Project develops lip reading algorithm for limited English.
problem Limited audio information for lip reading.
method Extract lip positions from video frames, classify visemes and phonemes, use HMMs to predict words.
result Algorithm predicts words from lip movements for a subset of English.
Researchers compare two GAN models for lip-synchronization tasks.
problem Improving the quality of generated lip-sync videos.
method Reimplemented and trained two GAN models (LipGAN and L1WGAN-GP) on the GRID dataset.
result L1WGAN-GP outperformed LipGAN in generating more realistic lip-sync videos.
We reformulate LIPs as min-max problems for easier solution.
problem Recovering signals from few linear measurements.
method Proposed a min-max reformulation of LIPs.
result Saddle points characterize solutions to LIPs.
Lips and swallow-tails are generic local moves of singularities of a smooth map to a 2-manifold. We prove that these moves of singularities of the product map of two functions on a 3-manifold can be realized by isotopies of the functions.
We show that at generic points blow-ups/tangents of differentiability spaces are still differentiability spaces; this implies that an analytic condition introduced by Keith as an inequality (and later proved to actually be an equality) passes to tangents. As an application, we characterize the p-weak gradient on iter…
The paper explores how close two Lipschitz functions can be without their difference exceeding a certain bound.
problem Understanding the closeness of two Lipschitz functions and their difference.
method Investigates the relationship between two Lip(γ) functions being close throughout a subset of their domain and the bound on the difference's Lipschitz norm. result The Lipschitz norm of the difference between two functions is bounded by a small value when the distance to a subset is small.
We provide a sufficient condition for the nontriviality of the Lipschitz homotopy group of the Heisenberg group, πmLip(Hn), in terms of properties of the classical homotopy group of the sphere, πm(Sn). As an application we provide a new simplified proof of the fact that πnLip(Hn)=0, n=1,2,..., a…
Paper tackles cold-start domain adaptation with language descriptions.
problem Cold-start domain adaptation failure with scarce target data.
method Leverages textual descriptions to learn preferences from LLM, integrates into EM algorithm.
result Framework guides source selection for weak target signals, improves as data accumulates.
Solves signal recovery from few linear measurements using convex duality.
problem Recovering signals from limited linear measurements in various applications.
method Develops a convex-concave min-max reformulation for linear inverse problems.
result Simple ascent-descent algorithms for solving linear inverse problems.
A new estimator reduces variance in slate bandit OPE.
problem Large action spaces in slate bandits cause high variance in OPE.
method Develops Latent IPS (LIPS) to optimize slate abstractions for low variance and bias.
result LIPS substantially outperforms existing estimators in scenarios with non-linear rewards and large slate spaces.
Let us consider a Riemannian manifold M (either separable or non-separable). We prove that, for every ε>0, every Lipschitz function f:M→R can be uniformly approximated by a Lipschitz, C1-smooth function g with $\Lip(g)\le \Lip(f)+ε$. As a consequence, every Riemannian manifold is uniformly …
Sharp estimates for heat flow on nonconvex domains.
problem Quantitative estimates for heat flow on nonconvex domains.
method Sharp gradient and transport estimates with novel dependence on time.
result Equivalent characterization of lower bound on second fundamental form.
We construct a smooth compact n-dimensional manifold Y with one point singularity such that all its Lipschitz homotopy groups are trivial, but Lipschitz mappings Lip(S^n,Y) are not dense in the Sobolev space W^{1,n}(S^n,Y). On the other hand we show that if a metric space Y is Lipschitz (n-1)-connected, then Lipschitz …
We show that for every Lipschitz function f defined on a separable Riemannian manifold M (possibly of infinite dimension), for every continuous ε:M→(0,+∞), and for every positive number r>0, there exists a C∞ smooth Lipschitz function g:M→R such that ∣f(p)−g(p)∣≤ε(p) for every …
Paper finds gender classification accuracy varies by skin type, not ethnicity.
problem Unequal performance of face classification services across skin types and genders.
method Stability experiments, image manipulation, and post-hoc explanation techniques.
result Lip, eye, and cheek structure differences, not skin type, cause gender classification discrepancies.
We prove that the groups of orientation preserving quasiconformal or bilipschitz homeomorphisms of S^n are simple in dimensions 2 and higher.
We relate generalized Lebesgue decompositions of measures in terms of curve fragments (Alberti representations) and Weaver derivations. This correspondence leads to a geometric characterization of the local norm on the Weaver cotangent bundle of a metric measure space (X,μ): the local norm of a form df sees how fas…
This paper determines the flexible exponent for non-geometric 3-manifolds.
problem Bounding the mapping degree in terms of the Lipschitz constant for non-geometric 3-manifolds.
method Analyzing the infimum of α such that the inequality holds for any Lipschitz map.
result The flexible exponent for non-geometric 3-manifolds is determined.
Let M be a PL 2-manifold and X be a compact subpolyhedron of M and let E(X, M) denote the space of embeddings of X into M with the compact-open topology. In this paper we study an extension property of embeddings of X into M and show that the restriction map from the homeomorphism group of M to E(X, M) is a principal b…
Structural-Jump-LSTM speeds up reading by skipping and jumping text.
problem Sequential inference in RNNs makes reading time linearly dependent on input length.
method Introduces a novel LSTM model with agents for skipping and jumping text.
result Structural-Jump-LSTM achieves best FLOP reduction and maintains or improves accuracy.
Solves regularity problem for Lie groups with asymptotic estimate Lie algebras.
problem Regularity problem for Milnor's infinite dimensional Lie groups.
method Analyzes Lie groups with asymptotic estimate Lie algebras and their evolution maps.
result Shows C∞-continuity of evolution map on specific domains. Researchers prove hot spots conjecture for Gaussian spaces.
problem Hot spots conjecture for Gaussian domains.
method Variational principle for Hodge Laplacian on weighted manifolds and Hodge decomposition.
result First nontrivial eigenfunction extrema are on the boundary for specified domains.
We shall give useful criteria of lips, beaks and swallowtail singularities of smooth map from the plane into the plane. As an application of criteria, we will discuss the singularities of Cauchy problem of single conservation law.
Paper uses genome Markov structure for outlier detection and read classification.
problem Identifying outliers and classifying reads in genome databases.
method Applying second-order Markov models to triplet base distributions.
result Improved accuracy in outlier identification and read classification.
XFlow deep neural networks improve audiovisual classification.
problem Multimodal data classification with better feature extraction.
method Cross-modal deep neural networks with dataflow between feature extractors.
result XFlow models achieve better performances than non-cross-modality models.
Semi-supervised deep learning detects problematic reads for genome assembly.
problem De novo genome assembly is hindered by specific types of reads.
method Analysis of coverage graphs converted to 1D-signals using semi-supervised deep learning models.
result Semi-supervised deep learning models can detect problematic reads with minimal labeled data.
We investigate singularities of all parallel surfaces to a given regular surface. In generic context, the types of singularities of parallel surfaces are cuspidal edge, swallowtail, cuspidal lips, cuspidal beaks, cuspidal butterfly and 3-dimensional D4± singularities. We give criteria for these singularities type…
The paper addresses fitting smooth functions to noisy data using a specific class of functions.
problem Fitting smooth functions to noisy data with minimal Lipschitz gradient.
method Provides an algorithm to interpolate a function in C1,1(Rd) with minimal Lipschitz gradient, and estimates the optimal Lipschitz constant. result Uniform bounds relating empirical risk and true risk over the class C1,1(Rd) are obtained. Jack the Reader is a framework for machine reading tasks.
problem Machine reading tasks require reading supporting text.
method A framework for quick model prototyping, evaluation, and integration of new datasets.
result Jack supports three tasks: QA, NLI, and Link Prediction.
Dataset for measuring reading levels in India's children.
problem Measuring reading levels in India's vast population.
method Developed ASER dataset with 5,301 subjects in Hindi, Marathi, and English.
result Achieved 86% accuracy in English language classification.
Improved speech recognition with audio-visual fusion.
problem Enhance speech recognition accuracy in noisy conditions.
method Proposes an attention-based audio-visual fusion strategy to align and learn from acoustic and lip motion data.
result Significant improvements in recognition accuracy (7-30%) on TCD-TIMIT dataset.
This paper sets baselines for reading comprehension benchmarks, finding simple models often perform well.
problem Understanding the difficulty of popular reading comprehension benchmarks.
method Established baselines for bAbI, SQuAD, CBT, CNN, and Who-did-What datasets.
result Simple models often outperform complex models on many benchmarks.
Dead-Direction Signatures (DDS) provide a cheap, closed-form spectral reading of a network's singular complexity.
problem Estimating the complexity of deep networks through their loss singularities.
method DDS replaces the SGLD posterior chain with spectral linear algebra.
result DDS observables rank-track the network's singular complexity at the framework-predicted sign.
Paper explores zero-shot cross-lingual reading comprehension using pre-trained multi-lingual model.
problem Lack of training data for every language in reading comprehension tasks.
method Systematic exploration of zero-shot cross-lingual transfer learning with a multi-lingual language representation model.
result Zero-shot cross-lingual transfer learning is feasible and translating source data into target language is not necessary.
QAInfomax improves reading comprehension by maximizing mutual information, achieving state-of-the-art performance.
problem Distractor sentences in question answering datasets are hard to distinguish from relevant ones.
method QAInfomax regularizes reading comprehension models to learn mutual information among passages, questions, and answers.
result QAInfomax achieves state-of-the-art performance on Adversarial-SQuAD dataset.
A novel conLSH algorithm improves alignment of noisy SMRT reads.
problem High error probability in SMRT sequencing data.
method Context-based Locality Sensitive Hashing (conLSH) for efficient alignment.
result Comprehensive improvement in speed and memory requirements compared to rHAT.
Formalizes interpreting natural language rules for answering questions, collecting 32k task instances.
problem Interpreting regulations and answering 'Can I...?' or 'Do I have to...?' questions.
method Formalization of task, crowd-sourcing strategy to collect 32k instances, analysis of challenges, evaluation of performance.
result Promising results when no background knowledge is needed, substantial room for improvement when background knowledge is needed.
META2 improves taxonomic classification and abundance estimation in metagenomics with deep learning and memory efficiency.
problem Memory constraints and inefficiencies in taxonomic classification and abundance estimation for metagenomics.
method Developed a novel memory-efficient read classification technique combining deep learning and locality-sensitive hashing, and formulated abundance estimation as a Multiple Instance Learning problem.
result Our approach outperforms conventional methods in both single-read taxonomic classification and abundance estimation, especially when memory is limited.
New approach speeds up DNA sequence alignment.
problem Efficiently estimating alignment scores for large sets of reads.
method Rank-one crowdsourcing models and multi-armed bandit algorithm.
result Adaptive algorithm identifies pairs with large alignment scores.
We present a graphical criterion for reading dependencies from the minimal directed independence map G of a graphoid p when G is a polytree and p satisfies composition and weak transitivity. We prove that the criterion is sound and complete. We argue that assuming composition and weak transitivity is not too restrictiv…
A new VAD method uses respiration patterns from video to detect speech.
problem Improving VAD performance in noisy audio recordings.
method Extract respiration patterns from video, use neural models to detect speech.
result Efficacy demonstrated through experiments on real acoustic environments.
New bounds on mapping degrees for geometric 3-manifolds.
problem Bounding the mapping degree in terms of Lipschitz constant for geometric 3-manifolds.
method Constructing Legendrian maps to prove bounds on flexible exponent.
result Complete result for flexible exponent of geometric 3-manifolds.
A metric space (X,d) has the de Groot property GPn if for any points x0,x1,...,xn+2∈X there are positive indices i,j,k≤n+2 such that i=j and d(xi,xj)≤d(x0,xk). If, in addition, k∈{i,j} then X is said to have the Nagata property NPn. It is known that a compact metrizable spac…
Shannon's theory sets limits on information transmission.
problem Limits of information transmission in systems.
method Mathematical theory of communication.
result Defines fundamental limits on information transmission.
Neural network optimizes learning sequence for reading words.
problem Children struggle with learning to read words due to inconsistent spelling-sound correspondences.
method Used a neural network to structure learning trials to optimize generalization accuracy.
result Significant improvement in generalization accuracy compared to random or frequency-based sequences.
We show how the rotation and translation fields of a surface, introduced by G. Darboux, may be used to obtain short proofs of a well-known theorem (that reads that the total mean curvature of a surface is stationary under an infinitesimal bending) and a new theorem (that reads that every infinitesimal flex of any simpl…
This work tackles missing annotations in large sensor datasets.
problem Missing annotations in large sensor datasets negatively affect supervised learning system performance.
method Proposes and evaluates three paradigms to handle gaps: dropping, single label, and unique label, along with a hybrid combination.
result Evaluation of proposed paradigms and hybrid combination shows significant performance improvement.
EMR learns to read and remember from streaming data for QA.
problem Scalable QA from streaming data without knowing questions.
method End-to-end deep network model (EMR) with RL agent for memory management.
result EMR achieves significant improvements over existing methods on synthetic and real-world datasets.