AI Art Detection

AI Art Detection — independent reviews, comparisons, pricing and step-by-step guides on Aizhi.

• 

  Sanchar Saathi

  Sanchar Saathi (lit. 'Communication Partner' or 'Communication Companion') is an Indian state-owned app and web portal, operated by the Department of Telecommunications, designed to assist Indian mobile users in tracking and blocking stolen or lost mobile devices. In late 2025, a government order requiring Sanchar Saathi to be pre-installed on all mobile devices sold nationwide, with explicit provisions on preventing users from deleting the app or disabling any of its broad functionalities, triggered widespread backlash. The order was subsequently withdrawn. == Background == The Telecommunications Act 2023 introduced an exceptionally broad definition of the term "telecommunications" and conferred wide-ranging powers on the government. Although the Department of Telecommunications (DoT) assured reporters that this definition would not be used to justify government overreach, a November 2024 amendment to the Telecom Cyber Security Rules expanded it further and introduced the concept of the Telecommunication Identifier User Entity (TIEU), enabling users to be personally identified through their phone numbers. Sanchar Saathi was launched amid a widespread rise in cybercrime and hacking, as part of the Indian government's effort to prevent stolen phones from being used for fraud and to promote a state-backed application. In an official statement, the DoT said, "India has big second-hand mobile device market. Cases have also been observed where stolen or blacklisted devices are being re-sold. It makes the purchaser abettor in crime and causes financial loss to them." == Launch == Sanchar Saathi was originally launched as a web portal in May 2023. It was later launched as a mobile app in January 2025. Describing itself as a "citizen-centric" safety tool, Sanchar Saathi allows users to check a device's IMEI, report and block lost or stolen phones, and flag suspected fraud communications. Under Sanchar Saathi's privacy policy, it can make and manage phone calls, view and send messages, read call logs, access photos and files, access the location and camera of the device in which the app is used, as well as read and write into the device's storage. According to official government data, by December 2025, the Sanchar Saathi app had helped recover more than 700,000 lost and stolen mobile devices across India. Users report around 2,000 fraud incidents through the app each day. == Pre-installation controversy == On 28 November 2025, the Bharatiya Janata Party government, led by prime minister Narendra Modi, privately ordered phone manufacturers, including Apple, Samsung, Xiaomi, Vivo, Oppo, among others, to pre-install the Sanchar Saathi app on new devices sold in the country, alongside mandating that old devices get issued a software update for the installation of the app. The order had a 90-day deadline and further included explicit provisions to ensure that the app is to be "readily visible and accessible to the end users at the time of first use or device setup" and that users should neither be able to delete the app nor disable or restrict any of its broad functionalities. The order caused widespread political backlash. K. C. Venugopal, a general secretary of the main opposition party, the Indian National Congress (or simply the Congress), called the order "beyond unconstitutional" and said, "A pre-loaded government app that cannot be uninstalled is a dystopian tool to monitor every Indian. It is a means to watch over every movement, interaction and decision of each citizen", adding, "Big Brother cannot watch us." Another Congress general secretary, Priyanka Gandhi, termed Sanchar Saathi a "snooping app", and attacked the government for "turning this country into a dictatorship". Uddhav Thackeray, former chief minister of Maharashtra, compared Sanchar Saathi to the Pegasus spyware. Sanjay Hegde, a senior advocate at the Supreme Court of India, said "Here in the garb of security, the intrusion is vast, unfettered, unguided and is totally disproportionate. The app ought to be struck down on that account". The Internet Freedom Foundation (IFF), an Indian digital rights advocacy organisation, said, "Forcing every smartphone to carry a permanent government app for a simple verification task is excessive and violates the Puttaswamy proportionality standard", referring to Puttaswamy v. Union of India, a 2017 landmark decision of the Supreme Court, which asserted that the right to privacy should be protected as a fundamental right. The IFF further said, "For this to work in practice, the app will almost certainly need system level or root level access, similar to carrier or OEM system apps, so that it cannot be disabled. That design choice erodes the protections that normally prevent one app from peering into the data of others, and turns Sanchar Saathi into a permanent, non-consensual point of access sitting inside the operating system of every Indian smartphone user." Moreover, the organisation said that while the app was being "framed as a benign IMEI checker", a server-side update could allow the app to engage in "client side scanning for 'banned' applications, flag VPN usage, correlate SIM activity, or trawl SMS logs in the name of fraud detection. Nothing in the order constrains these possibilities." In reaction to the controversy, Jyotiraditya Scindia, the union minister of communications, said, "There is no snooping or call monitoring", adding, "Obviously you can delete it. There is no problem. This is a matter of customer protection. It is not mandatory. If you don't want to register, and don't want to use the app, don't use it; don't register, and it will lay dormant." Scindia compared the app to other pre-installed mobile apps such as Google Maps, which he said could be deleted if users wished so. However, contrary to Scindia's statement, on many phone brands, such pre-installed apps cannot be deleted, although users can disable them. Furthermore, upon enquiry, Scindia did not clarify whether his remarks applied to the app after the order took effect, making no comment on the provision in the order that would prevent users from deleting the app. When Congress member Renuka Chowdhury submitted an adjournment motion notice in the Rajya Sabha seeking the suspension of all other matters to discuss the Sanchar Saathi issue, Kiren Rijiju, the union minister of parliamentary affairs, accused the opposition of "manufacturing issues" to stall session proceedings. By 2 December, it had been reported that Apple did not plan to comply with the order, citing privacy and security concerns for the iOS ecosystem and the fact that the order would violate its internal policy against the pre-installation of third-party software in iPhones. Although it was clarified that Apple did not intend to take the matter to court or publicly oppose the government, it was said that Apple "can't do this. Period." The order would have also required Google to create a custom version of Android solely for India which would include the Sanchar Saathi app, a requirement described to "not be acceptable to the company". Following the backlash, the order was revoked on 3 December 2025. In a press release, the government said, "Given Sanchar Saathi's increasing acceptance, Government has decided not to make the pre-installation mandatory for mobile manufacturers".

  Read more →
• 

  Shape context

  Shape context is a feature descriptor used in object recognition. Serge Belongie and Jitendra Malik proposed the term in their paper "Matching with Shape Contexts" in 2000. == Theory == The shape context is intended to be a way of describing shapes that allows for measuring shape similarity and the recovering of point correspondences. The basic idea is to pick n points on the contours of a shape. For each point pi on the shape, consider the n − 1 vectors obtained by connecting pi to all other points. The set of all these vectors is a rich description of the shape localized at that point but is far too detailed. The key idea is that the distribution over relative positions is a robust, compact, and highly discriminative descriptor. So, for the point pi, the coarse histogram of the relative coordinates of the remaining n − 1 points, h i ( k ) = # { q ≠ p i : ( q − p i ) ∈ bin ( k ) } {\displaystyle h_{i}(k)=\#\{q\neq p_{i}:(q-p_{i})\in {\mbox{bin}}(k)\}} is defined to be the shape context of p i {\displaystyle p_{i}} . The bins are normally taken to be uniform in log-polar space. The fact that the shape context is a rich and discriminative descriptor can be seen in the figure below, in which the shape contexts of two different versions of the letter "A" are shown. (a) and (b) are the sampled edge points of the two shapes. (c) is the diagram of the log-polar bins used to compute the shape context. (d) is the shape context for the point marked with a circle in (a), (e) is that for the point marked as a diamond in (b), and (f) is that for the triangle. As can be seen, since (d) and (e) are the shape contexts for two closely related points, they are quite similar, while the shape context in (f) is very different. For a feature descriptor to be useful, it needs to have certain invariances. In particular it needs to be invariant to translation, scaling, small perturbations, and, depending on the application, rotation. Translational invariance comes naturally to shape context. Scale invariance is obtained by normalizing all radial distances by the mean distance α {\displaystyle \alpha } between all the point pairs in the shape although the median distance can also be used. Shape contexts are empirically demonstrated to be robust to deformations, noise, and outliers using synthetic point set matching experiments. One can provide complete rotational invariance in shape contexts. One way is to measure angles at each point relative to the direction of the tangent at that point (since the points are chosen on edges). This results in a completely rotationally invariant descriptor. But of course this is not always desired since some local features lose their discriminative power if not measured relative to the same frame. Many applications in fact forbid rotational invariance e.g. distinguishing a "6" from a "9". == Use in shape matching == A complete system that uses shape contexts for shape matching consists of the following steps (which will be covered in more detail in the Details of Implementation section): Randomly select a set of points that lie on the edges of a known shape and another set of points on an unknown shape. Compute the shape context of each point found in step 1. Match each point from the known shape to a point on an unknown shape. To minimize the cost of matching, first choose a transformation (e.g. affine, thin plate spline, etc.) that warps the edges of the known shape to the unknown (essentially aligning the two shapes). Then select the point on the unknown shape that most closely corresponds to each warped point on the known shape. Calculate the "shape distance" between each pair of points on the two shapes. Use a weighted sum of the shape context distance, the image appearance distance, and the bending energy (a measure of how much transformation is required to bring the two shapes into alignment). To identify the unknown shape, use a nearest-neighbor classifier to compare its shape distance to shape distances of known objects. == Details of implementation == === Step 1: Finding a list of points on shape edges === The approach assumes that the shape of an object is essentially captured by a finite subset of the points on the internal or external contours on the object. These can be simply obtained using the Canny edge detector and picking a random set of points from the edges. Note that these points need not and in general do not correspond to key-points such as maxima of curvature or inflection points. It is preferable to sample the shape with roughly uniform spacing, though it is not critical. === Step 2: Computing the shape context === This step is described in detail in the Theory section. === Step 3: Computing the cost matrix === Consider two points p and q that have normalized K-bin histograms (i.e. shape contexts) g(k) and h(k). As shape contexts are distributions represented as histograms, it is natural to use the χ2 test statistic as the "shape context cost" of matching the two points: C S = 1 2 ∑ k = 1 K [ g ( k ) − h ( k ) ] 2 g ( k ) + h ( k ) {\displaystyle C_{S}={\frac {1}{2}}\sum _{k=1}^{K}{\frac {[g(k)-h(k)]^{2}}{g(k)+h(k)}}} The values of this range from 0 to 1. In addition to the shape context cost, an extra cost based on the appearance can be added. For instance, it could be a measure of tangent angle dissimilarity (particularly useful in digit recognition): C A = 1 2 ‖ ( cos ⁡ ( θ 1 ) sin ⁡ ( θ 1 ) ) − ( cos ⁡ ( θ 2 ) sin ⁡ ( θ 2 ) ) ‖ {\displaystyle C_{A}={\frac {1}{2}}{\begin{Vmatrix}{\dbinom {\cos(\theta _{1})}{\sin(\theta _{1})}}-{\dbinom {\cos(\theta _{2})}{\sin(\theta _{2})}}\end{Vmatrix}}} This is half the length of the chord in unit circle between the unit vectors with angles θ 1 {\displaystyle \theta _{1}} and θ 2 {\displaystyle \theta _{2}} . Its values also range from 0 to 1. Now the total cost of matching the two points could be a weighted-sum of the two costs: C = ( 1 − β ) C S + β C A {\displaystyle C=(1-\beta )C_{S}+\beta C_{A}\!\,} Now for each point pi on the first shape and a point qj on the second shape, calculate the cost as described and call it Ci,j. This is the cost matrix. === Step 4: Finding the matching that minimizes total cost === Now, a one-to-one matching π ( i ) {\displaystyle \pi (i)} that matches each point pi on shape 1 and qj on shape 2 that minimizes the total cost of matching, H ( π ) = ∑ i C ( p i , q π ( i ) ) {\displaystyle H(\pi )=\sum _{i}C\left(p_{i},q_{\pi (i)}\right)} is needed. This can be done in O ( N 3 ) {\displaystyle O(N^{3})} time using the Hungarian method, although there are more efficient algorithms. To have robust handling of outliers, one can add "dummy" nodes that have a constant but reasonably large cost of matching to the cost matrix. This would cause the matching algorithm to match outliers to a "dummy" if there is no real match. === Step 5: Modeling transformation === Given the set of correspondences between a finite set of points on the two shapes, a transformation T : R 2 → R 2 {\displaystyle T:\mathbb {R} ^{2}\to \mathbb {R} ^{2}} can be estimated to map any point from one shape to the other. There are several choices for this transformation, described below. ==== Affine ==== The affine model is a standard choice: T ( p ) = A p + o {\displaystyle T(p)=Ap+o\!} . The least squares solution for the matrix A {\displaystyle A} and the translational offset vector o is obtained by: o = 1 n ∑ i = 1 n ( p i − q π ( i ) ) , A = ( Q + P ) t {\displaystyle o={\frac {1}{n}}\sum _{i=1}^{n}\left(p_{i}-q_{\pi (i)}\right),A=(Q^{+}P)^{t}} Where P = ( 1 p 11 p 12 ⋮ ⋮ ⋮ 1 p n 1 p n 2 ) {\displaystyle P={\begin{pmatrix}1&p_{11}&p_{12}\\\vdots &\vdots &\vdots \\1&p_{n1}&p_{n2}\end{pmatrix}}} with a similar expression for Q {\displaystyle Q\!} . Q + {\displaystyle Q^{+}\!} is the pseudoinverse of Q {\displaystyle Q\!} . ==== Thin plate spline ==== The thin plate spline (TPS) model is the most widely used model for transformations when working with shape contexts. A 2D transformation can be separated into two TPS function to model a coordinate transform: T ( x , y ) = ( f x ( x , y ) , f y ( x , y ) ) {\displaystyle T(x,y)=\left(f_{x}(x,y),f_{y}(x,y)\right)} where each of the ƒx and ƒy have the form: f ( x , y ) = a 1 + a x x + a y y + ∑ i = 1 n ω i U ( ‖ ( x i , y i ) − ( x , y ) ‖ ) , {\displaystyle f(x,y)=a_{1}+a_{x}x+a_{y}y+\sum _{i=1}^{n}\omega _{i}U\left({\begin{Vmatrix}(x_{i},y_{i})-(x,y)\end{Vmatrix}}\right),} and the kernel function U ( r ) {\displaystyle U(r)\!} is defined by U ( r ) = r 2 log ⁡ r 2 {\displaystyle U(r)=r^{2}\log r^{2}\!} . The exact details of how to solve for the parameters can be found elsewhere but it essentially involves solving a linear system of equations. The bending energy (a measure of how much transformation is needed to align the points) will also be easily obtained. ==== Regularized TPS ==== The TPS formulation above has exact matching requirement for the pairs of points on the two shapes. For noisy data, it is best to

  Read more →
• 

  Connected-component labeling

  Connected-component labeling (CCL), connected-component analysis (CCA), blob extraction, region labeling, blob discovery, or region extraction is an algorithmic application of graph theory, where subsets of connected components are uniquely labeled based on a given heuristic. Connected-component labeling is not to be confused with segmentation. Connected-component labeling is used in computer vision to detect connected regions in binary digital images, although color images and data with higher dimensionality can also be processed. When integrated into an image recognition system or human-computer interaction interface, connected component labeling can operate on a variety of information. Blob extraction is generally performed on the resulting binary image from a thresholding step, but it can be applicable to gray-scale and color images as well. Blobs may be counted, filtered, and tracked. Blob extraction is related to but distinct from blob detection. == Overview == A graph, containing vertices and connecting edges, is constructed from relevant input data. The vertices contain information required by the comparison heuristic, while the edges indicate connected 'neighbors'. An algorithm traverses the graph, labeling the vertices based on the connectivity and relative values of their neighbors. Connectivity is determined by the medium; image graphs, for example, can be 4-connected neighborhood or 8-connected neighborhood. Following the labeling stage, the graph may be partitioned into subsets, after which the original information can be recovered and processed . == Definition == The usage of the term connected-component labeling (CCL) and its definition is quite consistent in the academic literature, whereas connected-component analysis (CCA) varies both in terminology and in its definition of the problem. Rosenfeld et al. define connected components labeling as the “[c]reation of a labeled image in which the positions associated with the same connected component of the binary input image have a unique label.” Shapiro et al. define CCL as an operator whose “input is a binary image and [...] output is a symbolic image in which the label assigned to each pixel is an integer uniquely identifying the connected component to which that pixel belongs.” There is no consensus on the definition of CCA in the academic literature. It is often used interchangeably with CCL. A more extensive definition is given by Shapiro et al.: “Connected component analysis consists of connected component labeling of the black pixels followed by property measurement of the component regions and decision making.” The definition for connected-component analysis presented here is more general, taking the thoughts expressed in into account. == Algorithms == The algorithms discussed can be generalised to arbitrary dimensions, albeit with increased time and space complexity. === One component at a time === This is a fast and very simple method to implement and understand. It is based on graph traversal methods in graph theory. In short, once the first pixel of a connected component is found, all the connected pixels of that connected component are labelled before going onto the next pixel in the image. This algorithm is part of Vincent and Soille's watershed segmentation algorithm, other implementations also exist. In order to do that a linked list is formed that will keep the indexes of the pixels that are connected to each other, steps (2) and (3) below. The method of defining the linked list specifies the use of a depth or a breadth first search. For this particular application, there is no difference which strategy to use. The simplest kind of a last in first out queue implemented as a singly linked list will result in a depth first search strategy. It is assumed that the input image is a binary image, with pixels being either background or foreground and that the connected components in the foreground pixels are desired. The algorithm steps can be written as: Start from the first pixel in the image. Set current label to 1. Go to (2). If this pixel is a foreground pixel and it is not already labelled, give it the current label and add it as the first element in a queue, then go to (3). If it is a background pixel or it was already labelled, then repeat (2) for the next pixel in the image. Pop out an element from the queue, and look at its neighbours (based on any type of connectivity). If a neighbour is a foreground pixel and is not already labelled, give it the current label and add it to the queue. Repeat (3) until there are no more elements in the queue. Go to (2) for the next pixel in the image and increment current label by 1. Note that the pixels are labelled before being put into the queue. The queue will only keep a pixel to check its neighbours and add them to the queue if necessary. This algorithm only needs to check the neighbours of each foreground pixel once and doesn't check the neighbours of background pixels. The pseudocode is: algorithm OneComponentAtATime(data) input : imageData[xDim][yDim] initialization : label = 0, labelArray[xDim][yDim] = 0, statusArray[xDim][yDim] = false, queue1, queue2; for i = 0 to xDim do for j = 0 to yDim do if imageData[i][j] has not been processed do if imageData[i][j] is a foreground pixel do check its four neighbors(north, south, east, west) : if neighbor is not processed do if neighbor is a foreground pixel do add it to queue1 else update its status to processed end if labelArray[i][j] = label (give label) statusArray[i][j] = true (update status) while queue1 is not empty do For each pixel in the queue do : check its four neighbors if neighbor is not processed do if neighbor is a foreground pixel do add it to queue2 else update its status to processed end if give it the current label update its status to processed remove the current element from queue1 copy queue2 into queue1 end While increase the label end if else update its status to processed end if end if end if end for end for === Two-pass === Relatively simple to implement and understand, the two-pass algorithm, (also known as the Hoshen–Kopelman algorithm) iterates through 2-dimensional binary data. The algorithm makes two passes over the image: the first pass to assign temporary labels and record equivalences, and the second pass to replace each temporary label by the smallest label of its equivalence class. The input data can be modified in situ (which carries the risk of data corruption), or labeling information can be maintained in an additional data structure. Connectivity checks are carried out by checking neighbor pixels' labels (neighbor elements whose labels are not assigned yet are ignored), or say, the north-east, the north, the north-west and the west of the current pixel (assuming 8-connectivity). 4-connectivity uses only north and west neighbors of the current pixel. The following conditions are checked to determine the value of the label to be assigned to the current pixel (4-connectivity is assumed) Conditions to check: Does the pixel to the left (west) have the same value as the current pixel? Yes – We are in the same region. Assign the same label to the current pixel No – Check next condition Do both pixels to the north and west of the current pixel have the same value as the current pixel but not the same label? Yes – We know that the north and west pixels belong to the same region and must be merged. Assign the current pixel the minimum of the north and west labels, and record their equivalence relationship No – Check next condition Does the pixel to the left (west) have a different value and the one to the north the same value as the current pixel? Yes – Assign the label of the north pixel to the current pixel No – Check next condition Do the pixel's north and west neighbors have different pixel values than current pixel? Yes – Create a new label id and assign it to the current pixel The algorithm continues this way, and creates new region labels whenever necessary. The key to a fast algorithm, however, is how this merging is done. This algorithm uses the union-find data structure which provides excellent performance for keeping track of equivalence relationships. Union-find essentially stores labels which correspond to the same blob in a disjoint-set data structure, making it easy to remember the equivalence of two labels by the use of an interface method E.g.: findSet(l). findSet(l) returns the minimum label value that is equivalent to the function argument 'l'. Once the initial labeling and equivalence recording is completed, the second pass merely replaces each pixel label with its equivalent disjoint-set representative element. A faster-scanning algorithm for connected-region extraction is presented below. On the first pass: Iterate through each element of the data by column, then by row (Raster Scanning) If the element is not the background Get the neighboring elements of the current element If there are no neighbors, uniquely

  Read more →
• 

  Apertus (LLM)

  Apertus is a public large language model, developed by the Swiss AI Initiative (a collaboration between EPFL, ETH Zurich, and the Swiss National Supercomputing Centre). It was released on September 2, 2025, under the free and open-source Apache 2.0 license. Designed initially for business and research use cases around the world, Apertus was trained on over 1800 languages, and comes in 8 billion or 70 billion parameter versions and is available on Hugging Face for download. The model was developed aiming to adhere to European copyright law, and is one of the first examples of AI as a public good in the vein of AI Sovereignty. It is also the first large model to comply with the European Union's Artificial Intelligence Act. At its launch, the model creators emphasized multilinguality, transparency, and auditability as priorities in contrast to commercial frontier model. While international reception was largely positive, the first iteration was significantly behind the capabilities of frontier models and needs adaptation for many use cases with chatbots being a secondary but not a primary use case. As of late 2025, it was considered the largest and most capable fully open model. The capability of future models will depend in part on how much more funding can be secured.

  Read more →
• 

  Absher (application)

  Absher (Arabic: أبشر ‘Absher, roughly meaning "good tidings" or "yes, done") is a smartphone application and web portal which allows citizens and residents of Saudi Arabia to use a variety of governmental services. Amongst several other services with the Absher app, it can be used to apply for jobs and Hajj permits, passport info can be updated, and electronic crimes can be reported. The application provides around 280 services for residents of Saudi Arabia including but not limited to making appointments, renewing passports, residents' cards, IDs, driver's licenses and others, and, controversially, enables Saudi men to track the whereabouts of women they control as part of the country's male guardianship system. The app can be downloaded from the Google Play Store and Apple App Store and is supplied by the Saudi Interior Ministry. According to the Ministry of the Interior, Absher has more than 20 million users. As of February 2019, Absher has been downloaded 4.2 million times from the App Store. Some services provided through Absher can also be accessed through the website absher.sa. In March 2021, Saudi Arabia launched the digital version of the Absher for individuals app through which the users can download a copy of their digital ID. Then, new services were added to the platform such as online birth and death registration services, requesting amendments to academic credentials, correcting names in English and marital status and requesting civil records of children. == Impact on women's rights == The app has been criticized by various human rights activists, human rights organisations and international communities. The US and European countries have also condemned the app and urged the kingdom to end its male guardianship system. Absher gained media attention in 2019 for its functions supporting the Saudi policy of male guardianship following an investigation by Business Insider. The app allows for designated guardians to receive notifications if a woman under their guardianship passes through an airport and subsequently gives them the option to withdraw her right to travel. In a few cases, this system has been circumvented by women who have been able to gain control over its settings and use it to allow themselves to travel. US Senator Ron Wyden of Oregon wrote a letter to the CEO's of Apple and Google, criticizing the app and demanding for its removal immediately. Wyden said "American companies should not enable or facilitate the Saudi government's patriarchy," and called the Saudi system of control over women "abhorrent". According to the EU lawmakers, current rules imposed over the women by the Saudi government make women “second-class citizens”. The lawmakers also asked the EU states to continue to build pressure on Riyadh so as to improve the conditions of women and human rights. Amnesty International and Human Rights Watch accused Apple and Google of helping "enforce gender apartheid" by hosting the app. US congresswomen Rep. Katherine Clark and Rep. Carolyn B. Maloney condemned the kingdom's male guardianship system that reflected from the app, calling Absher a "patriarchal weapon" and asking for its removal. In response to the criticism received by Absher, Apple chief executive officer Tim Cook stated in February 2019 that he intended to investigate the situation. Similarly, Google announced that it would also review the application. After a prompt review, Google declined to remove the app from Google Play, citing that it did not violate the agreed upon terms and conditions of the store. Saudi doctor Khawla Al-Kuraya supported this app an editorial in Bloomberg News. Kuraya wrote that Absher helped Saudi women avoid governmental bureaucracy as it allows their male guardians to process their travel permits anywhere and anytime through Absher. Although she believes that the guardianship system needs to be reconsidered, she thinks that Absher is an important step towards facilitating women-guardians related issues in Saudi Arabia. Absher manager Atiyah Al-Anazy announced in 2019 that two million women were using the application in Saudi Arabia to facilitate their transactions. Some female users stated that the application has made their movement and travel-related issues easier. New measures were introduced that year to allow Saudi women above the age of 18 to travel without their male guardians, which ultimately released male authoritative rights on women. A law was subsequently passed allowing women over the age of 21 to receive a passport and travel without prior male permission.

  Read more →
• 

  Vision transformer

  A vision transformer (ViT) is a transformer designed for computer vision. A ViT decomposes an input image into a series of patches (rather than text into tokens), serializes each patch into a vector, and maps it to a smaller dimension with a single matrix multiplication. These vector embeddings are then processed by a transformer encoder as if they were token embeddings. ViTs were designed as alternatives to convolutional neural networks (CNNs) in computer vision applications. They have different inductive biases, training stability, and data efficiency. Compared to CNNs, ViTs are less data efficient, but have higher capacity. Some of the largest modern computer vision models are ViTs, such as one with 22B parameters. Subsequent to its publication, many variants were proposed, with hybrid architectures with both features of ViTs and CNNs. ViTs have found application in image recognition, image segmentation, weather prediction, and autonomous driving. == History == Transformers were introduced in Attention Is All You Need (2017), and have found widespread use in natural language processing. A 2019 paper applied ideas from the Transformer to computer vision. Specifically, they started with a ResNet, a standard convolutional neural network used for computer vision, and replaced all convolutional kernels by the self-attention mechanism found in a Transformer. It resulted in superior performance. However, it is not a Vision Transformer. In 2020, an encoder-only Transformer was adapted for computer vision, yielding the ViT, which reached state of the art in image classification, overcoming the previous dominance of CNN. The masked autoencoder (2022) extended ViT to work with unsupervised training. The vision transformer and the masked autoencoder, in turn, stimulated new developments in convolutional neural networks. Subsequently, there was cross-fertilization between the previous CNN approach and the ViT approach. In 2021, some important variants of the Vision Transformers were proposed. These variants are mainly intended to be more efficient, more accurate or better suited to a specific domain. Two studies improved efficiency and robustness of ViT by adding a CNN as a preprocessor. The Swin Transformer achieved state-of-the-art results on some object detection datasets such as COCO, by using convolution-like sliding windows of attention mechanism, and the pyramid process in classical computer vision. == Overview == The basic architecture, used by the original 2020 paper, is as follows. In summary, it is a BERT-like encoder-only Transformer. The input image is of type R H × W × C {\displaystyle \mathbb {R} ^{H\times W\times C}} , where H , W , C {\displaystyle H,W,C} are height, width, channel (RGB). It is then split into square-shaped patches of type R P × P × C {\displaystyle \mathbb {R} ^{P\times P\times C}} . For each patch, the patch is pushed through a linear operator, to obtain a vector ("patch embedding"). The position of the patch is also transformed into a vector by "position encoding" (the paper tried no embedding, 1D embedding, 2D embedding, and relative embedding: 1D was adopted). The two vectors are added, then pushed through several Transformer encoders. The attention mechanism in a ViT repeatedly transforms representation vectors of image patches, incorporating more and more semantic relations between image patches in an image. This is analogous to how in natural language processing, as representation vectors flow through a transformer, they incorporate more and more semantic relations between words, from syntax to semantics. The above architecture turns an image into a sequence of vector representations. To use these for downstream applications, an additional head needs to be trained to interpret them. For example, to use it for classification, one can add a shallow MLP on top of it that outputs a probability distribution over classes. The original paper uses a linear-GeLU-linear-softmax network. == Variants == === Original ViT === The original ViT was an encoder-only Transformer supervise-trained to predict the image label from the patches of the image. As in the case of BERT, it uses a special token in the input side, and the corresponding output vector is used as the only input of the final output MLP head. The special token is an architectural hack to allow the model to compress all information relevant for predicting the image label into one vector. Transformers found their initial applications in natural language processing tasks, as demonstrated by language models such as BERT and GPT-3. By contrast the typical image processing system uses a convolutional neural network (CNN). Well-known projects include Xception, ResNet, EfficientNet, DenseNet, and Inception. Transformers measure the relationships between pairs of input tokens (words in the case of text strings), termed attention. The cost is quadratic in the number of tokens. For images, the basic unit of analysis is the pixel. However, computing relationships for every pixel pair in a typical image is prohibitive in terms of memory and computation. Instead, ViT computes relationships among pixels in various small sections of the image (e.g., 16x16 pixels), at a drastically reduced cost. The sections (with positional embeddings) are placed in a sequence. The embeddings are learnable vectors. Each section is arranged into a linear sequence and multiplied by the embedding matrix. The result, with the position embedding is fed to the transformer. === Architectural improvements === ==== Pooling ==== After the ViT processes an image, it produces some embedding vectors. These must be converted to a single class probability prediction by some kind of network. In the original ViT and Masked Autoencoder, they used a dummy [CLS] token, in emulation of the BERT language model. The output at [CLS] is the classification token, which is then processed by a LayerNorm-feedforward-softmax module into a probability distribution. Global average pooling (GAP) does not use the dummy token, but simply takes the average of all output tokens as the classification token. It was mentioned in the original ViT as being equally good. Multihead attention pooling (MAP) applies a multiheaded attention block to pooling. Specifically, it takes as input a list of vectors x 1 , x 2 , … , x n {\displaystyle x_{1},x_{2},\dots ,x_{n}} , which might be thought of as the output vectors of a layer of a ViT. The output from MAP is M u l t i h e a d e d A t t e n t i o n ( Q , V , V ) {\displaystyle \mathrm {MultiheadedAttention} (Q,V,V)} , where q {\displaystyle q} is a trainable query vector, and V {\displaystyle V} is the matrix with rows being x 1 , x 2 , … , x n {\displaystyle x_{1},x_{2},\dots ,x_{n}} . This was first proposed in the Set Transformer architecture. Later papers demonstrated that GAP and MAP both perform better than BERT-like pooling. A variant of MAP was proposed as class attention, which applies MAP, then feedforward, then MAP again. Re-attention was proposed to allow training deep ViT. It changes the multiheaded attention module. === Masked Autoencoder === The Masked Autoencoder took inspiration from denoising autoencoders and context encoders. It has two ViTs put end-to-end. The first one ("encoder") takes in image patches with positional encoding, and outputs vectors representing each patch. The second one (called "decoder", even though it is still an encoder-only Transformer) takes in vectors with positional encoding and outputs image patches again. ==== Training ==== During training, input images (224px x 224 px in the original implementation) are split along a designated number of lines on each axis, producing image patches. A certain percentage of patches are selected to be masked out by mask tokens, while all others are retained in the image. The network is tasked with reconstructing the image from the remaining unmasked patches. Mask tokens in the original implementation are learnable vector quantities. A linear projection with positional embeddings is then applied to the vector of unmasked patches. Experiments varying mask ratio on networks trained on the ImageNet-1K dataset found 75% mask ratios achieved high performance on both finetuning and linear-probing of the encoder's latent space. The MAE processes only unmasked patches during training, increasing the efficiency of data processing in the encoder and lowering the memory usage of the transformer. A less computationally-intensive ViT is used for the decoder in the original implementation of the MAE. Masked patches are added back to the output of the encoder block as mask tokens and both are fed into the decoder. A reconstruction loss is computed for the masked patches to assess network performance. ==== Prediction ==== In prediction, the decoder architecture is discarded entirely. The input image is split into patches by the same algorithm as in training, but no patches are masked out. A linear projection wi

  Read more →
• 

  Kleene star

  In formal language theory, the Kleene star (or Kleene operator or Kleene closure) refers to two related unary operations, that can be applied either to an alphabet of symbols or to a formal language, a set of strings (finite sequences of symbols). The Kleene star operator on an alphabet V generates the set V of all finite-length strings over V, that is, finite sequences whose elements belong to V; in mathematics, it is more commonly known as the free monoid construction. The Kleene star operator on a language L generates another language L, the set of all strings that can be obtained as a concatenation of zero or more members of L. In both cases, repetitions are allowed. The Kleene star operators are named after American mathematician Stephen Cole Kleene, who first introduced and widely used it to characterize automata for regular expressions. == Of an alphabet == Given an alphabet V {\displaystyle V} , define V 0 = { ε } {\displaystyle V^{0}=\{\varepsilon \}} (the set consists only of the empty string), V 1 = V , {\displaystyle V^{1}=V,} and define recursively the set V i + 1 = { w v : w ∈ V i and v ∈ V } {\displaystyle V^{i+1}=\{wv:w\in V^{i}{\text{ and }}v\in V\}} for each i > 0 , {\displaystyle i>0,} where w v {\displaystyle wv} denotes the string obtained by appending the single character v {\displaystyle v} to the end of w {\displaystyle w} . Here, V i {\displaystyle V^{i}} can be understood to be the set of all strings of length exactly i {\displaystyle i} , with characters from V {\displaystyle V} . The definition of Kleene star on V {\displaystyle V} is V ∗ = ⋃ i ≥ 0 V i = V 0 ∪ V 1 ∪ V 2 ∪ V 3 ∪ V 4 ∪ ⋯ . {\displaystyle V^{}=\bigcup _{i\geq 0}V^{i}=V^{0}\cup V^{1}\cup V^{2}\cup V^{3}\cup V^{4}\cup \cdots .} == Of a language == Given a language L {\displaystyle L} (any finite or infinite set of strings), define L 0 = { ε } {\displaystyle L^{0}=\{\varepsilon \}} (the language consisting only of the empty string), L 1 = L , {\displaystyle L^{1}=L,} and define recursively the set L i + 1 = { w v : w ∈ L i and v ∈ L } {\displaystyle L^{i+1}=\{wv:w\in L^{i}{\text{ and }}v\in L\}} for each i > 0 , {\displaystyle i>0,} where w v {\displaystyle wv} denotes the string obtained by concatenating w {\displaystyle w} and v {\displaystyle v} . Here, L i {\displaystyle L^{i}} can be understood to be the set of all strings that can be obtained by concatenating exactly i {\displaystyle i} strings from L {\displaystyle L} , allowing repetitions. The definition of Kleene star on L {\displaystyle L} is L ∗ = ⋃ i ≥ 0 L i = L 0 ∪ L 1 ∪ L 2 ∪ L 3 ∪ L 4 ∪ ⋯ . {\displaystyle L^{}=\bigcup _{i\geq 0}L^{i}=L^{0}\cup L^{1}\cup L^{2}\cup L^{3}\cup L^{4}\cup \cdots .} == Kleene plus == In some formal language studies, (e.g. AFL theory) a variation on the Kleene star operation called the Kleene plus is used. The Kleene plus omits the V 0 {\displaystyle V^{0}} or L 0 {\displaystyle L^{0}} term in the above unions. In other words, the Kleene plus on V {\displaystyle V} is V + = ⋃ i ≥ 1 V i = V 1 ∪ V 2 ∪ V 3 ∪ ⋯ , {\displaystyle V^{+}=\bigcup _{i\geq 1}V^{i}=V^{1}\cup V^{2}\cup V^{3}\cup \cdots ,} or V + = V ∗ V . {\displaystyle V^{+}=V^{}V.} == Examples == Example of Kleene star applied to a set of strings: {"ab","c"} = { ε, "ab", "c", "abab", "abc", "cab", "cc", "ababab", "ababc", "abcab", "abcc", "cabab", "cabc", "ccab", "ccc", ...}. Example of Kleene star applied to a set of strings without the prefix property: {"a","ab","b"} = { ε, "a", "ab", "b", "aa", "aab", "aba", "abab", "abb", "ba", "bab", "bb", ...};In this example, the string "aab" can be obtained in two different ways. The Sardinas-Patterson algorithm can be used to check for a given V whether any member of V can be obtained in more than one way. Example of Kleene and Kleene plus applied to a set of characters (following the C programming language convention where a character is denoted by single quotes and a string is denoted by double quotes): {'a', 'b', 'c'} = { ε, "a", "b", "c", "aa", "ab", "ac", "ba", "bb", "bc", "ca", "cb", "cc", "aaa", "aab", ...}. {'a', 'b', 'c'}+ = { "a", "b", "c", "aa", "ab", "ac", "ba", "bb", "bc", "ca", "cb", "cc", "aaa", "aab", ...}. == Properties == If V {\displaystyle V} is any finite or countably infinite set of characters, then V ∗ {\displaystyle V^{}} is a countably infinite set. As a result, each formal language over a finite or countably infinite alphabet Σ {\displaystyle \Sigma } is countable, since it is a subset of the countably infinite set Σ ∗ {\displaystyle \Sigma ^{}} . ( L ∗ ) ∗ = L ∗ {\displaystyle (L^{})^{}=L^{}} , which means that the Kleene star operator is an idempotent unary operator, as ( L ∗ ) i = L ∗ {\displaystyle (L^{})^{i}=L^{}} for every i ≥ 1 {\displaystyle i\geq 1} . V ∗ = { ε } {\displaystyle V^{}=\{\varepsilon \}} , if V {\displaystyle V} is the empty set ∅. For the version of the Kleene star operator on languages, L ∗ = { ε } {\displaystyle L^{}=\{\varepsilon \}} when L {\displaystyle L} is either the empty set ∅ or the singleton set { ε } {\displaystyle \{\varepsilon \}} . == Generalization == Strings form a monoid with concatenation as the binary operation and ε the identity element. In addition to strings, the Kleene star is defined for any monoid. More precisely, let (M, ⋅) be a monoid, and S ⊆ M. Then S is the smallest submonoid of M containing S; that is, S contains the neutral element of M, the set S, and is such that if x,y ∈ S, then x⋅y ∈ S. Furthermore, the Kleene star is generalized by including the -operation (and the union) in the algebraic structure itself by the notion of complete star semiring.

  Read more →
• 

  Eigenface

  An eigenface ( EYE-gən-) is the name given to a set of eigenvectors when used in the computer vision problem of human face recognition. The approach of using eigenfaces for recognition was developed by Sirovich and Kirby and used by Matthew Turk and Alex Pentland in face classification. The eigenvectors are derived from the covariance matrix of the probability distribution over the high-dimensional vector space of face images. The eigenfaces themselves form a basis set of all images used to construct the covariance matrix. This produces dimension reduction by allowing the smaller set of basis images to represent the original training images. Classification can be achieved by comparing how faces are represented by the basis set. == History == The eigenface approach began with a search for a low-dimensional representation of face images. Sirovich and Kirby showed that principal component analysis could be used on a collection of face images to form a set of basis features. These basis images, known as eigenpictures, could be linearly combined to reconstruct images in the original training set. If the training set consists of M images, principal component analysis could form a basis set of N images, where N < M. The reconstruction error is reduced by increasing the number of eigenpictures; however, the number needed is always chosen less than M. For example, if you need to generate a number of N eigenfaces for a training set of M face images, you can say that each face image can be made up of "proportions" of all the K "features" or eigenfaces: Face image1 = (23% of E1) + (2% of E2) + (51% of E3) + ... + (1% En). In 1991 M. Turk and A. Pentland expanded these results and presented the eigenface method of face recognition. In addition to designing a system for automated face recognition using eigenfaces, they showed a way of calculating the eigenvectors of a covariance matrix such that computers of the time could perform eigen-decomposition on a large number of face images. Face images usually occupy a high-dimensional space and conventional principal component analysis was intractable on such data sets. Turk and Pentland's paper demonstrated ways to extract the eigenvectors based on matrices sized by the number of images rather than the number of pixels. Once established, the eigenface method was expanded to include methods of preprocessing to improve accuracy. Multiple manifold approaches were also used to build sets of eigenfaces for different subjects and different features, such as the eyes. == Generation == A set of eigenfaces can be generated by performing a mathematical process called principal component analysis (PCA) on a large set of images depicting different human faces. Informally, eigenfaces can be considered a set of "standardized face ingredients", derived from statistical analysis of many pictures of faces. Any human face can be considered to be a combination of these standard faces. For example, one's face might be composed of the average face plus 10% from eigenface 1, 55% from eigenface 2, and even −3% from eigenface 3. Remarkably, it does not take many eigenfaces combined together to achieve a fair approximation of most faces. Also, because a person's face is not recorded by a digital photograph, but instead as just a list of values (one value for each eigenface in the database used), much less space is taken for each person's face. The eigenfaces that are created will appear as light and dark areas that are arranged in a specific pattern. This pattern is how different features of a face are singled out to be evaluated and scored. There will be a pattern to evaluate symmetry, whether there is any style of facial hair, where the hairline is, or an evaluation of the size of the nose or mouth. Other eigenfaces have patterns that are less simple to identify, and the image of the eigenface may look very little like a face. The technique used in creating eigenfaces and using them for recognition is also used outside of face recognition: handwriting recognition, lip reading, voice recognition, sign language/hand gestures interpretation and medical imaging analysis. Therefore, some do not use the term eigenface, but prefer to use 'eigenimage'. === Practical implementation === To create a set of eigenfaces, one must: Prepare a training set of face images. The pictures constituting the training set should have been taken under the same lighting conditions, and must be normalized to have the eyes and mouths aligned across all images. They must also be all resampled to a common pixel resolution (r × c). Each image is treated as one vector, simply by concatenating the rows of pixels in the original image, resulting in a single column with r × c elements. For this implementation, it is assumed that all images of the training set are stored in a single matrix T, where each column of the matrix is an image. Subtract the mean. The average image a has to be calculated and then subtracted from each original image in T. Calculate the eigenvectors and eigenvalues of the covariance matrix S. Each eigenvector has the same dimensionality (number of components) as the original images, and thus can itself be seen as an image. The eigenvectors of this covariance matrix are therefore called eigenfaces. They are the directions in which the images differ from the mean image. Usually this will be a computationally expensive step (if at all possible), but the practical applicability of eigenfaces stems from the possibility to compute the eigenvectors of S efficiently, without ever computing S explicitly, as detailed below. Choose the principal components. Sort the eigenvalues in descending order and arrange eigenvectors accordingly. The number of principal components k is determined arbitrarily by setting a threshold ε on the total variance. Total variance ⁠ v = ( λ 1 + λ 2 + . . . + λ n ) {\displaystyle v=(\lambda _{1}+\lambda _{2}+...+\lambda _{n})} ⁠, n = number of components, and λ {\displaystyle \lambda } represents component eigenvalue. k is the smallest number that satisfies ( λ 1 + λ 2 + . . . + λ k ) v > ϵ {\displaystyle {\frac {(\lambda _{1}+\lambda _{2}+...+\lambda _{k})}{v}}>\epsilon } These eigenfaces can now be used to represent both existing and new faces: we can project a new (mean-subtracted) image on the eigenfaces and thereby record how that new face differs from the mean face. The eigenvalues associated with each eigenface represent how much the images in the training set vary from the mean image in that direction. Information is lost by projecting the image on a subset of the eigenvectors, but losses are minimized by keeping those eigenfaces with the largest eigenvalues. For instance, working with a 100 × 100 image will produce 10,000 eigenvectors. In practical applications, most faces can typically be identified using a projection on between 100 and 150 eigenfaces, so that most of the 10,000 eigenvectors can be discarded. === Matlab example code === Here is an example of calculating eigenfaces with Extended Yale Face Database B. To evade computational and storage bottleneck, the face images are sampled down by a factor 4×4=16. Note that although the covariance matrix S generates many eigenfaces, only a fraction of those are needed to represent the majority of the faces. For example, to represent 95% of the total variation of all face images, only the first 43 eigenfaces are needed. To calculate this result, implement the following code: === Computing the eigenvectors === Performing PCA directly on the covariance matrix of the images is often computationally infeasible. If small images are used, say 100 × 100 pixels, each image is a point in a 10,000-dimensional space and the covariance matrix S is a matrix of 10,000 × 10,000 = 108 elements. However the rank of the covariance matrix is limited by the number of training examples: if there are N training examples, there will be at most N − 1 eigenvectors with non-zero eigenvalues. If the number of training examples is smaller than the dimensionality of the images, the principal components can be computed more easily as follows. Let T be the matrix of preprocessed training examples, where each column contains one mean-subtracted image. The covariance matrix can then be computed as S = TTT and the eigenvector decomposition of S is given by S v i = T T T v i = λ i v i {\displaystyle \mathbf {Sv} _{i}=\mathbf {T} \mathbf {T} ^{T}\mathbf {v} _{i}=\lambda _{i}\mathbf {v} _{i}} However TTT is a large matrix, and if instead we take the eigenvalue decomposition of T T T u i = λ i u i {\displaystyle \mathbf {T} ^{T}\mathbf {T} \mathbf {u} _{i}=\lambda _{i}\mathbf {u} _{i}} then we notice that by pre-multiplying both sides of the equation with T, we obtain T T T T u i = λ i T u i {\displaystyle \mathbf {T} \mathbf {T} ^{T}\mathbf {T} \mathbf {u} _{i}=\lambda _{i}\mathbf {T} \mathbf {u} _{i}} Meaning that, if ui is an eigenvector of TTT, then vi = Tui is an eigenvector of S. If we have

  Read more →
• 

  Business Controls Corporation

  Business Controls Corporation is a privately held computer company that developed an application-program-generator and also a series of accounting software packages. These packages were widely enough used for various business magazines to have back-of-the-book ads for companies seeking accountants with experience in one or more of them. Computer magazines ran coverage for their SB-5 application-program-generator as from time to time new versions were released, each with new or improved features. == Early days == The company's initial offerings were packages for the DEC PDP-8, although Business Controls Corporation also wrote custom-written programs for customers. Large customers with mainframes who also used smaller systems for departmental use and distributed processing also used BCC's services. == SB-5 == The addition of an application-program-generator named SB-5 that, from specifications, could generate COBOL code was a major step forward. Although this began with supporting the DEC PDP-11, they subsequently began to support COBOL on DEC's DECsystem-10 & DECSYSTEM-20. VAX support came later. The specifications also permitted COBOL inserts and overrides: SB-5 could build an application that was all COBOL, yet only code the portions that varied from BCC's "vanilla" accounting packages. === Similar offerings === A similar idea was done for the IBM mainframe world in the form of a series of application-program-generators from Dylakor Corporation. They were named DYL-250, DYL-260, DYL-270 & DYL-280. Dylakor was acquired by Computer Associates. The specific syntax was different, but it had wider use, and - a mark of success and recognition in the industry - syntax-compatible implementations were released by a competitor. Still another alternative was Peat Marwick Mitchell's PMM2170 application-program-generator package. Like the others, it supported COBOL inserts and overrides. === Extended integration === Business Controls Corporation subsequently extended SB-5's feature set to provide support for System 1022, a product for the DECsystem-10 & DECSYSTEM-20; 1022's vendor also had a VAX/VMS (later OpenVMS) product, System 1032.

  Read more →
• 

  International Conference on Language Resources and Evaluation

  The International Conference on Language Resources and Evaluation is an international conference organised by the ELRA Language Resources Association every other year (on even years) with the support of institutions and organisations involved in Natural language processing. The series of LREC conferences was launched in Granada in 1998. == History of conferences == The survey of the LREC conferences over the period 1998-2013 was presented during the 2014 conference in Reykjavik as a closing session. It appears that the number of papers and signatures is increasing over time. The average number of authors per paper is higher as well. The percentage of new authors is between 68% and 78%. The distribution between male (65%) and female (35%) authors is stable over time. The most frequent technical term is "annotation", then comes "part-of-speech". == The LRE Map == The LRE Map was introduced at LREC 2010 and is now a regular feature of the LREC submission process for both the conference papers and the workshop papers. At the submission stage, the authors are asked to provide some basic information about all the resources (in a broad sense, i.e. including tools, standards and evaluation packages), either used or created, described in their papers. All these descriptors are then gathered in a global matrix called the LRE Map. This feature has been extended to several other conferences.

  Read more →
• 

  Brave Leo

  Brave Leo is a large language model-based chatbot developed by Brave Software and included with the Brave browser. == History == In November 2023, the company said versions for iOS and Android would be available "in the coming months". == Features == Since January 2024, Leo has used the open-source Mixtral 8x7B from Mistral AI as its default large language model, in addition to LLaMA 2 from Meta Platforms and Claude from Anthropic, both of which have been used previously. Leo can suggest follow-up questions, and summarize webpages, PDFs, and videos. Leo has a $15 (US) per month premium version that enables more requests and uses larger LLMs. == Privacy == The answers given by Leo are not saved. Brave uses the slogan Love Privacy to emphasize its focus on user privacy and data protection. The phrase has been featured in Brave's official marketing campaigns and has been cited in media coverage of the browser's privacy-first approach. == Controversies == In 2023, PC World reported that Leo evades questions about US elections.

  Read more →
• 

  TipTop Technologies

  TipTop Technologies is a real-time web and social search engine with a platform for semantic analysis of natural language. Tip-Top Search provides results capturing individual and group sentiment, opinions, and experiences there from the content of various sorts such as real-time messages from Twitter or consumer product reviews on Amazon.com. TipTop Technologies and ITC Infotech collaborated to create a search interface suitable for both enterprise and consumer applications. Tip-Top's products are part of the "emerging Web 3.0 applications which use semantic technologies to augment the underlying Web system's functionalities." Their main product is 360, an AI tool that incorporates multiple AI applications under one wing. Jonathan AlBright professor at Elon University, found videos generated by TipTop Technologies software on YouTube in his research into artificial intelligence, described it as AI-generated "fake news". Through semantic analysis of large data sets, TipTop gleaned behavioral insights from Tweets around events like Halloween, Thanksgiving, Holiday Gifting, the Super Bowl, and the Oscar Nominees for the Academy Awards coverage. Sentiment analysis, concept trend tracking, and real-time market research are other applications included in the TipTop Search product. TipTop's insight engine solves the problem of real-time data noise, and its ability to "sort the 'good tweets' from the 'bad tweets' when it comes to a product, service, or a region..." In addition, products like TipTop Shopping with customizable search widgets bring together consumer reviews, social search, and sentiment analysis enabling product comparisons across attributes like the overall value and aiding purchasing decisions through user-driven product tips and pits. TipTop Finance adds another complexity to real-time search results by incorporating corporate sentiment, company stock tickers, and social media into TipTop's existing social search platform. Additional success applying semantic technologies has been with polling, "if you compare these Gallup results with TipTop, a sentiment engine based on Twitter, the results are not way off. It does surprise you but it tells me that sentiment analysis in case of public opinion about a burning social issue or a famous personality is relatively easier." With the increasing amount of unstructured, opinion-oriented, and user-generated content available on the Web, TipTop's technology aims to make sense of all this data, and deliver it in a useful way for consumer and enterprise users alike. TipTop Technologies is a privately held company with its headquarters in the San Francisco Bay Area, and team members are located globally.

  Read more →
• 

  SWILE

  SWILE (formerly: Lunchr) is a French app-based company that focuses on improving the employee experience. Among others, the platform offers meal vouchers, gift vouchers, mobility vouchers, and business travel solutions. In March 2020, it was renamed SWILE and entered the lunch break and meal voucher market. == History == The company was founded as Lunchr by Loïc Soubeyrand in 2016. Originally, Lunchr was an app for pre-ordering lunch on the spot or to go. In January 2017, the company raised €2.5 million in seed funding from Daphni. In 2018, the company raised €11 million (series A) from Idinvest, followed by another €30 million in February 2019 (series B), notably from Index Ventures and Kima Ventures. In January 2020, Lunchr became one of the first startups to join the French Tech 120. A few months later, in March, Lunchr diversified its services, adding team life management tools and changing its brand name to Swile. In June 2020, the company raised €70 million more in a new round of financing (Series C) from the same investors and the BPI. In November 2020, Swile acquired Briq, a startup specializing in employee engagement. In January 2021, Swile won a tender with Carrefour and distributed 62,000 Swile cards to its employees. In early October 2021, a new $200 million (€175 million) fundraising round, in which Japanese Softbank joined other investors, allowed Swile to capitalize on $1 billion. President Emmanuel Macron cited the company as "a further proof that FrenchTech is at the forefront internationally." In May 2022, the company acquired the travel management start-up Okarito for €6 million. == Overview == Swile operates in two countries (France and Brazil) and has a total of 1000 employees, 5.5 million users and 85,000 corporate customers, including Carrefour, Le Monde, JCDECAUX, PSG, Airbnb, Spotify, Red Bull, and TikTok in the private sector, as well as numerous local authorities and ministerial references in the public sector.

  Read more →
• 

  Retrieval-augmented generation

  Retrieval-augmented generation (RAG) is a technique that enables large language models (LLMs) to retrieve and incorporate new information from external data sources. With RAG, LLMs first refer to a specified set of documents, then respond to user queries. These documents supplement information from the LLM's pre-existing training data. This allows LLMs to use domain-specific and/or updated information that is not available in the training data. For example, this enables LLM-based chatbots to access internal company data or generate responses based on authoritative sources. RAG improves LLMs by incorporating information retrieval before generating responses. Unlike LLMs that rely on static training data, RAG pulls relevant text from databases, uploaded documents, or web sources. According to Ars Technica, "RAG is a way of improving LLM performance, in essence by blending the LLM process with a web search or other document look-up process to help LLMs stick to the facts." This method helps reduce AI hallucinations, which have caused chatbots to describe policies that don't exist, or recommend nonexistent legal cases to lawyers that are looking for citations to support their arguments. RAG also reduces the need to retrain LLMs with new data, saving on computational and financial costs. Beyond efficiency gains, RAG also allows LLMs to include sources in their responses, so users can verify the cited sources. This provides greater transparency, as users can cross-check retrieved content to ensure accuracy and relevance. The term retrieval-augmented generation (RAG) was introduced in a 2020 paper that described combining a parametric language model with a non-parametric external memory accessed through retrieval at inference time. == RAG and LLM limitations == LLMs can provide incorrect information. For example, when Google first demonstrated its LLM tool "Google Bard" (later re-branded to Gemini), the LLM provided incorrect information about the James Webb Space Telescope. This error contributed to a $100 billion decline in Google's stock value. RAG is used to prevent these errors, but it does not solve all the problems. For example, LLMs can generate misinformation even when pulling from factually correct sources if they misinterpret the context. MIT Technology Review gives the example of an AI-generated response stating, "The United States has had one Muslim president, Barack Hussein Obama." The model retrieved this from an academic book rhetorically titled Barack Hussein Obama: America's First Muslim President? The LLM did not "know" or "understand" the context of the title, generating a false statement. LLMs with RAG are programmed to prioritize new information. This technique has been called "prompt stuffing." Without prompt stuffing, the LLM's input is generated by a user; with prompt stuffing, additional relevant context is added to this input to guide the model's response. This approach provides the LLM with key information early in the prompt, encouraging it to prioritize the supplied data over pre-existing training knowledge. == Process == Retrieval-augmented generation (RAG) enhances large language models (LLMs) by incorporating an information-retrieval mechanism that allows models to access and utilize additional data beyond their original training set. Ars Technica notes that "when new information becomes available, rather than having to retrain the model, all that's needed is to augment the model's external knowledge base with the updated information" ("augmentation"). IBM states that "in the generative phase, the LLM draws from the augmented prompt and its internal representation of its training data to synthesize" an answer. === RAG key stages === Typically, the data to be referenced is converted into LLM embeddings, numerical representations in the form of a large vector space. RAG can be used on unstructured (usually text), semi-structured, or structured data (for example knowledge graphs). These embeddings are then stored in a vector database to allow for document retrieval. Given a user query, a document retriever is first called to select the most relevant documents that will be used to augment the query. This comparison can be done using a variety of methods, which depend in part on the type of indexing used. The model feeds this relevant retrieved information into the LLM via prompt engineering of the user's original query. Newer implementations (as of 2023) can also incorporate specific augmentation modules with abilities such as expanding queries into multiple domains and using memory and self-improvement to learn from previous retrievals. Finally, the LLM can generate output based on both the query and the retrieved documents. Some models incorporate extra steps to improve output, such as the re-ranking of retrieved information, context selection, and fine-tuning. == Applications == Retrieval-augmented generation is used in applications where generated responses need to be grounded in external or frequently updated information. Commonly cited use cases include search engines, question-answering systems, customer support chatbots, enterprise knowledge assistants, content generation, recommendation systems, retail and e-commerce, and industrial or manufacturing workflows. In healthcare, RAG has been studied as a way to ground large language model outputs in external medical knowledge sources, although reviews have noted continuing challenges around evaluation, ethics, and clinical reliability. == Improvements == Improvements to the basic process above can be applied at different stages in the RAG flow. === Encoder === These methods focus on the encoding of text as either dense or sparse vectors. Sparse vectors, which encode the identity of a word, are typically dictionary-length and contain mostly zeros. Dense vectors, which encode meaning, are more compact and contain fewer zeros. Various enhancements can improve the way similarities are calculated in the vector stores (databases). Performance improves by optimizing how vector similarities are calculated. Dot products enhance similarity scoring, while approximate nearest neighbor (ANN) searches improve retrieval efficiency over K-nearest neighbors (KNN) searches. Accuracy may be improved with Late Interactions, which allow the system to compare words more precisely after retrieval. This helps refine document ranking and improve search relevance. Hybrid vector approaches may be used to combine dense vector representations with sparse one-hot vectors, taking advantage of the computational efficiency of sparse dot products over dense vector operations. Other retrieval techniques focus on improving accuracy by refining how documents are selected. Some retrieval methods combine sparse representations, such as SPLADE, with query expansion strategies to improve search accuracy and recall. === Retriever-centric methods === These methods aim to enhance the quality of document retrieval in vector databases: Pre-training the retriever using the Inverse Cloze Task (ICT), a technique that helps the model learn retrieval patterns by predicting masked text within documents. Supervised retriever optimization aligns retrieval probabilities with the generator model's likelihood distribution. This involves retrieving the top-k vectors for a given prompt, scoring the generated response's perplexity, and minimizing KL divergence between the retriever's selections and the model's likelihoods to refine retrieval. Reranking techniques can refine retriever performance by prioritizing the most relevant retrieved documents during training. === Language model === By redesigning the language model with the retriever in mind, a 25-time smaller network can get comparable perplexity as its much larger counterparts. Because it is trained from scratch, this method (Retro) incurs the high cost of training runs that the original RAG scheme avoided. The hypothesis is that by giving domain knowledge during training, Retro needs less focus on the domain and can devote its smaller weight resources only to language semantics. The redesigned language model is shown here. It has been reported that Retro is not reproducible, so modifications were made to make it so. The more reproducible version is called Retro++ and includes in-context RAG. === Chunking === Chunking involves various strategies for breaking up the data into vectors so the retriever can find details in it. Three types of chunking strategies are: Fixed length with overlap. This is fast and easy. Overlapping consecutive chunks helps to maintain semantic context across chunks. Syntax-based chunks can break the document up into sentences. Libraries such as spaCy or NLTK can also help. File format-based chunking. Certain file types have natural chunks built in, and it's best to respect them. For example, code files are best chunked and vectorized as whole functions or classes. HTML files should leave

  or base64 encoded elements

  Read more →

  

  Loebner Prize

  The Loebner Prize was an annual competition in artificial intelligence that awarded prizes to the computer programs considered by the judges to be the most human-like. The format of the competition was that of a standard Turing test. In each round, a human judge simultaneously held textual conversations with a computer program and a human being via computer. Based upon the responses, the judge would attempt to determine which was which. The contest was launched in 1990 by Hugh Loebner in conjunction with the Cambridge Center for Behavioral Studies, Massachusetts, United States. In 2004 and 2005, it was held in Loebner's apartment in New York City. Within the field of artificial intelligence, the Loebner Prize is somewhat controversial; the most prominent critic, Marvin Minsky, called it a publicity stunt that does not help the field along. Beginning in 2014, it was organised by the AISB at Bletchley Park. It has also been associated with Flinders University, Dartmouth College, the Science Museum in London, University of Reading and Ulster University, Magee Campus, Derry, UK City of Culture. For the final 2019 competition, the format changed. There was no panel of judges. Instead, the chatbots were judged by the public and there were to be no human competitors. The prize has been reported as defunct as of 2020. == Prizes == Originally, $2,000 was awarded for the most human-seeming program in the competition. The prize was $3,000 in 2005 and $2,250 in 2006. In 2008, $3,000 was awarded. In addition, there were two one-time-only prizes that have never been awarded. $25,000 is offered for the first program that judges cannot distinguish from a real human and which can convince judges that the human is the computer program. $100,000 is the reward for the first program that judges cannot distinguish from a real human in a Turing test that includes deciphering and understanding text, visual, and auditory input. The competition was planned to end after the achievement of this prize. == Competition rules and restrictions == The rules varied over the years and early competitions featured restricted conversation Turing tests but since 1995 the discussion has been unrestricted. For the three entries in 2007, Robert Medeksza, Noah Duncan and Rollo Carpenter, some basic "screening questions" were used by the sponsor to evaluate the state of the technology. These included simple questions about the time, what round of the contest it is, etc.; general knowledge ("What is a hammer for?"); comparisons ("Which is faster, a train or a plane?"); and questions demonstrating memory for preceding parts of the same conversation. "All nouns, adjectives and verbs will come from a dictionary suitable for children or adolescents under the age of 12." Entries did not need to respond "intelligently" to the questions to be accepted. For the first time in 2008 the sponsor allowed introduction of a preliminary phase to the contest opening up the competition to previously disallowed web-based entries judged by a variety of invited interrogators. The available rules do not state how interrogators are selected or instructed. Interrogators (who judge the systems) have limited time: 5 minutes per entity in the 2003 competition, 20+ per pair in 2004–2007 competitions, 5 minutes to conduct simultaneous conversations with a human and the program in 2008–2009, increased to 25 minutes of simultaneous conversation since 2010. == Criticisms == The prize has long been scorned by experts in the field, for a variety of reasons. It is regarded by many as a publicity stunt. Marvin Minsky scathingly offered a "prize" to anyone who could stop the competition. Loebner responded by jokingly observing that Minsky's offering a prize to stop the competition effectively made him a co-sponsor. The rules of the competition have encouraged poorly qualified judges to make rapid judgements. Interactions between judges and competitors was originally very brief, for example effectively 2.5 mins of questioning, which permitted only a few questions. Questioning was initially restricted to a single topic of the contestant's choice, such as "whimsical conversation", a domain suiting standard chatbot tricks. Competition entrants do not aim at understanding or intelligence but resort to basic ELIZA style tricks, and successful entrants find deception and pretense is rewarded. == Contests == See article history for more details of some earlier contests. A very incomplete listing of a few of the contests: === 2003 === In 2003, the contest was organised by Professor Richard H. R. Harper and Dr. Lynne Hamill from the Digital World Research Centre at the University of Surrey. Although no bot passed the Turing test, the winner was Jabberwock, created by Juergen Pirner. Second was Elbot (Fred Roberts, Artificial Solutions). Third was Jabberwacky, (Rollo Carpenter). === 2006 === In 2006, the contest was organised by Tim Child (CEO of Televirtual) and Huma Shah. On August 30, the four finalists were announced: Rollo Carpenter Richard Churchill and Marie-Claire Jenkins Noah Duncan Robert Medeksza The contest was held on 17 September in the VR theatre, Torrington Place campus of University College London. The judges included the University of Reading's cybernetics professor, Kevin Warwick, a professor of artificial intelligence, John Barnden (specialist in metaphor research at the University of Birmingham), a barrister, Victoria Butler-Cole and a journalist, Graham Duncan-Rowe. The latter's experience of the event can be found in an article in Technology Review. The winner was 'Joan', based on Jabberwacky, both created by Rollo Carpenter. === 2007 === The 2007 competition was held on October 21 in New York City. The judges were: computer science professor Russ Abbott, philosophy professor Hartry Field, psychology assistant professor Clayton Curtis and English lecturer Scott Hutchins. No bot passed the Turing test, but the judges ranked the three contestants as follows: 1st: Robert Medeksza, creator of Ultra Hal 2nd: Noah Duncan, a private entry, creator of Cletus 3rd: Rollo Carpenter from Icogno, creator of Jabberwacky The winner received $2,250 and the annual medal. The runners-up received $250 each. === 2008 === The 2008 competition was organised by professor Kevin Warwick, coordinated by Huma Shah and held on October 12 at the University of Reading, UK. After testing by over one hundred judges during the preliminary phase, in June and July 2008, six finalists were selected from thirteen original entrant artificial conversational entities (ACEs). Five of those invited competed in the finals: Brother Jerome, Peter Cole and Benji Adams Elbot, Fred Roberts / Artificial Solutions Eugene Goostman, Vladimir Veselov, Eugene Demchenko and Sergey Ulasen Jabberwacky, Rollo Carpenter Ultra Hal, Robert Medeksza In the finals, each of the judges was given five minutes to conduct simultaneous, split-screen conversations with two hidden entities. Elbot of Artificial Solutions won the 2008 Loebner Prize bronze award, for most human-like artificial conversational entity, through fooling three of the twelve judges who interrogated it (in the human-parallel comparisons) into believing it was human. This is coming very close to the 30% traditionally required to consider that a program has actually passed the Turing test. Eugene Goostman and Ultra Hal both deceived one judge each that it was the human. Will Pavia, a journalist for The Times, has written about his experience; a Loebner finals' judge, he was deceived by Elbot and Eugene. Kevin Warwick and Huma Shah have reported on the parallel-paired Turing tests. === 2009 === The 2009 Loebner Prize Competition was held September 6, 2009, at the Brighton Centre, Brighton UK in conjunction with the Interspeech 2009 conference. The prize amount for 2009 was $3,000. Entrants were David Levy, Rollo Carpenter, and Mohan Embar, who finished in that order. The writer Brian Christian participated in the 2009 Loebner Prize Competition as a human confederate, and described his experiences at the competition in his book The Most Human Human. === 2010 === The 2010 Loebner Prize Competition was held on October 23 at California State University, Los Angeles. The 2010 competition was the 20th running of the contest. The winner was Bruce Wilcox with Suzette. === 2011 === The 2011 Loebner Prize Competition was held on October 19 at the University of Exeter, Devon, United Kingdom. The prize amount for 2011 was $4,000. The four finalists and their chatterbots were Bruce Wilcox (Rosette), Adeena Mignogna (Zoe), Mohan Embar (Chip Vivant) and Ron Lee (Tutor), who finished in that order. That year there was an addition of a panel of junior judges, namely Georgia-Mae Lindfield, William Dunne, Sam Keat and Kirill Jerdev. The results of the junior contest were markedly different from the main contest, with chatterbots Tutor and Zoe tying for first place and Chip Vivant and Rosette coming in third and fourt

  Read more →