44 Matching Annotations
  1. Feb 2019
  2. www.iro.umontreal.ca www.iro.umontreal.ca
    1. For the slot filling task, the input is the sentence consisting of a sequence of words, L, and the output is a sequence of slot/concept IDs, S, one for each word. In the statistical SLU systems, the task is often formalized as a pattern recognition problem: Given the word sequence L, the goal of SLU is to find the semantic representation of the slot sequence 푆that has the maximum a posterioriprobability 푃(푆|퐿).

      对于填槽任务,输入是一个有一系列词组成的语句,输出是每个词对应的slot/concept IDs。在统计SLU系统里,这个任务可以看作是:给定词序列L,SLU的目标是找到一个slot 序列来最大化后验概率P(S/L).

    2. Using Recurrent Neural Networksfor Slot Filling in Spoken Language Understanding

  3. Apr 2017
    1. WALTER 1. ONG, SJ

      Former SLU faculty and a really remarkable guy. I've been trawling through the archives and interviewing former colleagues and students of Ong for an ongoing project, and I'm continuously struck by the personality that comes across. You don't really see it in his published works, but his lectures have these corny, not-really-a-joke jokes ("it's agreed that these epic poems were either written by Homer, or by another man of the same name") in them. Also, this essay is a good example of him touching on history, archaeology, musicology, ancient Greek, and a bit of Freud, so if you're like me and a hugely disorganized mess of interests, Ong's a lot of fun.

      Seriously, if you get the chance, check out the archives, digitized or the whole thing. You can find his poetry from before the US entered WW2 all the way to discussions about flame wars in the early 90s.

  4. Mar 2017
    1. or instance, young animals are endearing, even when the adults of the species are not; so the child may feel encouraged to derive not only the obvious contextual implication that he is dirty, but also the further contextual implication that he is, nevertheless, endearing.

      In this case we have a scenario like the Juliet one where the metaphor is very rich and can have many possibly characteristics to pick out. The related characteristics each have to have a certain weight so that the hearer can have a clue on which one is the most relevant in the instance. It seems to me that in this example, void of context, uttering these words can leave the hearer puzzled if the reference is for being dirty or being endearing, while in the Juliet example which is much more rich in context, we can exclude that Romeo is saying Juliet is a celestial object.

    2. The greater effort imposed indicates that greater effect is intended. By uttering (13), the speaker thus encourages the hearer to look for a range of further contextual implications not shared by (14), and to assume that within this range there are some that she is prepared to endors

      So would the authors say that even though the hyperbole is blatantly false (because we haven't gone to every person in the world and measured their niceness), the intended meaning is still safe because the relevance of the statement still holds that the utterer wanted to convey this higher degree of intended effect?

    3. f we are right, loose uses are non-literal uses in the sense described above: they are based on resemblance relations among representations, and involve interpretive rather than descriptive dimensions of language use

      In order to determine the "loosness" of a statement however, we need context, that is we must call upon pragmatics. If the speaker utters "I live in Paris" and we have no context about the exact location of her residence, then we might as well believe she lives right in the center of the city, and then the truth-conditional semantics seem to be false.

    4. The qualification 'near Paris' demands some processing effort, which, given the presumption of relevance, should be offset by some cognitive effect

      The problem I have with "near Paris" is that regarding each city, there are norms of what is included within the city and what is not. For example, many of you may have heard from philosophical readings of Piraeus, the port of Athens. Now, Piraeus is not in Athens (and I believe never was, not even in Aristotle's time let's say). However, in contemporary common conversation, Piraeus is often said to be in Athens, for sake of simplicity, even though it is located quite some time from the center of Athens, similar to the distance I imagine Issy-les-Moulineaux to be located from the center of Paris. If someone tells me "I live in Athens" and someone else tells me "I live in Piraeus", even though the two are picking out different locations in essence, to me this has no significance, partly because everything is centered around a big city, including its peripheries, but also because I know the people in Piraeus and the people in Athens live live quite the same lives. In my opinion the statements I live in Paris and I live near Paris don't have much difference and their relevance are more or less the same

    5. Hence, Peter is entitled to assume that Marie intended him to interpret her utterance in this way, which is consistent with the principle of relevance

      I get the point being made in this paragraph and the her statement is heavily based on the relevance to the hearer, Peter. But if we consider this example in a normal everyday conversation, one living in this area just outside of Paris would most likely say that she/he lived in Paris no matter what, because the distance from Paris is insignificant, in my opinion, to make this distinction. I say this because someone living in the outskirts of the city or just outside the city is still closely related to the city. To make my point a little more clear, when I try to explain to my friends in Greece where in America I live, more often than not I simply say "45 minutes out of New York". To the hearer, especially one that's never been to New York, the relevance factor is focused on the city, even though I have some distance from it and I'm by no means living in the city. But my point is not undermined, because despite me living outside the city, I still have gone to the city many times and I can still make this point even though I know my hearer will find it more relevant that I live within distance of New York rather than that I live in Connecticut.

    6. n a nutshell, for an utterance to be understood, it must have one and only one interpretation consistent with the fact that the speaker intended it to seem relevant to the hearer-adequately relevant on the effect side and maximally relevant on the effort side

      This is confusing to me, because when I think back to the Juliet example, the utterance "Juliet is the sun" there is certainly not one and only one interpretation. Further, there is a different level of relevance of each interpretation of the utterance.

    7. conveyed, but not asserted

      I am confused as to the difference between these two terms in this context...

    8. Speakers may try hard or not at all to be relevant to their audience

      Is it that difficult to be relevant to your audience?

    9. In such a process, the hearer is taking a large share of the responsibility for the conclusions he arrives at. As a result, different hearers with different background knowledge and different imaginations will follow somewhat different routes.

      So different people will get different things out of the same metaphor

    10. We want to argue that they differ not in kind but only in degree of looseness

      First impression: this thesis sets up an argument that seems nit-picky and seemingly unnecessary. Differentiating between degrees and types is a way of putting things into boxes that doesn't necessarily represent the way things are well, just the way in which an individual views them.

    11. We claim that interests are simply by-products of the general search for relevance: as a result of our cognitive history, some topics in our memory are richer in information and, either temporarily or permanently, more accessible than others, so that information relating to them is likely to produce greater effect for less effort, i.e. be more relevant as defined.

      Interesting point

    12. On this approach, the claim that a man with one hair is bald is just as false as the claim that a man with a full head of hair is bald. What distinguishes them is not the fact that one is true and the other false, but the fact that one is an acceptable loose use because many of its logical and contextual implications are true, whereas the other is unacceptable since a hearer would be able to derive from it virtually no true descriptive information about the state of affairs it purports to represent

      useful distinction

    13. nd implication (9), which is contextually implied by both

      I wouldn't agree that (4) implies (9), even though it is a possibility, it doesn't necessarily follow from "it is winter" that "we should stay at home"

    14. Marie's first answer, 'I live in Paris', is effective enough to convey just what she wants; it may be more effective than the literally true second answer, 'I live near Paris'.

      Another helpful example

    15. ny object in the world can, in principle, be used to represent any other object that it resembles. For instance, a piece of rope can be used to represent a snake which it resembles in shape

      I agree with this point, but just playing a little devils advocate, what allows us to do this? That is, why can we use one object to represent another that it resembles? What allows us to do this? Is it just because of their sheer resemblance?

    16. but also calculable

      What is entailed by an implicature being calculable?

    17. : it is an exceptionless generalisation about human communicative behaviour.

      What makes the principle of relevance different from every other principle, etc. in modern pragmatics

    18. Any utterance addressed to someone automatically conveys a presumption of its own relevance. This fact, we call the principle of relevanc

      The principle of relevance

    19. We claim that humans automatically aim at maximal relevance, i.e. maximal cognitive effect for minimal processing effort.

      Claim concerning human information processing

    20. For instance, when I draw you a diagram of how to get to my house, you do not infer that I intend you to travel across white paper, in two dimensions, past landmarks clearly labelled CHURCH and NEWSPAPER SHOP, a distance of 8 inches from door to door. You have to make some assumption about which properties of the representation carry over to the original.

      Helpful example

    21. Every utterance used in verbal communication interpretively represents a thought entertained by the speaker-the very thought that the speaker wants to communicate. That much the hearer is entitled to expect; that much is necessary for verbal communication to be possible at all. However, the hearer is not invariably entitled to expect a literal interpretation of the speaker's thought, nor is such an interpretation always necessary for successful communi- cation to take place. A less-than-literal interpretation of the speaker's thought may be good enough: may indeed be better on some occasions than a strictly literal one.

      So a literal interpretation may not be what a hearer gets, but they can expect that what's being said is an interpretive representation of what the speaker is thinking

    22. Let us say that when one representation is interpretively used to represent another, all of whose implications it shares, it is a literal interpretation of that other representation. On this account, literalness is just a limiting case of interpretive resemblance.

      So two representations with all the same implications are literal interpretations of each other

    23. By our definition, propositions (4) and (5) resemble one another more in context (6a-b) than in context (7a-b): in (6a-b) they share implication (8), which is contexually implied by (4) and analytically implied by (5), and implication (9), which is contextually implied by both; whereas in (7a-b), (4) and (5) share no implications at all. This seems to match our intuitions, insofar as intuitions are possible given the artificiality of the example.

      Could use some clarification here

    24. We are thus defining interpretive resemblance as a context- dependent notion: two propositions P and Qmay resemble one another closely in one context and less closely or not at all in another context.

      Interpretive resemblance depends on context

    25. interpretati

      Representation in virtue of resemblance

    26. r descripti

      Representation in virtue of truth-conditions

    27. Any object in the world can, in principle, be used to represent any other object that it resembles. For instance, a piece of rope can be used to represent a snake which it resembles in shape. An utterance can be used to represent another utterance which it resembles in meaning-either closely, as in the case of a paraphrase or translation, or more distantly, as in the case of a summary. Generally speaking, an utterance can be used to represent any representation which it resembles in content, whether a public representation such as another utterance, or a mental representation such as a thought.

      New idea

    28. Rather, she intends him to entertain the proposition(s) most salient in her mind and to construct around it (or them) a complex thought which merely bears some similarity to her own. For instance, the mother wants the child to realise quite clearly that she thinks he is dirty, and to get at least an inkling of her accompanying thoughts

      So a speaker using a metaphor doesn't intend for their hearer to get their exact same thoughts/associations, rather they want the hearer to understand the most important associations and then get a general idea of the other accompanying thoughts

    29. In her modest way, the mother who calls her child a piglet achieves some unparaphrasable effects: for instance, she seems more indulgent than if she had called him a dirty child.

      Similar to what we were talking about last class - metaphors seem to have meaning beyond what can be translated into literal language

    30. Such views are no longer considered adequate to account for other cognitive abilities, but are still called upon, for want of any alternative, when it comes to explaining what is evoked by a metonymy, a synecdoche, a metaphor, or an irony. No other explanation is given of how figurative interpretations are recovered. Grice's account merely adds an inferential step of confirmation to these mysteriously retrieved figurative interpre- tations.

      So association isn't enough to explain how the figurative meanings of metaphors, etc. come to a hearer's mind

    31. The initial implausibility of any hypothetical rule of literal truthfulness might be overlooked if the appeal to such a rule had useful theoretical consequences; if it helped to explain how not only literal talk, but also loose talk and metaphor are understood. But in this respect, modern accounts are neither essentially different from, nor superior to, classical rhetorical accounts

      The problem with hypothetical rules of literal truthfulness : they don't account for loose talk and metaphor are understood

    32. One generally accepted answer is that there is a rule (or norm, or principle, or maxim, or convention, or presumption) of literal truthfulness whereby the utterer of a declarative sentence, in expressing a certain proposition, automatically vouches for its truth (similar rules of literal commitment can be formulated for non-declarative utterances)

      So Peter would assume that the car is actually in the garage simply because Mary said so

    33. Literal talk, loose talk and metaphorical talk are often seen as different in kind. We want to argue that they differ not in kind but only in degree of looseness, and that they are understood in essentially the same way.

      Seems like this is going to be the main argument of the paper

  5. Feb 2017
    1. or the task was to give the meaning of all expressions in a certain infinite set on the basis of the meaning of the parts ;

      So how does this infinite set of expressions apply to the synonyms of language or the ambiguous nature of language itself that is a necessary implicitly?

    2. rege's theor

      This idea was formulated in non-symbolic terms in his The Foundations of Arithmetic (1884). Later, in his Basic Laws of Arithmetic (vol. 1, 1893; vol. 2, 1903; vol. 2 was published at his own expense), Frege attempted to derive, by use of his symbolism, all of the laws of arithmetic from axioms he asserted as logical. Most of these axioms were carried over from his Begriffsschrift, though not without some significant changes. The one truly new principle was one he called the Basic Law V: the "value-range" of the function f(x) is the same as the "value-range" of the function g(x) if and only if ∀x[f(x) = g(x)]. The crucial case of the law may be formulated in modern notation as follows. Let {x|Fx} denote the extension of the predicate Fx, i.e., the set of all Fs, and similarly for Gx. Then Basic Law V says that the predicates Fx and Gx have the same extension iff ∀x[Fx ↔ Gx]. The set of Fs is the same as the set of Gs just in case every F is a G and every G is an F. (The case is special because what is here being called the extension of a predicate, or a set, is only one type of "value-range" of a function.) (stanford.edu)

    3. ost philosophers of language, and recently even by some linguists,

      What's the difference between the fields of philosophy of language and linguistics? This is a really interesting potential distinction.

  6. Jan 2017
    1. . There is only one world, the “real”world:

      This is a phenominological view, not necessarily an objective one...

    2. Logic, I should maintain, must no moreadmit a unicorn than zoology can; for logic is concerned with thereal world just as truly as zoology, though with its more abstract andgeneral features.

      Is logic built on science or is science built on logic?

    3. Who did you meet?” “I met a man.” “That is a very indefinitedescription.” We are therefore not departing from usage in our termi-nology. Our question is: What do I really assert when I assert “I meta man”? Let us assume, for the moment, that my assertion is true,and that in fact I met Jones. It is clear that what I assert isnot“I metJones.” I may say “I met a man, but it was not Jones”; in that case,though I lie, I do not contradict myself, as I should do if when I sayI met aj168man I really mean that I met Jones. It is clear also that theperson to whom I am speaking can understand what I say, even if heis a foreigner and has never heard of Jones

      Are there varying degrees of how definite things can be? For example, You can say that you met Jones, but Jones is not the Jones, he is simply a Jones, as you have not provided any distinction such as last name or characteristic.