auditory-semantic information is more readily available in them than in text words. The rate of learning in both types of words, however, may be dictated by the orthography of the to be learned word.

Procedures as a Representation for Data in a Computer Program for Understanding Natural Language

This paper describes a system for the computer understanding of English. The system answers questions, executes commands, and accepts information in normal English dialog. It uses semantic information and context to understand discourse and to disambiguate sentences. It combines a complete syntactic analysis of each sentence with a "heuristic understander" which uses different kinds of information about a sentence, other parts of the discourse, and general information about the world in deciding what the sentence means. It is based on the belief that a computer cannot deal reasonably with language unless it can "understand" the subject it is discussing. The program is given a detailed model of the knowledge needed by a simple robot having only a hand and an eye. We can give it instructions to manipulate toy objects, interrogate it about the scene, and give it information it will use in deduction. In addition to knowing the properties of toy objects, the program has a simple model of its own mentality. It can remember and discuss its plans and actions as well as carry them out. It enters into a dialog with a person, responding to English sentences with actions and English replies, and asking for clarification when its heuristic programs cannot understand a sentence through use of context and physical knowledge.

Distributional semantics and machine learning for statistical machine translation

[EU]Lan honetan semantika distribuzionalaren eta ikasketa automatikoaren erabilera aztertzen dugu itzulpen automatiko estatistikoa hobetzeko. Bide horretan, erregresio logistikoan oinarritutako ikasketa automatikoko eredu bat proposatzen dugu hitz-segiden itzulpen- probabilitatea modu dinamikoan modelatzeko. Proposatutako eredua itzulpen automatiko estatistikoko ohiko itzulpen-probabilitateen orokortze bat dela frogatzen dugu, eta testuinguruko nahiz semantika distribuzionaleko informazioa barneratzeko baliatu ezaugarri lexiko, hitz-cluster eta hitzen errepresentazio bektorialen bidez. Horretaz gain, semantika distribuzionaleko ezagutza itzulpen automatiko estatistikoan txertatzeko beste hurbilpen bat lantzen dugu: hitzen errepresentazio bektorial elebidunak erabiltzea hitz-segiden itzulpenen antzekotasuna modelatzeko. Gure esperimentuek proposatutako ereduen baliagarritasuna erakusten dute, emaitza itxaropentsuak eskuratuz oinarrizko sistema sendo baten gainean. Era berean, gure lanak ekarpen garrantzitsuak egiten ditu errepresentazio bektorialen mapaketa elebidunei eta hitzen errepresentazio bektorialetan oinarritutako hitz-segiden antzekotasun neurriei dagokienean, itzulpen automatikoaz haratago balio propio bat dutenak semantika distribuzionalaren arloan.

The neural response to emotional prosody, as revealed by functional magnetic resonance imaging

Prosody is an important feature of language, comprising intonation, loudness, and tempo. Emotional prosodic processing forms an integral part of our social interactions. The main aim of this study was to use bold contrast fMRI to clarify the normal functional neuroanatomy of emotional prosody, in passive and active contexts. Subjects performed six separate scanning studies, within which two different conditions were contrasted: (1) "pure" emotional prosody versus rest; (2) congruent emotional prosody versus 'neutral' sentences; (3) congruent emotional prosody versus rest; (4) incongruent emotional prosody versus rest; (5) congruent versus incongruent emotional prosody; and (6) an active experiment in which subjects were instructed to either attend to the emotion conveyed by semantic content or that conveyed by tone of voice. Data resulting from these contrasts were analysed using SPM99. Passive listening to emotional prosody consistently activated the lateral temporal lobe (superior and/or middle temporal gyri). This temporal lobe response was relatively right-lateralised with or without semantic information. Both the separate and direct comparisons of congruent and incongruent emotional prosody revealed that subjects used fewer brain regions to process incongruent emotional prosody than congruent. The neural response to attention to semantics, was left lateralised, and recruited an extensive network not activated by attention to emotional prosody. Attention to emotional prosody modulated the response to speech, and induced right-lateralised activity, including the middle temporal gyrus. In confirming the results of lesion and neuropsychological studies, the current study emphasises the importance of the right hemisphere in the processing of emotional prosody, specifically the lateral temporal lobes. (C) 2003 Elsevier Science Ltd. All rights reserved.

Motor brain regions are involved in the encoding of delayed intentions: a fMRI study

In studies of prospective memory, recall of the content of delayed intentions is normally excellent, probably because they contain actions that have to be enacted at a later time. Action words encoded for later enactment are more accessible from memory than those encoded for later verbal report [Freeman, J.E., and Ellis, J.A. 2003a. The representation of delayed intentions: A prospective subject-performed task? Journal of Experimental Psychology: Learning, Memory, and Cognition, 29, 976-992.]. As this higher assessibility is lost when the intended actions have to be enacted during encoding, or when a motor interference task is introduced concurrent to intention encoding, Freeman and Ellis suggested that the advantage of to-be-enacted actions is due to additional preparatory motor operations during encoding. Accordingly, in a fMRI study with 10 healthy young participants, we investigated whether motor brain regions are differentially activated during verbal encoding of actions for later enactment with the right hand in contrast to verbal encoding of actions for later verbal report. We included an additional condition of verbal encoding of abstract verbs for later verbal report to investigate whether the semantic motor information inherent in action verbs in contrast to abstract verbs activates motor brain regions different from those involved in the verbal encoding of actions for later enactment. Differential activation for the verbal encoding of to-be-enacted actions in contrast to to-be-reported actions was found in brain regions known to be involved in covert motor preparation for hand movements, i.e. the postcentral gyrus, the precuneus, the dorsal and ventral premotor cortex, the posterior middle temporal gyrus and the inferior parietal lobule. There was no overlap between these brain regions and those differentially activated during the verbal encoding of actions in contrast to abstract verbs for later verbal report. Consequently, the results of this fMRI study suggest the presence of preparatory motor operations during the encoding of delayed intentions requiring a future motor response, which cannot be attributed to semantic information inherent to action verbs. (c) 2006 Elsevier B.V. All rights reserved.

Verb retrieval in fluent aphasia: a clinical study

Background: Problems with lexical retrieval are common across all types of aphasia but certain word classes are thought to be more vulnerable in some aphasia types. Traditionally, verb retrieval problems have been considered characteristic of non-fluent aphasias but there is growing evidence that verb retrieval problems are also found in fluent aphasia. As verbs are retrieved from the mental lexicon with syntactic as well as phonological and semantic information, it is speculated that an improvement in verb retrieval should enhance communicative abilities in this population as in others. We report on an investigation into the effectiveness of verb treatment for three individuals with fluent aphasia. Methods & Procedures: Multiple pre-treatment baselines were established over 3 months in order to monitor language change before treatment. The three participants then received twice-weekly verb treatment over approximately 4 months. All pre-treatment assessments were administered immediately after treatment and 3 months post-treatment. Outcome & Results: Scores fluctuated in the pre-treatment period. Following treatment, there was a significant improvement in verb retrieval for two of the three participants on the treated items. The increase in scores for the third participant was statistically nonsignificant but post-treatment scores moved from below the normal range to within the normal range. All participants were significantly quicker in the verb retrieval task following treatment. There was an increase in well-formed sentences in the sentence construction test and in some samples of connected speech. Conclusions: Repeated systematic treatment can produce a significant improvement in verb retrieval of practised items and generalise to unpractised items for some participants. An increase in well-formed sentences is seen for some speakers. The theoretical and clinical implications of the results are discussed.

Visual motion interferes with lexical decision on motion words

Embodied theories of cognition propose that neural substrates used in experiencing the referent of a word, for example perceiving upward motion, should be engaged in weaker form when that word, for example ‘rise’, is comprehended. Motivated by the finding that the perception of irrelevant background motion at near-threshold, but not supra-threshold, levels interferes with task execution, we assessed whether interference from near-threshold background motion was modulated by its congruence with the meaning of words (semantic content) when participants completed a lexical decision task (deciding if a string of letters is a real word or not). Reaction times for motion words, such as ‘rise’ or ‘fall’, were slower when the direction of visual motion and the ‘motion’ of the word were incongruent — but only when the visual motion was at nearthreshold levels. When motion was supra-threshold, the distribution of error rates, not reaction times, implicated low-level motion processing in the semantic processing of motion words. As the perception of near-threshold signals is not likely to be influenced by strategies, our results support a close contact between semantic information and perceptual systems.

Processing empty categories in a second language: When naturalistic exposure fills the (intermediate) gap

An ongoing debate on second language (L2) processing revolves around whether or not L2 learners process syntactic information similarly to monolinguals (L1), and what factors lead to a native-like processing. According to the Shallow Structure Hypothesis (Clahsen & Felser, 2006a), L2 learners’ processing does not include abstract syntactic features, such as intermediate gaps of wh-movement, but relies more on lexical/semantic information. Other researchers have suggested that naturalistic L2 exposure can lead to native-like processing (Dussias, 2003). This study investigates the effect of naturalistic exposure in processing wh-dependencies. Twenty-six advanced Greek learners of L2 English with an average nine years of naturalistic exposure, 30 with classroom exposure, and 30 native speakers of English completed a self-paced reading task with sentences involving intermediate gaps. L2 learners with naturalistic exposure showed evidence of native-like processing of the intermediate gaps, suggesting that linguistic immersion can lead to native-like abstract syntactic processing in the L2.

The time-course of morphological constraints: Evidence from eye-movements during reading

Lexical compounds in English are constrained in that the non-head noun can be an irregular but not a regular plural (e.g. mice eater vs. *rats eater), a contrast that has been argued to derive from a morphological constraint on modifiers inside compounds. In addition, bare nouns are preferred over plural forms inside compounds (e.g. mouse eater vs. mice eater), a contrast that has been ascribed to the semantics of compounds. Measuring eyemovements during reading, this study examined how morphological and semantic information become available over time during the processing of a compound. We found that the morphological constraint affected both early and late eye-movement measures, whereas the semantic constraint for singular non-heads only affected late measures of processing. These results indicate that morphological information becomes available earlier than semantic information during the processing of compounds.

When semantics aids phonology: a processing advantage for iconic word forms in aphasia

Iconicity is the non-arbitrary relation between properties of a phonological form and semantic content (e.g. “moo”, “splash”). It is a common feature of both spoken and signed languages, and recent evidence shows that iconic forms confer an advantage during word learning. We explored whether iconic forms conferred a processing advantage for 13 individuals with aphasia following left-hemisphere stroke. Iconic and control words were compared in four different tasks: repetition, reading aloud, auditory lexical decision and visual lexical decision. An advantage for iconic words was seen for some individuals in all tasks, with consistent group effects emerging in reading aloud and auditory lexical decision. Both these tasks rely on mapping between semantics and phonology. We conclude that iconicity aids spoken word processing for individuals with aphasia. This advantage may be due to a stronger connection between semantic information and phonological forms.

Selecting object pairs for action: is the active object always first?

Perception is linked to action via two routes: a direct route based on affordance information in the environment and an indirect route based on semantic knowledge about objects. The present study explored the factors modulating the recruitment of the two routes, in particular which factors affecting the selection of paired objects. In Experiment 1, we presented real objects among semantically related or unrelated distracters. Participants had to select two objects that can interact. The presence of distracters affected selection times, but not the semantic relations of the objects with the distracters. Furthermore, participants first selected the active object (e.g. teaspoon) with their right hand, followed by the passive object (e.g. mug), often with their left hand. In Experiment 2, we presented pictures of the same objects with no hand grip, congruent or incongruent hand grip. Participants had to decide whether the two objects can interact. Action decisions were faster when the presentation of the active object preceded the presentation of the passive object, and when the grip was congruent. Interestingly, participants were slower when the objects were semantically but not functionally related; this effect increased with congruently gripped objects. Our data showed that action decisions in the presence of strong affordance cues (real objects, pictures of congruently gripped objects) relied on sensory-motor representation, supporting the direct route from perception-to-action that bypasses semantic knowledge. However, in the case of weak affordance cues (pictures), semantic information interfered with action decisions, indicating that semantic knowledge impacts action decisions. The data support the dual-route account from perception-to-action.

Verbs in space: axis and direction of motion norms for 299 English verbs

A strong body of work has explored the interaction between visual perception and language comprehension; for example, recent studies exploring predictions from embodied cognition have focused particularly on the common representation of sensory—motor and semantic information. Motivated by this background, we provide a set of norms for the axis and direction of motion implied in 299 English verbs, collected from approximately 100 native speakers of British English. Until now, there have been no freely available norms of this kind for a large set of verbs that can be used in any area of language research investigating the semantic representation of motion. We have used these norms to investigate the interaction between language comprehension and low-level visual processes involved in motion perception, validating the norming procedure’s ability to capture the motion content of individual verbs. Supplemental materials for this study may be downloaded from brm.psychonomic-journals.org/content/supplemental.