BoningKnife: Joint Entity Mention Detection and Typing for Nested NER via prior Boundary Knowledge

While named entity recognition (NER) is a key task in natural language processing, most approaches only target flat entities, ignoring nested structures which are common in many scenarios. Most existing nested NER methods traverse all sub-sequences which is both expensive and inefficient, and also dont well consider boundary knowledge which is significant for nested entities. In this paper, we propose a joint entity mention detection and typing model via prior boundary knowledge (BoningKnife) to better handle nested NER extraction and recognition tasks. BoningKnife consists of two modules, MentionTagger and TypeClassifier. MentionTagger better leverages boundary knowledge beyond just entity start/end to improve the handling of nesting levels and longer spans, while generating high quality mention candidates. TypeClassifier utilizes a two-level attention mechanism to decouple different nested level representations and better distinguish entity types. We jointly train both modules sharing a common representation and a new dual-info attention layer, which leads to improved representation focus on entity-related information. Experiments over different datasets show that our approach outperforms previous state of the art methods and achieves 86.41, 85.46, and 94.2 F1 scores on ACE2004, ACE2005, and NNE, respectively.

Chaotic diffusion in multi-scale turbulence

Turbulence plays a very important role in determining the transport of energy and particles in tokamaks. This work is devoted to studying the chaotic diffusion in multi-scale turbulence in the context of the nonlinear wave-particle interaction. Turbulent waves with different scales of characteristic wavelengths can interact with the same group of charged particles when their phase velocity is close to the velocities of the charged particles. A multi-wavenumber standard mapping is developed to model the chaotic diffusion in multi-scale turbulence. The diffusion coefficient is obtained by calculating the correlation functions analytically. It is found that the contribution of the largest scale turbulence dominates the deviation from the quasi-linear diffusion coefficient. Increasing the overlap parameters of the smaller scale turbulence by just the increasing the wavenumber cannot make the diffusion coefficient to be the quasi-linear diffusion coefficient for a finite wave amplitude. Especially, in two-scale turbulence, the diffusion coefficient is mostly over the quasi-linear diffusion coefficient in the large wavenumber (of the smaller scale turbulence) limit. As more scales of components are added in the turbulence, the diffusion coefficient approaches the quasi-linear diffusion coefficient. The results can also be applied to other resonance-induced multi-scale turbulence in Hamiltonian systems with 1.5 or 2 degrees of freedom.

AdvPicker: Effectively Leveraging Unlabeled Data via Adversarial Discriminator for Cross-Lingual NER

Neural methods have been shown to achieve high performance in Named Entity Recognition (NER), but rely on costly high-quality labeled data for training, which is not always available across languages. While previous works have shown that unlabeled data in a target language can be used to improve cross-lingual model performance, we propose a novel adversarial approach (AdvPicker) to better leverage such data and further improve results. We design an adversarial learning framework in which an encoder learns entity domain knowledge from labeled source-language data and better shared features are captured via adversarial training - where a discriminator selects less language-dependent target-language data via similarity to the source language. Experimental results on standard benchmark datasets well demonstrate that the proposed method benefits strongly from this data selection process and outperforms existing state-of-the-art methods; without requiring any additional external resources (e.g., gazetteers or via machine translation). The code is available at https://aka.ms/AdvPicker

High-fidelity Face Tracking for AR/VR via Deep Lighting Adaptation

3D video avatars can empower virtual communications by providing compression, privacy, entertainment, and a sense of presence in AR/VR. Best 3D photo-realistic AR/VR avatars driven by video, that can minimize uncanny effects, rely on person-specific models. However, existing person-specific photo-realistic 3D models are not robust to lighting, hence their results typically miss subtle facial behaviors and cause artifacts in the avatar. This is a major drawback for the scalability of these models in communication systems (e.g., Messenger, Skype, FaceTime) and AR/VR. This paper addresses previous limitations by learning a deep learning lighting model, that in combination with a high-quality 3D face tracking algorithm, provides a method for subtle and robust facial motion transfer from a regular video to a 3D photo-realistic avatar. Extensive experimental validation and comparisons to other state-of-the-art methods demonstrate the effectiveness of the proposed framework in real-world scenarios with variability in pose, expression, and illumination. Please visit https://www.youtube.com/watch?v=dtz1LgZR8cc for more results. Our project page can be found at https://www.cs.rochester.edu/u/lchen63.

Equal Affection or Random Selection: the Quality of Subjective Feedback from a Group Perspective

In the setting where a group of agents is asked a single subjective multi-choice question (e.g. which one do you prefer? cat or dog?), we are interested in evaluating the quality of the collected feedback. However, the collected statistics are not sufficient to reflect how informative the feedback is since fully informative feedback (equal affection of the choices) and fully uninformative feedback (random selection) have the same uniform statistics. Here we distinguish the above two scenarios by additionally asking for respondents predictions about others choices. We assume that informative respondents predictions strongly depend on their own choices while uninformative respondents do not. With this assumption, we propose a new definition for uninformative feedback and correspondingly design a family of evaluation metrics, called f-variety, for group-level feedback which can 1) distinguish informative feedback and uninformative feedback (separation) even if their statistics are both uniform and 2) decrease as the ratio of uninformative respondents increases (monotonicity). We validate our approach both theoretically and numerically. Moreover, we conduct two real-world case studies about 1) comparisons about athletes and 2) comparisons about stand-up comedians to show the superiority of our approach.

Weakly Supervised Learning with Side Information for Noisy Labeled Images

In many real-world datasets, like WebVision, the performance of DNN based classifier is often limited by the noisy labeled data. To tackle this problem, some image related side information, such as captions and tags, often reveal underlying relationships across images. In this paper, we present an efficient weakly supervised learning by using a Side Information Network (SINet), which aims to effectively carry out a large scale classification with severely noisy labels. The proposed SINet consists of a visual prototype module and a noise weighting module. The visual prototype module is designed to generate a compact representation for each category by introducing the side information. The noise weighting module aims to estimate the correctness of each noisy image and produce a confidence score for image ranking during the training procedure. The propsed SINet can largely alleviate the negative impact of noisy image labels, and is beneficial to train a high performance CNN based classifier. Besides, we released a fine-grained product dataset called AliProducts, which contains more than 2.5 million noisy web images crawled from the internet by using queries generated from 50,000 fine-grained semantic classes. Extensive experiments on several popular benchmarks (i.e. Webvision, ImageNet and Clothing-1M) and our proposed AliProducts achieve state-of-the-art performance. The SINet has won the first place in the classification task on WebVision Challenge 2019, and outperformed other competitors by a large margin.

Central limit theorems for spatial averages of the stochastic heat equation via Malliavin-Steins method

Suppose that ${u(t,, x)}_{t >0, x inmathbb{R}^d}$ is the solution to a $d$-dimensional stochastic heat equation driven by a Gaussian noise that is white in time and has a spatially homogeneous covariance that satisfies Dalangs condition. The purpose of this paper is to establish quantitative central limit theorems for spatial averages of the form $N^{-d} int_{[0,N]^d} g(u(t,,x)), mathrm{d} x$, as $Nrightarrowinfty$, where $g$ is a Lipschitz-continuous function or belongs to a class of locally-Lipschitz functions, using a combination of the Malliavin calculus and Steins method for normal approximations. Our results include a central limit theorem for the {it Hopf-Cole} solution to KPZ equation. We also establish a functional central limit theorem for these spatial averages.

Talking-head Generation with Rhythmic Head Motion

When people deliver a speech, they naturally move heads, and this rhythmic head motion conveys prosodic information. However, generating a lip-synced video while moving head naturally is challenging. While remarkably successful, existing works either generate still talkingface videos or rely on landmark/video frames as sparse/dense mapping guidance to generate head movements, which leads to unrealistic or uncontrollable video synthesis. To overcome the limitations, we propose a 3D-aware generative network along with a hybrid embedding module and a non-linear composition module. Through modeling the head motion and facial expressions1 explicitly, manipulating 3D animation carefully, and embedding reference images dynamically, our approach achieves controllable, photo-realistic, and temporally coherent talking-head videos with natural head movements. Thoughtful experiments on several standard benchmarks demonstrate that our method achieves significantly better results than the state-of-the-art methods in both quantitative and qualitative comparisons. The code is available on https://github.com/ lelechen63/Talking-head-Generation-with-Rhythmic-Head-Motion.

Spatial ergodicity and central limit theorems for parabolic Anderson model with delta initial condition

Let ${u(t,, x)}_{t >0, x inmathbb{R}}$ denote the solution to the parabolic Anderson model with initial condition $delta_0$ and driven by space-time white noise on $mathbb{R}_+timesmathbb{R}$, and let $bm{p}_t(x):= (2pi t)^{-1/2}exp{-x^2/(2t)}$ denote the standard Gaussian heat kernel on the line. We use a non-trivial adaptation of the methods in our companion papers cite{CKNP,CKNP_b} in order to prove that the random field $xmapsto u(t,,x)/bm{p}_t(x)$ is ergodic for every $t >0$. And we establish an associated quantitative central limit theorem following the approach based on the Malliavin-Stein method introduced in Huang, Nualart, and Viitasaari cite{HNV2018}.

A CLT for dependent random variables, with an application to an infinite system of interacting diffusion processes

We present a central limit theorem for stationary random fields that are short-range dependent and asymptotically independent. As an application, we present a central limit theorem for an infinite family of interacting It^o-type diffusion processes.