Proposes a model to improve multi-domain recommender systems.
problem Challenges in transferring knowledge between domains in recommender systems.
method Generative adversarial networks (GANs), Variational Autoencoders (VAEs), and Cycle-Consistency (CC) for weight-sharing.
result Improves performance of multi-domain recommender systems by capturing both similarities and differences among domains.
We address the problem of domain generalization where a decision function is learned from the data of several related domains, and the goal is to apply it on an unseen domain successfully. It is assumed that there is plenty of labeled data available in source domains (also called as training domain), but no labeled dat…
Large batch sizes improve model performance but have limits; we found a simple predictor.
problem Understanding the limits of large batch sizes across different domains.
method Empirical model using gradient noise scale to predict optimal batch size.
result Gradient noise scale predicts the largest useful batch size across various domains.
Empirical law predicts accuracy of Google Translate's translation chains.
problem Predicting accuracy in machine translation with multiple hops.
method Empirical testing of Google Translate's sequential translation.
result Accuracy decreases with the number of translating hops, following a power law.
The authors of (Cho et al., 2014a) have shown that the recently introduced neural network translation systems suffer from a significant drop in translation quality when translating long sentences, unlike existing phrase-based translation systems. In this paper, we propose a way to address this issue by automatically se…
Paper proposes a method to learn word translations bidirectionally.
problem Word translation between languages.
method Jointly learns translations in both directions with minimal supervision.
result Improves accuracy of translations over previous methods.
Study on rigidity of translating hypersurfaces not in graphical direction.
problem Rigidity of translating hypersurfaces not in graphical direction.
method Proved rigidity results for complete graphical translating hypersurfaces under specific conditions.
result Entire graphical translating surfaces are flat under certain conditions.
Forward translation improves neural machine translation for sentences originally in source language.
problem Improving neural machine translation quality using synthetic data.
method Case study with French-English news translation, separating test sets by original language, analyzing domains, translationese, and noise.
result Forward translation delivers superior gains on sentences originally in source language, complementing back-translation on target language sentences.
The natural automorphism group of a translation surface is its group of translations. For finite translation surfaces of genus g > 1 the order of this group is naturally bounded in terms of g due to a Riemann-Hurwitz formula argument. In analogy with classical Hurwitz surfaces, we call surfaces which achieve the maxima…
Study on stable translation lengths of surface homeomorphisms and their approximations.
problem Understanding stable translation lengths of homeomorphisms and their finite approximations.
method Comparing stable translation lengths of homeomorphisms and their finite approximations on curve graphs.
result Stable translation length of homeomorphisms with dense periodic points equals the supremum of their approximations.
Researchers classify and describe Kα-translators in Euclidean space.
problem Classifying and describing Kα-translators in Euclidean space. method Rotationally symmetric and helicoidal motions.
result For each α, there is a Kα-translator intersecting orthogonally the rotation axis. The quality of machine translation is rapidly evolving. Today one can find several machine translation systems on the web that provide reasonable translations, although the systems are not perfect. In some specific domains, the quality may decrease. A recently proposed approach to this domain is neural machine translat…
Neural machine translation models trained for 5 South African languages.
problem Lack of resources and research for machine translation in African languages.
method Training neural machine translation models for 5 South African languages using modern techniques.
result Promises of neural machine translation for African languages.
Study measures gender bias in machine translation using multiple reference points.
problem Measuring and identifying gender bias in machine translation.
method Used an optimal non-biased translator, reference points from occupational statistics and survey.
result Found bias against both genders, but more against women, and found occupations have a greater effect than adjectives.
Classifies and constructs translators for curvature flows.
problem Understanding translating solitons in curvature flows.
method Developed rotational theory, introduced signed-neck framework.
result Classified and constructed catenoidal-type translators.
This paper uses LLMs and cycle consistency for better machine translation evaluation.
problem Evaluating translation quality and LLM capabilities without ground truth.
method Generate translation candidates, back-translate, and evaluate cycle consistency.
result Larger LLMs or more inference passes improve cycle consistency.
Neural machine translation used to convert CUDA to OpenCL.
problem Translating CUDA to OpenCL programs.
method Training input set generation, pre/post processing, case study.
result Improved accuracy in translating CUDA to OpenCL.
Paper finds a non-existence theorem for certain translators in high dimensions.
problem Non-existence of certain translators in high-dimensional spaces.
method Developed a non-existence theorem and found an example of a translator.
result Non-existence of entire Qn−1-translators in Rn+1. Constructing translating solitons from Lagrangian Grim Reapers.
problem Creating Lagrangian translating solitons from intersections of Grim Reapers.
method Desingularizing intersections with special Lagrangian Lawlor necks.
result Constructing Lagrangian translating solitons with multiple ends and loops.
Study classifies translators for mean curvature flow in 3D.
problem Classifying semigraphical translators for mean curvature flow in R3. method Morse-Radó theory and angular maximum principle.
result No solution to the translator equation on the upper half-plane with alternating boundary values.
In an era when big data are becoming the norm, there is less concern with the quantity but more with the quality and completeness of the data. In many disciplines, data are collected from heterogeneous sources, resulting in multi-view or multi-modal datasets. The missing data problem has been challenging to address in …
Study on singular points of translation surfaces under linearly dependent conditions.
problem Investigate singular points of translation surfaces under linearly dependent conditions.
method Use theories of generalised framed surfaces and framed surfaces.
result Introduce translation generalised framed surfaces and investigate their singular points.
Proves uniqueness of translators in 3D space.
problem Uniqueness of pitchfork and helicoid translators in mean curvature flow.
method Arc-counting argument and rotational maximum principle.
result Proves conjecture on uniqueness of translators.
An NMT system for Indic languages outperforms Google Translate.
problem Challenges in translating Indic languages efficiently.
method Encoder-decoder with attention mechanism for neural machine translation.
result Outperforms Google Translate with a 6 BLEU score margin on English-Gujarati translation.
New families of translation surfaces with multiple oblivious points discovered.
problem Identifying points on translation surfaces without nearby closed geodesics.
method Constructing new families of translation surfaces and proving existence in higher genera.
result Translation surfaces in every genus ≥3 have at least one oblivious point.
In this article we prove two non-existence results for translating solitons of the mean curvature flow (translators for short) in Rm+1. We also obtain an upper bound to the maximum height that a compact embedded translator in R3 can achieve. On the other hand, we study graphical perturbation…
CrossNet uses cross-consistency to improve unpaired image translation.
problem Image-to-image translation without paired data.
method Introduced a novel architecture with cross-translators and latent cross-consistency constraints.
result CrossNet outperforms state-of-the-art on various image translation tasks.
27 problems identified in automating movie/TV subtitle translation.
problem Challenges in translating movie/TV subtitles.
method Categorized problems into three categories and evaluated translation quality.
result Frontier NLP systems struggle with subtitles and require post-processing.
This paper classifies grim reapers in a specific product space.
problem Classifying grim reapers in a product space.
method Analyzing mean curvature flow and translations in $\h^2 imes
$.
result A full classification of grim reapers in $\h^2 imes
$.
Neural machine translation is a recently proposed approach to machine translation. Unlike the traditional statistical machine translation, the neural machine translation aims at building a single neural network that can be jointly tuned to maximize the translation performance. The models proposed recently for neural ma…
Study invariant λ-translators in Lorentz-Minkowski space.
problem Characterize λ-translators invariant under translations and rotations. method Analyze 1-parameter group of translations and rotations, find explicit parametrizations, and solve non-linear autonomous systems.
result Explicit parametrizations and qualitative properties of invariant λ-translators. Unique minimal translation surfaces found in Heisenberg group.
problem Classifying minimal translation surfaces in Heisenberg group.
method Complete classification through generating curves analysis.
result Uniqueness of minimal translation surfaces up to isometries.
Paper tackles NAT translation issues with auxiliary regularization.
problem Improves NAT translation quality by addressing repeated and incomplete translations.
method Improves decoder hidden representations via two auxiliary regularization terms.
result Significant improvement in NAT model accuracy with better inference efficiency.
Study classifies and characterizes translators in hyperbolic static universe.
problem Classifying and characterizing translators in hyperbolic static universe.
method Classified and characterized translators foliated by horospheres and rotationally invariant ones, both space-like and time-like.
result Obtained a characterization of the bowl and certain translators foliated by horospheres.
The study proves stability of various graphical translators in mean curvature flow.
problem Stability of graphical translators in mean curvature flow.
method Existence of longtime solution to mean curvature flow, dynamical stability results for various graphical translators.
result Dynamical stability of various types of graphical translators.
Paper tackles zero-shot translation by encouraging consistent agreement in models.
problem Challenges of generalizing multilingual translation without parallel data.
method Reformulated as probabilistic inference, introduced consistent agreement-based training.
result Agreement-based learning improves zero-shot translation by 2-3 BLEU points.
Study translators in Generalised Robertson-Walker spacetimes, identifying warping functions and classifying examples.
problem Understanding translators in Generalised Robertson-Walker spacetimes.
method Analyzing translators as submanifolds satisfying the mean curvature flow equation with a specific vector field.
result Identification of three one-parameter families of warping functions and classification of translators.
The paper shows how integrating categorical semantics can enhance unsupervised domain translation.
problem Improving unsupervised domain translation between perceptually different domains.
method Learning invariant categorical semantic features in an unsupervised manner and conditioning them on the style encoder.
result Conditioning the style encoder on learned categorical semantics improves translation and stylization.
The paper explores new translating solitons and their relation to Δ-wings.
problem Understanding translating solitons and their properties.
method Description of new annular examples and proving related results.
result Proves several results about translating solitons and their relation to Δ-wings. Study shows only grim reaper cylinder for certain self-translating surfaces.
problem Characterizing self-translating surfaces in 3D space.
method Used parabolicity in a weighted setting and universally L-superharmonic functions.
result Characterized the grim reaper cylinder as the only finite entropy self-translating 2-surface in R^3 of width π and bounded from below.
Study proves conditions for translating solitons to be planar.
problem Conditions for translating solitons to be planar.
method Investigates translating solitons as sums and products of curves, extends results to Lorentz-Minkowski space.
result Translating solitons can be planar if one component is planar.
We study volume growth, entropy and stability for translating solitons of mean curvature flow. First, we prove that every complete properly immersed translator has at least linear volume growth. Then, by using Huisken's monotonicity formula, we compute the entropy of the grim reaper and the bowl solitons. We also give …
Researchers create annular translators for mean curvature flow.
problem Constructing complete, properly embedded annular translators.
method Constructing a family of complete, properly embedded, annular translators in a slab.
result For each inner width \( b \ge \pi/2 \) and necksize \( s > 0 \), there exists a translator.
We introduce and describe the results of a novel shared task on bandit learning for machine translation. The task was organized jointly by Amazon and Heidelberg University for the first time at the Second Conference on Machine Translation (WMT 2017). The goal of the task is to encourage research on learning machine tra…
The paper classifies special surfaces in a 3D space.
problem Classifying λ-translators in S2imesR. method Investigates surfaces with specific curvature properties under group actions.
result Identifies all λ-translators invariant by rotations and vertical translations. Improved understanding of translating solitons using new techniques.
problem Understanding translating solitons in geometry.
method Using a new test function and gradient estimate technique.
result Better Bernstein type result of translating solitons.
Analyzes translating soliton equation with convex domains.
problem Analyzing translating soliton equation in convex domains.
method Analytic approach to solving translating soliton equation.
result Study of Dirichlet problem in convex domains of the plane.
In this paper, we compute the first and second variation formulas for the F-functional of translating solitons and study the Hamiltonian L-stability of Lagrangian translating solitons. We prove that any Lagrangian translating soliton is Hamiltonian L-stable.