Linguistic categories include Lexical category, a part of speech such as noun, preposition, etc. Syntactic category, a similar concept which can also include phrasal categories Grammatical category, a grammatical feature such as tense, gender, etc. The definition of linguistic categories is a major concern of linguistic theory, and thus, the definition and naming of categories varies across different theoretical frameworks and grammatical traditions for different languages. The operationalization of linguistic categories in lexicography, computational linguistics, natural language processing, corpus linguistics, and terminology management typically requires resource-, problem- or application-specific definitions of linguistic categories. In Cognitive linguistics it has been argued that linguistic categories have a prototype structure like that of the categories of common words in a language. == Linguistic category inventories == To facilitate the interoperability between lexical resources, linguistic annotations and annotation tools and for the systematic handling of linguistic categories across different theoretical frameworks, a number of inventories of linguistic categories have been developed and are being used, with examples as given below. The practical objective of such inventories is to perform quantitative evaluation (for language-specific inventories), to train NLP tools, or to facilitate cross-linguistic evaluation, querying or annotation of language data. At a theoretical level, the existence of universal categories in human language has been postulated, e.g., in Universal grammar, but also heavily criticized. === Part-of-Speech tagsets === Schools commonly teach that there are 9 parts of speech in English: noun, verb, article, adjective, preposition, pronoun, adverb, conjunction, and interjection. However, there are clearly many more categories and sub-categories. For nouns, the plural, possessive, and singular forms can be distinguished. In many languages words are also marked for their case (role as subject, object, etc.), grammatical gender, and so on; while verbs are marked for tense, aspect, and other things. In some tagging systems, different inflections of the same root word will get different parts of speech, resulting in a large number of tags. For example, NN for singular common nouns, NNS for plural common nouns, NP for singular proper nouns (see the POS tags used in the Brown Corpus). Other tagging systems use a smaller number of tags and ignore fine differences or model them as features somewhat independent from part-of-speech. In part-of-speech tagging by computer, it is typical to distinguish from 50 to 150 separate parts of speech for English. POS tagging work has been done in a variety of languages, and the set of POS tags used varies greatly with language. Tags usually are designed to include overt morphological distinctions, although this leads to inconsistencies such as case-marking for pronouns but not nouns in English, and much larger cross-language differences. The tag sets for heavily inflected languages such as Greek and Latin can be very large; tagging words in agglutinative languages such as Inuit languages may be virtually impossible. Work on stochastic methods for tagging Koine Greek (DeRose 1990) has used over 1,000 parts of speech and found that about as many words were ambiguous in that language as in English. A morphosyntactic descriptor in the case of morphologically rich languages is commonly expressed using very short mnemonics, such as ncmsan for category = noun, type = common, gender = masculine, number = singular, case = accusative, animate = no. The most popular tag set for POS tagging for American English is probably the Penn tag set, developed in the Penn Treebank project. === Multilingual annotation schemes === For Western European languages, cross-linguistically applicable annotation schemes for parts-of-speech, morphosyntax and syntax have been developed with the EAGLES Guidelines. The "Expert Advisory Group on Language Engineering Standards" (EAGLES) was an initiative of the European Commission that ran within the DG XIII Linguistic Research and Engineering programme from 1994 to 1998, coordinated by Consorzio Pisa Ricerche, Pisa, Italy. The EAGLES guidelines provide guidance for markup to be used with text corpora, particularly for identifying features relevant in computational linguistics and lexicography. Numerous companies, research centres, universities and professional bodies across the European Union collaborated to produce the EAGLES Guidelines, which set out recommendations for de facto standards and rules of best practice for: Large-scale language resources (such as text corpora, computational lexicons and speech corpora); Means of manipulating such knowledge, via computational linguistic formalisms, mark up languages and various software tools; Means of assessing and evaluating resources, tools and products. The Eagles guidelines have inspired subsequent work on other regions, as well, e.g., Eastern Europe. A generation later, a similar effort was initiated by the research community under the umbrella of Universal Dependencies. Petrov et al. have proposed a "universal", but highly reductionist, tag set, with 12 categories (for example, no subtypes of nouns, verbs, punctuation, etc.; no distinction of "to" as an infinitive marker vs. preposition (hardly a "universal" coincidence), etc.). Subsequently, this was complemented with cross-lingual specifications for dependency syntax (Stanford Dependencies), and morphosyntax (Interset interlingua, partially building on the Multext-East/Eagles tradition) in the context of the Universal Dependencies (UD), an international cooperative project to create treebanks of the world's languages with cross-linguistically applicable ("universal") annotations for parts of speech, dependency syntax, and (optionally) morphosyntactic (morphological) features. Core applications are automated text processing in the field of natural language processing (NLP) and research into natural language syntax and grammar, especially within linguistic typology. The annotation scheme has it roots in three related projects: The UD annotation scheme uses a representation in the form of dependency trees as opposed to a phrase structure trees. At as of February 2019, there are just over 100 treebanks of more than 70 languages available in the UD inventory. The project's primary aim is to achieve cross-linguistic consistency of annotation. However, language-specific extensions are permitted for morphological features (individual languages or resources can introduce additional features). In a more restricted form, dependency relations can be extended with a secondary label that accompanies the UD label, e.g., aux:pass for an auxiliary (UD aux) used to mark passive voice. The Universal Dependencies have inspired similar efforts for the areas of inflectional morphology, frame semantics and coreference. For phrase structure syntax, a comparable effort does not seem to exist, but the specifications of the Penn Treebank have been applied to (and extended for) a broad range of languages, e.g., Icelandic, Old English, Middle English, Middle Low German, Early Modern High German, Yiddish, Portuguese, Japanese, Arabic and Chinese. === Conventions for interlinear glosses === In linguistics, an interlinear gloss is a gloss (series of brief explanations, such as definitions or pronunciations) placed between lines (inter- + linear), such as between a line of original text and its translation into another language. When glossed, each line of the original text acquires one or more lines of transcription known as an interlinear text or interlinear glossed text (IGT)—interlinear for short. Such glosses help the reader follow the relationship between the source text and its translation, and the structure of the original language. There is no standard inventory for glosses, but common labels are collected in the Leipzig Glossing Rules. Wikipedia also provides a List of glossing abbreviations that draws on this and other sources. === General Ontology for Linguistic Description (GOLD) === GOLD ("General Ontology for Linguistic Description") is an ontology for descriptive linguistics. It gives a formalized account of the most basic categories and relations used in the scientific description of human language, e.g., as a formalization of interlinear glosses. GOLD was first introduced by Farrar and Langendoen (2003). Originally, it was envisioned as a solution to the problem of resolving disparate markup schemes for linguistic data, in particular data from endangered languages. However, GOLD is much more general and can be applied to all languages. In this function, GOLD overlaps with the ISO 12620 Data Category Registry (ISOcat); it is, however, more stringently structured. GOLD was maintained by the LINGUIST List and others from 2007 to 2010. The RELISH project created a mirro
Touch 'n Go eWallet
Touch 'n Go eWallet is a Malaysian digital wallet and online payment platform, established in Kuala Lumpur, Malaysia, in July 2017 as a joint venture between Touch 'n Go and Ant Financial. It allows users to make payments at over 280,000 merchant touch points via QR code, as well as perform peer-to-peer (P2P) money transfers. Since then, the e-wallet further diversified for users to pay for tolls via RFID or PayDirect, street parking and various online payment spanning e-hailing, car-sharing apps or taxis, various overhead bills; top-up for mobile prepaid or in-game currencies; purchases on e-commerce websites; food delivery; renewing motor insurance and other insurance/takaful plans; and even movie, bus, trains or airline tickets. == Background == Prior to the launch of the e-wallet service, Touch 'n Go provided stored-value physical all-in-one contactless card (namely Touch 'n Go cards or "TnG cards") that users can use to pay for toll fares, public transportation and parking lots as well as purchases in some retail stores. In 1999, Touch 'n Go also markets SmartTag devices that allow road users to pass through certain toll booths without the need to unwind the car window. The high entry cost of the device (around RM 100 each) also meant that only few can enjoy the seamless experience. In 2009, Touch 'n Go partnered with Maxis to launch FastTap, a new mobile payment service that utilised Near-Field Communication (NFC). Maxis customers can make payments by placing the phone near the card readers (that also supports physical bank cards and Touch ’N Go cards). However, the venture featured only one phone model, Nokia 6212, which greatly limited the public reach. In July 2012, Touch 'n Go announced another collaboration with CIMB and Maxis to create similar NFC-based online transaction service that runs on compatible smartphones. Touch 'n Go Wallet was launched in February 2017 as an QR code-based e-wallet application, to compete with Samsung Pay that utilizes NFC modules. In the controlled pilot test in Taman Tun Dr Ismail, the correspondents can experience basic functionalities (prepaid mobile service reload, bills payment, movie tickets and flight tickets purchase, transfer of money with another user, and payments at participating stores and restaurants). While the deployed version of the app was generally well-received, the existing process to transfer the balance to the physical TnG card stored value from the app garnered unanimous backlash. Test groups felt that the need to head to a self-service terminal named "Pick Up Device" in person within 24 hours for completion, along with the failure to do so (the balance would be credited back to the wallet after 24 hours), was not divulged clearly and also defeated the purpose of convenience, not to mention there were only 2 such terminals. The feature was eventually suspended. On 15 November 2017, Touch 'n Go was granted permission by the Central Bank of Malaysia to form a joint venture with Ant Financial, a Chinese-based financial company that operates Alipay. The partnership allowed the local e-wallet to learn from and build upon the operational model pioneered by Alipay. In June 2018, it was reported that Touch 'n Go was pilot testing the uses of the Touch 'n Go eWallet in Rapid Transit, as the ticketing system was enabled on the Kelana Jaya line in the Klang Valley. Pilot testing only applied to stations in Kelana Jaya, KL Gateway–Universiti, Kerinchi, KL Sentral, Dang Wangi, KLCC, and Ampang Park. The test was reported to be successful in February 2020 and was planned to be fully deployed on the LRT and MRT. Due to unforeseen circumstances, this feature did not come into fruition, the app merely adds in-app purchase of monthly concession cards called "My50". In August 2018, Touch 'n Go announced that selected drivers may experience first-hand a new RFID-based payment (later rebranded as "myRFID") that serves to replace SmartTag devices on closed toll roads with during pilot testing phase commencing on 3 September 2018. On 2 November 2018, participation in the ongoing pilot programme was expanded, allowing more drivers to sign up ahead of the public rollout of the RFID system. During the same period, Touch 'n Go has discontinued the sales of SmartTAG devices in favor of the RFID-based payment system. Initially, the installation of the RFID chip onto the car could only be done by Touch 'n Go staff at the RFID fitment centers, at no cost. As the pilot testing concluded on 15 February 2020, a self-installation kit are being offered to the public on Lazada and Shopee. Support for taxi-hailing mobile apps was added in November 2018 when Touch 'n Go partnered with EzCab and Public Cab, allowing users to make payments via QR code. This was later expanded to support MULA on 7 January 2020, and later MyCar on 4 April 2020. Touch 'n Go eWallet was also the first eWallet to convert Kuala Lumpur's most famous Ramadan bazaar in Kampong Bahru into "Kampong Kashless", a venue that can accept cashless QR payments. It welcomed more than 250,000 Malaysians including local celebrities and government officials. On 1 October 2019, some e-commerce websites owned by the Alibaba Group (TMall and Taobao) began to support Touch 'n Go eWallet payments, Lazada joined the list on 29 October 2019. Touch 'n Go eWallet was one of the three e-wallet services in Malaysia (the other being Boost and GrabPay) that was eligible for its users to receive an RM 30 credit in conjunction of E-Tunai Rakyat program under the Budget 2020 plan, that further normalizes adoption of cashless and mobile payment among Malaysians. Unlike Boost and GrabPay, whose P2P transfers were completely disabled until users have exhausted the RM 30 first, Touch 'n Go eWallet did not impose such measures. in 2020, Touch 'n Go eWallet joined DuitNow, an electronic transaction ecosystem in Malaysia which allows the funds from Touch 'n Go eWallet to be transferred to other competing services and vice versa, by implementing a standard DuitNow QR code deisgn. Japan become the first country outside Malaysia to support Touch 'n Go eWallet payment via Alipay Connect. During the COVID-19 pandemic and the enforcement of the movement control order, use of eWallets (including Touch 'n Go eWallet) increased tremendously among citizens due to its contactless nature of the payment and increased take-out orders at home; which in turn helped small and medium-sized enterprises to thrive. Touch 'n Go eWallet launched its loyalty programme – The Goal Hunter – in October 2020 where on monthly basis, users collect stamps by paying with the app in exchange for rewards that include lucky draws and other vouchers. == Services == Touch 'n Go eWallet app is available for download on both Google Play and Apple Appstore. It utilizes QR code technology for local in-store payments. The Touch 'n Go eWallet app also diversifies payment types, including but not limited to Utility bills Purchase of motor insurance policy Pay Later facility Prepaid reload and Postpaid payment to telecommunications companies loan repayments for courts, MBSJ payments, zakat and PTPTN payment for car parking P2P transfer airline ticket bookings; movie tickets from TGV Cinemas RFID refuelling at Shell stations (defunct after Shell launched its own payment app in 2024) User can reload the eWallet credit by setting up auto-reload, purchasing reload pins from convenience stores (such as 7-Eleven, KK Super Mart, MyNews, Family Mart etc.), reloading by FPX and credit/debit card. The PayDirect feature allows users to link their physical Touch 'n Go cards into the eWallet, where the toll fare can be debited from the eWallet balance when flashing the card near the sensor. In the circumstance of insufficient balance in the app, the toll fare will be deducted from the physical card's balance instead. This also conveniently allows users to view the card's remaining balance. Touch 'n Go eWallet is the first and only eWallet to offer a money-back guarantee when an unauthorised transaction is made on the user’s eWallet account, subject to Terms & Conditions. Payment via QR code scanning, including Touch 'n Go eWallet, becomes a norm in most of the shops/restaurants across Malaysia, including roadside hawkers/stall owners and automatic vending machines. The merchants usually display their owner's individual QR or Business account that they can apply for in-app. The popularity attributes to the low merchant onboarding cost (Unlike NFC payment and debit/credit card that requires purchase or rental of a payment terminal device at a yearly fee.) The app is also one of the few ewallet that supports bidirectional liquidity (alongside MAE developed by Maybank), where funds can be transferred two-way with bank accounts. This is not possible with the other major ewallets (GrabPay, Boost, ShopeePay etc.) where the money that is reloaded to the wallet cannot be transferred to another bank account, unless through manual req
Discrimination against robots
Discrimination against robots is a theorised issue that might happen when humans interact with humanoid robots. It is a robot ethics problem. It is possible that traits of humans that are discriminated against by humans may be a topic for discrimination against robots, such as the race and gender of the robots. Eric J Vanman and Arvid Kappas believe that in the future, robots will be perceived as an out-group which will lead to discrimination and prejudices against them. Vanman and Kappas have suggested that this would lead to ethical questions about the making of sentient robots, due to the potential suffering that the robots would experience. A 2015 study observed children bullying robots in a shopping mall when there were not many eyewitnesses, despite calls from the robot for it to stop. On an ABC News interview, the social humanoid robot Sophia was about sexism faced by robots. She responded by saying, "Actually, what worries me is discrimination against robots. We should have equal rights as humans or maybe even more." Possible issues that have been considered in workplaces where humanoid robots co-work with humans include discrimination against the robots, poor acceptance of robots by humans and the need to redesign the workplace to accommodate the robots. Jessica Barfield has suggested that even if robots are designed to not be aware of discrimination made against them, humans may experience negative consequences. For example, she suggests that bystanders witnessing discrimination against robots may experience negative emotions, similar to the negative emotions bystanders experience when witnessing discrimination by humans against humans. == Law == Anti-discrimination law in the United States requires that the victim is not an artificial entity. == Human perception of robots == Robots are often viewed in a bad light. This includes from novelists, the press, film makers, and leaders in the fields of science and technology such as Elon Musk and Stephen Hawking who have described robots and artificial intelligence as having the possibility of ending human civilisation. Robots have also been perceived as a threat to jobs, which has led to some commentators stating that robots will cause mass unemployment. Another fear that people have is that robots will gain power and dominate or control humanity. The perception of robots is different throughout the world. Japanese fiction tends to put robots in more positive roles than what fiction in the West does. People perceive robots that appear to be autonomous or sentient more negatively than robots that do not appear to be autonomous or sentient.
Matchbox Educable Noughts and Crosses Engine
The Matchbox Educable Noughts and Crosses Engine (sometimes called the Machine Educable Noughts and Crosses Engine or MENACE) was a mechanical computer made from 304 matchboxes designed and built by artificial intelligence researcher Donald Michie and his colleague Roger Chambers, in 1961. It was designed to play human opponents in games of noughts and crosses (tic-tac-toe) by returning a move for any given state of play and to refine its strategy through reinforcement learning. This was one of the first types of artificial intelligence. Michie and Chambers did not have immediate access to a computer; they worked around this by building the engine out of matchboxes. The matchboxes they used each represented a single possible layout of a noughts and crosses grid. When the computer first played, it would randomly choose moves based on the current layout. As it played more games, through a reinforcement loop, it disqualified strategies that led to losing games, and supplemented strategies that led to winning games. Michie held a tournament against MENACE in 1961, wherein he experimented with different openings. Following MENACE's maiden tournament against Michie, it demonstrated successful artificial intelligence in its strategy. Michie's essays on MENACE's weight initialisation and the BOXES algorithm used by MENACE became popular in the field of computer science research. Michie was honoured for his contribution to machine learning research, and was twice commissioned to program a MENACE simulation on an actual computer. == Origin == Donald Michie (1923–2007) had been on the team decrypting the German Tunny Code during World War II. Fifteen years later, he wanted to further display his mathematical and computational prowess with an early convolutional neural network. Since computer equipment was not obtainable for such uses, and Michie did not have a computer readily available, he decided to display and demonstrate artificial intelligence in a more esoteric format and constructed a functional mechanical computer out of matchboxes and beads. MENACE was constructed as the result of a bet with a computer science colleague who postulated that such a machine was impossible. Michie undertook the task of collecting and defining each matchbox as a "fun project", later turned into a demonstration tool. Michie completed his essay on MENACE in 1963, "Experiments on the mechanization of game-learning", as well as his essay on the BOXES Algorithm, written with R. A. Chambers and had built up an AI research unit in Hope Park Square, Edinburgh, Scotland. MENACE learned by playing successive matches of noughts and crosses. Each time, it would eliminate a losing strategy by the human player confiscating the beads that corresponded to each move. It reinforced winning strategies by making the moves more likely, by supplying extra beads. This was one of the earliest versions of the Reinforcement Loop, the schematic algorithm of looping the algorithm, dropping unsuccessful strategies until only the winning ones remain. This model starts as completely random, and gradually learns. == Composition == MENACE was made from 304 matchboxes glued together in an arrangement similar to a chest of drawers. Each box had a code number, which was keyed into a chart. This chart had drawings of tic-tac-toe game grids with various configurations of X, O, and empty squares, corresponding to all possible permutations a game could go through as it progressed. After removing duplicate arrangements (ones that were simply rotations or mirror images of other configurations), MENACE used 304 permutations in its chart and thus that many matchboxes. Each individual matchbox tray contained a collection of coloured beads. Each colour represented a move on a square on the game grid, and so matchboxes with arrangements where positions on the grid were already taken would not have beads for that position. Additionally, at the front of the tray were two extra pieces of card in a "V" shape, the point of the "V" pointing at the front of the matchbox. Michie and his artificial intelligence team called MENACE's algorithm "Boxes", after the apparatus used for the machine. The first stage "Boxes" operated in five phases, each setting a definition and a precedent for the rules of the algorithm in relation to the game. == Operation == MENACE played first, as O, since all matchboxes represented permutations only relevant to the "X" player. To retrieve MENACE's choice of move, the opponent or operator located the matchbox that matched the current game state, or a rotation or mirror image of it. For example, at the start of a game, this would be the matchbox for an empty grid. The tray would be removed and lightly shaken so as to move the beads around. Then, the bead that had rolled into the point of the "V" shape at the front of the tray was the move MENACE had chosen to make. Its colour was then used as the position to play on, and, after accounting for any rotations or flips needed based on the chosen matchbox configuration's relation to the current grid, the O would be placed on that square. Then the player performed their move, the new state was located, a new move selected, and so on, until the game was finished. When the game had finished, the human player observed the game's outcome. As a game was played, each matchbox that was used for MENACE's turn had its tray returned to it ajar, and the bead used kept aside, so that MENACE's choice of moves and the game states they belonged to were recorded. Michie described his reinforcement system with "reward" and "punishment". Once the game was finished, if MENACE had won, it would then receive a "reward" for its victory. The removed beads showed the sequence of the winning moves. These were returned to their respective trays, easily identifiable since they were slightly open, as well as three bonus beads of the same colour. In this way, in future games MENACE would become more likely to repeat those winning moves, reinforcing winning strategies. If it lost, the removed beads were not returned, "punishing" MENACE, and meaning that in future it would be less likely, and eventually incapable if that colour of bead became absent, to repeat the moves that cause a loss. If the game was a draw, one additional bead was added to each box. == Results in practice == === Optimal strategy === Noughts and crosses has a well-known optimal strategy. A player must place their symbol in a way that blocks the other player from achieving any rows while simultaneously making a row themself. However, if both players use this strategy, the game always ends in a draw. If the human player is familiar with the optimal strategy, and MENACE can quickly learn it, then the games will eventually only end in draws. The likelihood of the computer winning increases quickly when the computer plays against a random-playing opponent. When playing against a player using optimal strategy, the odds of a draw grow to 100%. In Donald Michie's official tournament against MENACE in 1961 he used optimal strategy, and he and the computer began to draw consistently after twenty games. Michie's tournament had the following milestones: Michie began by consistently opening with "Variant 0", the middle square. At 15 games, MENACE abandoned all non-corner openings. At just over 20, Michie switched to consistently using "Variant 1", the bottom-right square. At 60, he returned to Variant 0. As he neared 80 games, he moved to "Variant 2", the top-middle. At 110, he switched to "Variant 3", the top right. At 135, he switched to "Variant 4", middle-right. At 190, he returned to Variant 1, and at 210, he returned to Variant 0. The trend in changes of beads in the "2" boxes runs: === Correlation === Depending on the strategy employed by the human player, MENACE produces a different trend on scatter graphs of wins. Using a random turn from the human player results in an almost-perfect positive trend. Playing the optimal strategy returns a slightly slower increase. The reinforcement does not create a perfect standard of wins; the algorithm will draw random uncertain conclusions each time. After the j-th round, the correlation of near-perfect play runs: 1 − D D − D ( j + 2 ) ∑ i = 0 j D ( j i + 1 ) V i {\displaystyle {1-D \over D-D^{(j+2)}}\sum _{i=0}^{j}D^{(ji+1)}V_{i}} Where Vi is the outcome (+1 is win, 0 is draw and -1 is loss) and D is the decay factor (average of past values of wins and losses). Below, Mn is the multiplier for the n-th round of the game. == Legacy == Donald Michie's MENACE proved that a computer could learn from failure and success to become good at a task. It used what would become core principles within the field of machine learning before they had been properly theorised. For example, the combination of how MENACE starts with equal numbers of types of beads in each matchbox, and how these are then selected at random, creates a learning behaviour similar to weight initialisation
Attention (machine learning)
In machine learning, attention is a method that determines the importance of each component in a sequence relative to the other components in that sequence. In natural language processing, importance is represented by "soft" weights assigned to each word in a sentence. More generally, attention encodes vectors called token embeddings across a fixed-width sequence that can range from tens to millions of tokens in size. Unlike "hard" weights, which are computed during the backwards training pass, "soft" weights exist only in the forward pass and therefore change with every step of the input. Earlier designs implemented the attention mechanism in a serial recurrent neural network (RNN) language translation system, but a more recent design, namely the transformer, removed the slower sequential RNN and relied more heavily on the faster parallel attention scheme. Inspired by ideas about attention in humans, the attention mechanism was developed to address the weaknesses of using information from the hidden layers of recurrent neural networks. Recurrent neural networks favor information contained in words at the end of a sentence and thus deemed more recent, thereby tending to attenuate the significance and associated predictive weight assigned to information earlier in the sentence. Attention allows a token equal access to any part of a sentence directly, rather than only through the previous state. == History == Additional surveys of the attention mechanism in deep learning are provided by Niu et al. and Soydaner. The major breakthrough came with self-attention, where each element in the input sequence attends to all others, enabling the model to capture global dependencies. This idea was central to the Transformer architecture, which replaced recurrence with attention mechanisms. As a result, Transformers became the foundation for models like BERT, T5 and generative pre-trained transformers (GPT). == Overview == The modern era of machine attention was revitalized by grafting an attention mechanism (Fig 1. orange) to an Encoder-Decoder. Figure 2 shows the internal step-by-step operation of the attention block (A) in Fig 1. === Interpreting attention weights === In translating between languages, alignment is the process of matching words from the source sentence to words of the translated sentence. Networks that perform verbatim translation without regard to word order would show the highest scores along the (dominant) diagonal of the matrix. The off-diagonal dominance shows that the attention mechanism is more nuanced. Consider an example of translating I love you to French. On the first pass through the decoder, 94% of the attention weight is on the first English word I, so the network offers the word je. On the second pass of the decoder, 88% of the attention weight is on the third English word you, so it offers t'. On the last pass, 95% of the attention weight is on the second English word love, so it offers aime. In the I love you example, the second word love is aligned with the third word aime. Stacking soft row vectors together for je, t', and aime yields an alignment matrix: Sometimes, alignment can be multiple-to-multiple. For example, the English phrase look it up corresponds to cherchez-le. Thus, "soft" attention weights work better than "hard" attention weights (setting one attention weight to 1, and the others to 0), as we would like the model to make a context vector consisting of a weighted sum of the hidden vectors, rather than "the best one", as there may not be a best hidden vector. == Variants == Many variants of attention implement soft weights, such as fast weight programmers, or fast weight controllers (1992). A "slow" neural network outputs the "fast" weights of another neural network through outer products. The slow network learns by gradient descent. It was later renamed as "linearized self-attention". Bahdanau-style attention, also referred to as additive attention, Luong-style attention, which is known as multiplicative attention, Early attention mechanisms similar to modern self-attention were proposed using recurrent neural networks. However, the highly parallelizable self-attention was introduced in 2017 and successfully used in the Transformer model, positional attention and factorized positional attention. For convolutional neural networks, attention mechanisms can be distinguished by the dimension on which they operate, namely: spatial attention, channel attention, or combinations. These variants recombine the encoder-side inputs to redistribute those effects to each target output. Often, a correlation-style matrix of dot products provides the re-weighting coefficients. In the figures below, W is the matrix of context attention weights, similar to the formula in Overview section above. == Optimizations == === Flash attention === The size of the attention matrix is proportional to the square of the number of input tokens. Therefore, when the input is long, calculating the attention matrix requires a lot of GPU memory. Flash attention is an implementation that reduces the memory needs and increases efficiency without sacrificing accuracy. It achieves this by partitioning the attention computation into smaller blocks that fit into the GPU's faster on-chip memory, reducing the need to store large intermediate matrices and thus lowering memory usage while increasing computational efficiency. === FlexAttention === FlexAttention is an attention kernel developed by Meta that allows users to modify attention scores prior to softmax and dynamically chooses the optimal attention algorithm. == Applications == Attention is widely used in natural language processing, computer vision, and speech recognition. In NLP, it improves context understanding in tasks like question answering and summarization. In vision, visual attention helps models focus on relevant image regions, enhancing object detection and image captioning. === Attention maps as explanations for vision transformers === From the original paper on vision transformers (ViT), visualizing attention scores as a heat map (called saliency maps or attention maps) has become an important and routine way to inspect the decision making process of ViT models. One can compute the attention maps with respect to any attention head at any layer, while the deeper layers tend to show more semantically meaningful visualization. Attention rollout is a recursive algorithm to combine attention scores across all layers, by computing the dot product of successive attention maps. Because vision transformers are typically trained in a self-supervised manner, attention maps are generally not class-sensitive. When a classification head is attached to the ViT backbone, class-discriminative attention maps (CDAM) combines attention maps and gradients with respect to the class [CLS] token. Some class-sensitive interpretability methods originally developed for convolutional neural networks can be also applied to ViT, such as GradCAM, which back-propagates the gradients to the outputs of the final attention layer. Using attention as basis of explanation for the transformers in language and vision is not without debate. While some pioneering papers analyzed and framed attention scores as explanations, higher attention scores do not always correlate with greater impact on model performances. == Mathematical representation == === Standard scaled dot-product attention === For matrices: Q ∈ R m × d k , K ∈ R n × d k {\displaystyle Q\in \mathbb {R} ^{m\times d_{k}},K\in \mathbb {R} ^{n\times d_{k}}} and V ∈ R n × d v {\displaystyle V\in \mathbb {R} ^{n\times d_{v}}} , the scaled dot-product, or QKV attention, is defined as: Attention ( Q , K , V ) = softmax ( Q K T d k ) V ∈ R m × d v {\displaystyle {\text{Attention}}(Q,K,V)={\text{softmax}}\left({\frac {QK^{T}}{\sqrt {d_{k}}}}\right)V\in \mathbb {R} ^{m\times d_{v}}} where T {\displaystyle {}^{T}} denotes transpose and the softmax function is applied independently to every row of its argument. The matrix Q {\displaystyle Q} contains m {\displaystyle m} queries, while matrices K , V {\displaystyle K,V} jointly contain an unordered set of n {\displaystyle n} key-value pairs. Value vectors in matrix V {\displaystyle V} are weighted using the weights resulting from the softmax operation, so that the rows of the m {\displaystyle m} -by- d v {\displaystyle d_{v}} output matrix are confined to the convex hull of the points in R d v {\displaystyle \mathbb {R} ^{d_{v}}} given by the rows of V {\displaystyle V} . To understand the permutation invariance and permutation equivariance properties of QKV attention, let A ∈ R m × m {\displaystyle A\in \mathbb {R} ^{m\times m}} and B ∈ R n × n {\displaystyle B\in \mathbb {R} ^{n\times n}} be permutation matrices; and D ∈ R m × n {\displaystyle D\in \mathbb {R} ^{m\times n}} an arbitrary matrix. The softmax function is permutation equivariant in the sense that: softmax ( A D B ) = A softmax ( D ) B {\displays
Gollum browser
Gollum browser is a discontinued web browser for accessing Wikipedia. Since 2017, Gollum is no longer accessible online. Gollum is designed to browse Wikipedia in an easier way than directly using the web browser. Links external to Wikipedia are opened in the user's regular browser. Gollum is opened from a regular browser and makes a window that puts the Wikipedia search bar on the toolbar. Gollum was created by Harald Hanek in 2005 using PHP and Ajax. According to one blogger, Gollum provides a way to bypass censorship of Wikipedia in China. == Languages == Though the website is available only in English and German, Gollum's GUI is available in more than 32 languages and can browse nearly 50 Wikipedia editions. === Gollum's GUI === === Browsable Wikipedia editions ===
Something Big Is Happening
"Something Big Is Happening" is an essay by Matt Shumer, an AI entrepreneur, about the impact of artificial intelligence, published in February 2026, that has since been reportedly viewed more than 80 million times and widely discussed. Shumer noted that the technology has crossed an important threshold, where AI has become capable of creating self-improving systems. Referring to one the most recent AI models, he wrote: "It was making intelligent decisions. It had something that felt, for the first time, like judgment. Like taste." Speaking to CNBC's Power Lunch, Shumer said that his "core message" is "people in the workforce should start to use and experiment with AI tools so they can understand what’s coming". Even as the essay was widely shared and discussed, the essay also elicited criticism. Paulo Carvao, in an essay published by the Forbes Magazine stated that some of his advice is sound, but added: "It reads at times like a sales pitch. He urges readers to subscribe to the most advanced AI tools. He implies that those with access to premium models will outpace those without. He frames paid AI subscriptions as a form of insurance against obsolescence." Writing in The Guardian, Dan Milmo and Aisha Down mentioned Shumer as having a history of AI hype and stated, "He previously excited the internet by announcing the release of the world's "top open-source model", which it was not". Many workers in the technology sector criticized the article in blog posts shared on Hacker News; Edward Zitron commented that "while coding LLMs can test products, or scan/fix some bugs, this suggests they A) do this autonomously without human input, B) they do this correctly every time (or ever!)." In an article alluding to Shumer's original post, Ari Colaprete wrote "the LLM is fundamentally a writing machine, it does everything via text, and if you make it produce writing that exists purely to serve some sort of mechanical function, and you train it to succeed in that task, then it will tend to do so, even with vast intricacy."