arXiv subjects

Yue Yang

Publications and source records attributed to Yue Yang.

At least 163 records · Page 9Linked to original sources

arXiv2023

Align, Adapt and Inject: Sound-guided Unified Image Generation

Text-guided image generation has witnessed unprecedented progress due to the development of diffusion models. Beyond text and image, sound is a vital element within the sphere of human perception, offering vivid representations and naturally coinciding with corresponding scenes. Taking advantage of sound therefore presents a promising avenue for exploration within image generation research. However, the relationship between audio and image supervision remains significantly underdeveloped, and the scarcity of related, high-quality datasets brings further obstacles. In this paper, we propose a unified framework 'Align, Adapt, and Inject' (AAI) for sound-guided image generation, editing, and stylization. In particular, our method adapts input sound into a sound token, like an ordinary word, which can plug and play with existing powerful diffusion-based Text-to-Image (T2I) models. Specifically, we first train a multi-modal encoder to align audio representation with the pre-trained textual manifold and visual manifold, respectively. Then, we propose the audio adapter to adapt audio representation into an audio token enriched with specific semantics, which can be injected into a frozen T2I model flexibly. In this way, we are able to extract the dynamic information of varied sounds, while utilizing the formidable capability of existing T2I models to facilitate sound-guided image generation, editing, and stylization in a convenient and cost-effective manner. The experiment results confirm that our proposed AAI outperforms other text and sound-guided state-of-the-art methods. And our aligned multi-modal encoder is also competitive with other approaches in the audio-visual retrieval and audio-text retrieval tasks.

cs.GR↗

arXiv2023

Modeling the boundary-layer flashback of premixed hydrogen-enriched swirling flames at high pressures

We model the boundary-layer flashback (BLF) of CH$_4$/H$_2$/air swirling flames via large-eddy simulations with the flame-surface-density method (LES-FSD), in particular, at high pressures. A local displacement speed model tabulating the stretched flame speed is employed to account for the thermo-diffusive effects, flame surface curvature, and heat loss in LES-FSD. The LES-FSD well captures the propagation characteristics during the BLF of swirling flames. In the LES-FSD for lean CH$_4$/H$_2$/air flames at 2.5 bar, the critical equivalence ratio for flashback decreases with the increasing hydrogen volume fraction, consistent with the experiments. This is due to the improved modeling of effects of the flame stretch and heat loss on the local displacement speed. We also develop a simple model to predict the BLF limits of swirling flames. The model estimates the critical bulk velocity for given reactants and swirl number, via the balance between the flame-induced pressure rise and adverse pressure for boundary-layer separation. We validate the model against 14 datasets of CH$_4$/H$_2$/air swirling flame experiments, with the hydrogen volume fractions in fuel from 50% to 100%. The present model well estimates the flashback limits in various operating conditions.

physics.flu-dyn↗

arXiv2023

Language in a Bottle: Language Model Guided Concept Bottlenecks for Interpretable Image Classification

Concept Bottleneck Models (CBM) are inherently interpretable models that factor model decisions into human-readable concepts. They allow people to easily understand why a model is failing, a critical feature for high-stakes applications. CBMs require manually specified concepts and often under-perform their black box counterparts, preventing their broad adoption. We address these shortcomings and are first to show how to construct high-performance CBMs without manual specification of similar accuracy to black box models. Our approach, Language Guided Bottlenecks (LaBo), leverages a language model, GPT-3, to define a large space of possible bottlenecks. Given a problem domain, LaBo uses GPT-3 to produce factual sentences about categories to form candidate concepts. LaBo efficiently searches possible bottlenecks through a novel submodular utility that promotes the selection of discriminative and diverse information. Ultimately, GPT-3's sentential concepts can be aligned to images using CLIP, to form a bottleneck layer. Experiments demonstrate that LaBo is a highly effective prior for concepts important to visual recognition. In the evaluation with 11 diverse datasets, LaBo bottlenecks excel at few-shot classification: they are 11.7% more accurate than black box linear probes at 1 shot and comparable with more data. Overall, LaBo demonstrates that inherently interpretable models can be widely applied at similar, or better, performance than black box approaches.

cs.CV↗

arXiv2023

Semi-Supervised 2D Human Pose Estimation Driven by Position Inconsistency Pseudo Label Correction Module

In this paper, we delve into semi-supervised 2D human pose estimation. The previous method ignored two problems: (i) When conducting interactive training between large model and lightweight model, the pseudo label of lightweight model will be used to guide large models. (ii) The negative impact of noise pseudo labels on training. Moreover, the labels used for 2D human pose estimation are relatively complex: keypoint category and keypoint position. To solve the problems mentioned above, we propose a semi-supervised 2D human pose estimation framework driven by a position inconsistency pseudo label correction module (SSPCM). We introduce an additional auxiliary teacher and use the pseudo labels generated by the two teacher model in different periods to calculate the inconsistency score and remove outliers. Then, the two teacher models are updated through interactive training, and the student model is updated using the pseudo labels generated by two teachers. To further improve the performance of the student model, we use the semi-supervised Cut-Occlude based on pseudo keypoint perception to generate more hard and effective samples. In addition, we also proposed a new indoor overhead fisheye human keypoint dataset WEPDTOF-Pose. Extensive experiments demonstrate that our method outperforms the previous best semi-supervised 2D human pose estimation method. We will release the code and dataset at https://github.com/hlz0606/SSPCM.

cs.CV↗

arXiv2023

Safe Inverse Reinforcement Learning via Control Barrier Function

Learning from Demonstration (LfD) is a powerful method for enabling robots to perform novel tasks as it is often more tractable for a non-roboticist end-user to demonstrate the desired skill and for the robot to efficiently learn from the associated data than for a human to engineer a reward function for the robot to learn the skill via reinforcement learning (RL). Safety issues arise in modern LfD techniques, e.g., Inverse Reinforcement Learning (IRL), just as they do for RL; yet, safe learning in LfD has received little attention. In the context of agile robots, safety is especially vital due to the possibility of robot-environment collision, robot-human collision, and damage to the robot. In this paper, we propose a safe IRL framework, CBFIRL, that leverages the Control Barrier Function (CBF) to enhance the safety of the IRL policy. The core idea of CBFIRL is to combine a loss function inspired by CBF requirements with the objective in an IRL method, both of which are jointly optimized via gradient descent. In the experiments, we show our framework performs safer compared to IRL methods without CBF, that is $\sim15\%$ and $\sim20\%$ improvement for two levels of difficulty of a 2D racecar domain and $\sim 50\%$ improvement for a 3D drone domain.

cs.RO↗

arXiv2023

Watermarking in Secure Federated Learning: A Verification Framework Based on Client-Side Backdooring

Federated learning (FL) allows multiple participants to collaboratively build deep learning (DL) models without directly sharing data. Consequently, the issue of copyright protection in FL becomes important since unreliable participants may gain access to the jointly trained model. Application of homomorphic encryption (HE) in secure FL framework prevents the central server from accessing plaintext models. Thus, it is no longer feasible to embed the watermark at the central server using existing watermarking schemes. In this paper, we propose a novel client-side FL watermarking scheme to tackle the copyright protection issue in secure FL with HE. To our best knowledge, it is the first scheme to embed the watermark to models under the Secure FL environment. We design a black-box watermarking scheme based on client-side backdooring to embed a pre-designed trigger set into an FL model by a gradient-enhanced embedding method. Additionally, we propose a trigger set construction mechanism to ensure the watermark cannot be forged. Experimental results demonstrate that our proposed scheme delivers outstanding protection performance and robustness against various watermark removal attacks and ambiguity attack.

cs.CR↗

arXiv2023

Quantum computing of fluid dynamics using the hydrodynamic Schrödinger equation

Simulating fluid dynamics on a quantum computer is intrinsically difficult due to the nonlinear and non-Hamiltonian nature of the Navier-Stokes equation (NSE). We propose a framework for quantum computing of fluid dynamics based on the hydrodynamic Schrödinger equation (HSE), which can be promising in simulating three-dimensional turbulent flows in various engineering applications. The HSE is derived by generalizing the Madelung transform to compressible/incompressible flows with finite vorticity and dissipation. Since the HSE is expressed as a unitary operator on a two-component wave function, it is more suitable than the NSE for quantum computing. The flow governed by the HSE can resemble a turbulent flow consisting of tangled vortex tubes with the five-thirds scaling of energy spectrum. We develop a prediction-correction quantum algorithm to solve the HSE. This algorithm is implemented for simple flows on the quantum simulator Qiskit with exponential speedup.

physics.flu-dyn↗

arXiv2023

Causal Reasoning of Entities and Events in Procedural Texts

Entities and events are crucial to natural language reasoning and common in procedural texts. Existing work has focused either exclusively on entity state tracking (e.g., whether a pan is hot) or on event reasoning (e.g., whether one would burn themselves by touching the pan), while these two tasks are often causally related. We propose CREPE, the first benchmark on causal reasoning of event plausibility and entity states. We show that most language models, including GPT-3, perform close to chance at .35 F1, lagging far behind human at .87 F1. We boost model performance to .59 F1 by creatively representing events as programming languages while prompting language models pretrained on code. By injecting the causal relations between entities and events as intermediate reasoning steps in our representation, we further boost the performance to .67 F1. Our findings indicate not only the challenge that CREPE brings for language models, but also the efficacy of code-like prompting combined with chain-of-thought prompting for multihop event reasoning.

cs.CL↗

arXiv2023

An optimized online filter stack spectrometer

The spectrum of laser-plasma-generated X-rays is very important as it can characterize electron dynamics and also be useful for applications, and nowadays with the forthcoming high-repetition-rate laser-plasma experiments, there is a raising demand for online diagnosis for the X-ray spectrum. In this paper, scintillators and silicon PIN diodes are used to build a wideband online filter stack spectrometer. The genetic algorithm is used to optimize the arrangements of the X-ray sensors and filters by minimizing the condition number of the response matrix, thus the unfolding error can be significantly decreased according to the numerical experiments. The detector responses are quantitatively calibrated by irradiating the scintillator and PIN diode using different nuclides and comparing the measured gamma-ray peaks. Finally, a 15-channel spectrometer prototype has been implemented. The X-ray detector, front-end electronics, and back-end electronics are integrated into the prototype, and the prototype can determine the spectrum with 1 kHz repetition rates.

physics.ins-det↗

arXiv2022

On the Nonexistence of a Strong Minimal Pair

Two nonzero recursively enumerable (r.e.) degrees $\mathbf{a}$ and $\mathbf{b}$ form a strong minimal pair if $\mathbf{a} \wedge \mathbf{b}=\mathbf{0}$ and $\mathbf{b}\vee \mathbf{x}\geq \mathbf{a}$ for any nonzero r.e. degree $\mathbf{x}\leq \mathbf{a}$. We prove that there is no strong minimal pair in the r.e. degrees. Our construction goes beyond the usual $\mathbf{0}'''$-priority arguments and we give some evidence to show that it needs $\mathbf{0}^{(4)}$-priority arguments.

math.LO↗

arXiv2022

Visualizing the Obvious: A Concreteness-based Ensemble Model for Noun Property Prediction

Neural language models encode rich knowledge about entities and their relationships which can be extracted from their representations using probing. Common properties of nouns (e.g., red strawberries, small ant) are, however, more challenging to extract compared to other types of knowledge because they are rarely explicitly stated in texts. We hypothesize this to mainly be the case for perceptual properties which are obvious to the participants in the communication. We propose to extract these properties from images and use them in an ensemble model, in order to complement the information that is extracted from language models. We consider perceptual properties to be more concrete than abstract properties (e.g., interesting, flawless). We propose to use the adjectives' concreteness score as a lever to calibrate the contribution of each source (text vs. images). We evaluate our ensemble model in a ranking task where the actual properties of a noun need to be ranked higher than other non-relevant properties. Our results show that the proposed combination of text and images greatly improves noun property prediction compared to powerful text-based language models.

cs.CL↗

arXiv2022

Z-LaVI: Zero-Shot Language Solver Fueled by Visual Imagination

Large-scale pretrained language models have made significant advances in solving downstream language understanding tasks. However, they generally suffer from reporting bias, the phenomenon describing the lack of explicit commonsense knowledge in written text, e.g., ''an orange is orange''. To overcome this limitation, we develop a novel approach, Z-LaVI, to endow language models with visual imagination capabilities. Specifically, we leverage two complementary types of ''imaginations'': (i) recalling existing images through retrieval and (ii) synthesizing nonexistent images via text-to-image generation. Jointly exploiting the language inputs and the imagination, a pretrained vision-language model (e.g., CLIP) eventually composes a zero-shot solution to the original language tasks. Notably, fueling language models with imagination can effectively leverage visual knowledge to solve plain language tasks. In consequence, Z-LaVI consistently improves the zero-shot performance of existing language models across a diverse set of language tasks.

cs.CL↗

arXiv2022

Do Deep Neural Networks Always Perform Better When Eating More Data?

Data has now become a shortcoming of deep learning. Researchers in their own fields share the thinking that "deep neural networks might not always perform better when they eat more data," which still lacks experimental validation and a convincing guiding theory. Here to fill this lack, we design experiments from Identically Independent Distribution(IID) and Out of Distribution(OOD), which give powerful answers. For the purpose of guidance, based on the discussion of results, two theories are proposed: under IID condition, the amount of information determines the effectivity of each sample, the contribution of samples and difference between classes determine the amount of sample information and the amount of class information; under OOD condition, the cross-domain degree of samples determine the contributions, and the bias-fitting caused by irrelevant elements is a significant factor of cross-domain. The above theories provide guidance from the perspective of data, which can promote a wide range of practical applications of artificial intelligence.

cs.LG↗

arXiv2022

Show Me More Details: Discovering Hierarchies of Procedures from Semi-structured Web Data

Procedures are inherently hierarchical. To "make videos", one may need to "purchase a camera", which in turn may require one to "set a budget". While such hierarchical knowledge is critical for reasoning about complex procedures, most existing work has treated procedures as shallow structures without modeling the parent-child relation. In this work, we attempt to construct an open-domain hierarchical knowledge-base (KB) of procedures based on wikiHow, a website containing more than 110k instructional articles, each documenting the steps to carry out a complex procedure. To this end, we develop a simple and efficient method that links steps (e.g., "purchase a camera") in an article to other articles with similar goals (e.g., "how to choose a camera"), recursively constructing the KB. Our method significantly outperforms several strong baselines according to automatic evaluation, human judgment, and application to downstream tasks such as instructional video retrieval. A demo with partial data can be found at https://wikihow-hierarchy.github.io. The code and the data are at https://github.com/shuyanzhou/wikihow_hierarchy.

cs.CL↗

arXiv2022

Foreground Object Structure Transfer for Unsupervised Domain Adaptation

Unsupervised domain adaptation aims to train a classification model from the labeled source domain for the unlabeled target domain. Since the data distributions of the two domains are different, the model often performs poorly on the target domain. Existing methods align the feature distributions of the source and target domains and learn domain-invariant features to improve the performance of the model. However, the features are usually aligned as a whole, and the domain adaptation task fails to serve the classification, which will ignore the class information and lead to misalignment.In this paper, we investigate those features that should be used for domain alignment, introduce prior knowledge to extract foreground features to guide the domain adaptation task for classification tasks, and perform alignment in the local structure of objects. We propose a method called Foreground Object Structure Transfer(FOST). The key to FOST is the new clustering based condition, which combines the relative position relationship of foreground objects. Based on this conditions, FOST makes the data distribution of the same class more compact in geometry. In practice, since the label of the target domain is not available, we use the clustering information of the source domain to assign pseudo labels to the target domain samples, and then according to the source domain data prior knowledge guides those positive features to maximum the inter-class distance between different classes and mimimum the intra-class distance. Extensive experimental results on various benchmarks ($i.e.$ ImageCLEF-DA, Office-31, Office-Home, Visda-2017) under different domain adaptation settings prove that our FOST compares favorably against the existing state-of-the-art domain adaptation methods.

cs.CV↗

arXiv2021

Induce, Edit, Retrieve: Language Grounded Multimodal Schema for Instructional Video Retrieval

Schemata are structured representations of complex tasks that can aid artificial intelligence by allowing models to break down complex tasks into intermediate steps. We propose a novel system that induces schemata from web videos and generalizes them to capture unseen tasks with the goal of improving video retrieval performance. Our system proceeds in three major phases: (1) Given a task with related videos, we construct an initial schema for a task using a joint video-text model to match video segments with text representing steps from wikiHow; (2) We generalize schemata to unseen tasks by leveraging language models to edit the text within existing schemata. Through generalization, we can allow our schemata to cover a more extensive range of tasks with a small amount of learning data; (3) We conduct zero-shot instructional video retrieval with the unseen task names as the queries. Our schema-guided approach outperforms existing methods for video retrieval, and we demonstrate that the schemata induced by our system are better than those generated by other models.

cs.CV↗

arXiv2021

Conservation Strength of The Infinite Pigeonhole Principle for Trees

Let $\mathsf{TT}^1$ be the combinatorial principle stating that every finite coloring of the infinite full binary tree has a homogeneous isomorphic subtree. Let $\mathsf{RT}^2_2$ and $\mathsf{WKL}_0$ denote respectively the principles of Ramsey's theorem for pairs and weak König's lemma. It is proved that $\mathsf{TT}^1+\mathsf{RT}^2_2+\mathsf{WKL}_0$ is $Π^0_3$-conservative over the base system $\mathsf{RCA}_0$. Thus over $\mathsf{RCA}_0$, $\mathsf{TT}^1$ and Ramsey's theorem for pairs prove the same $Π^0_3$-sentences.

math.LO↗

arXiv2021

Visual Goal-Step Inference using wikiHow

Understanding what sequence of steps are needed to complete a goal can help artificial intelligence systems reason about human activities. Past work in NLP has examined the task of goal-step inference for text. We introduce the visual analogue. We propose the Visual Goal-Step Inference (VGSI) task, where a model is given a textual goal and must choose which of four images represents a plausible step towards that goal. With a new dataset harvested from wikiHow consisting of 772,277 images representing human actions, we show that our task is challenging for state-of-the-art multimodal models. Moreover, the multimodal representation learned from our data can be effectively transferred to other datasets like HowTo100m, increasing the VGSI accuracy by 15 - 20%. Our task will facilitate multimodal reasoning about procedural events.

cs.CV↗