arXiv subjects

Xinyu Hu

Publications and source records attributed to Xinyu Hu.

54 records · Page 3Linked to original sources

arXiv2024

What You See Is What Matters: A Novel Visual and Physics-Based Metric for Evaluating Video Generation Quality

As video generation models advance rapidly, assessing the quality of generated videos has become increasingly critical. Existing metrics, such as Fréchet Video Distance (FVD), Inception Score (IS), and ClipSim, measure quality primarily in latent space rather than from a human visual perspective, often overlooking key aspects like appearance and motion consistency to physical laws. In this paper, we propose a novel metric, VAMP (Visual Appearance and Motion Plausibility), that evaluates both the visual appearance and physical plausibility of generated videos. VAMP is composed of two main components: an appearance score, which assesses color, shape, and texture consistency across frames, and a motion score, which evaluates the realism of object movements. We validate VAMP through two experiments: corrupted video evaluation and generated video evaluation. In the corrupted video evaluation, we introduce various types of corruptions into real videos and measure the correlation between corruption severity and VAMP scores. In the generated video evaluation, we use state-of-the-art models to generate videos from carefully designed prompts and compare VAMP's performance to human evaluators' rankings. Our results demonstrate that VAMP effectively captures both visual fidelity and temporal consistency, offering a more comprehensive evaluation of video quality than traditional methods.

cs.CV↗

arXiv2024

MC-MKE: A Fine-Grained Multimodal Knowledge Editing Benchmark Emphasizing Modality Consistency

Multimodal large language models (MLLMs) are prone to non-factual or outdated knowledge issues, which can manifest as misreading and misrecognition errors due to the complexity of multimodal knowledge. Previous benchmarks have not systematically analyzed the performance of editing methods in correcting these two error types. To better represent and correct these errors, we decompose multimodal knowledge into its visual and textual components. Different error types correspond to different editing formats, which edit distinct parts of the multimodal knowledge. We present MC-MKE, a fine-grained Multimodal Knowledge Editing benchmark emphasizing Modality Consistency. Our benchmark facilitates independent correction of misreading and misrecognition errors by editing the corresponding knowledge component. We evaluate four multimodal knowledge editing methods on MC-MKE, revealing their limitations, particularly in terms of modality consistency. Our work highlights the challenges posed by multimodal knowledge editing and motivates further research in developing effective techniques for this task.

cs.CV↗

arXiv2024

New bounds of two hypergraph Ramsey problems

We focus on two hypergraph Ramsey problems. First, we consider the Erdős-Hajnal function $r_k(k+1,t;n)$. In 1972, Erdős and Hajnal conjectured that the tower growth rate of $r_k(k+1,t;n)$ is $t-1$ for each $2\le t\le k$. To finish this conjecture, it remains to show that the tower growth rate of $r_4(5,4;n)$ is three. We prove a superexponential lower bound for $r_4(5,4;n)$, which improves the previous best lower bound $r_4(5,4;n)\geq 2^{Ω(n^2)}$ from Mubayi and Suk (\emph{J. Eur. Math. Soc., 2020}). Second, we prove an upper bound for the hypergraph Erdős-Rogers function $f^{(k)}_{k+1,k+2}(N)$ that is an iterated $(k-3)$-fold logarithm in $N$ for each $k\geq 5$. This improves the previous upper bound that is an iterated $(k-13)$-fold logarithm in $N$ for $k\ge14$ due to Mubayi and Suk (\emph{J. London Math. Soc., 2018}), in which they conjectured that $f^{(k)}_{k+1,k+2}(N)$ is an iterated $(k-2)$-fold logarithm in $N$ for each $k\ge3$.

math.CO↗

arXiv2024

Themis: A Reference-free NLG Evaluation Language Model with Flexibility and Interpretability

The evaluation of natural language generation (NLG) tasks is a significant and longstanding research area. With the recent emergence of powerful large language models (LLMs), some studies have turned to LLM-based automatic evaluation methods, which demonstrate great potential to become a new evaluation paradigm following traditional string-based and model-based metrics. However, despite the improved performance of existing methods, they still possess some deficiencies, such as dependency on references and limited evaluation flexibility. Therefore, in this paper, we meticulously construct a large-scale NLG evaluation corpus NLG-Eval with annotations from both human and GPT-4 to alleviate the lack of relevant data in this field. Furthermore, we propose Themis, an LLM dedicated to NLG evaluation, which has been trained with our designed multi-perspective consistency verification and rating-oriented preference alignment methods. Themis can conduct flexible and interpretable evaluations without references, and it exhibits superior evaluation performance on various NLG tasks, simultaneously generalizing well to unseen tasks and surpassing other evaluation models, including GPT-4.

cs.CL↗

arXiv2024

SMART-RAG: Selection using Determinantal Matrices for Augmented Retrieval

Retrieval-Augmented Generation (RAG) has greatly improved large language models (LLMs) by enabling them to generate accurate, contextually grounded responses through the integration of external information. However, conventional RAG approaches, which prioritize top-ranked documents based solely on query-context relevance, often introduce redundancy and conflicting information. This issue is particularly evident in unsupervised retrieval settings, where there are no mechanisms to effectively mitigate these problems, leading to suboptimal context selection. To address this, we propose Selection using Matrices for Augmented Retrieval (SMART) in question answering tasks, a fully unsupervised and training-free framework designed to optimize context selection in RAG. SMART leverages Determinantal Point Processes (DPPs) to simultaneously model relevance, diversity and conflict, ensuring the selection of potentially high-quality contexts. Experimental results across multiple datasets demonstrate that SMART significantly enhances QA performance and surpasses previous unsupervised context selection methods, showing a promising strategy for RAG.

cs.CL↗

arXiv2024

Are LLM-based Evaluators Confusing NLG Quality Criteria?

Some prior work has shown that LLMs perform well in NLG evaluation for different tasks. However, we discover that LLMs seem to confuse different evaluation criteria, which reduces their reliability. For further verification, we first consider avoiding issues of inconsistent conceptualization and vague expression in existing NLG quality criteria themselves. So we summarize a clear hierarchical classification system for 11 common aspects with corresponding different criteria from previous studies involved. Inspired by behavioral testing, we elaborately design 18 types of aspect-targeted perturbation attacks for fine-grained analysis of the evaluation behaviors of different LLMs. We also conduct human annotations beyond the guidance of the classification system to validate the impact of the perturbations. Our experimental results reveal confusion issues inherent in LLMs, as well as other noteworthy phenomena, and necessitate further research and improvements for LLM-based evaluation.

cs.CL↗

arXiv2024

Task Oriented In-Domain Data Augmentation

Large Language Models (LLMs) have shown superior performance in various applications and fields. To achieve better performance on specialized domains such as law and advertisement, LLMs are often continue pre-trained on in-domain data. However, existing approaches suffer from two major issues. First, in-domain data are scarce compared with general domain-agnostic data. Second, data used for continual pre-training are not task-aware, such that they may not be helpful to downstream applications. We propose TRAIT, a task-oriented in-domain data augmentation framework. Our framework is divided into two parts: in-domain data selection and task-oriented synthetic passage generation. The data selection strategy identifies and selects a large amount of in-domain data from general corpora, and thus significantly enriches domain knowledge in the continual pre-training data. The synthetic passages contain guidance on how to use domain knowledge to answer questions about downstream tasks. By training on such passages, the model aligns with the need of downstream applications. We adapt LLMs to two domains: advertisement and math. On average, TRAIT improves LLM performance by 8% in the advertisement domain and 7.5% in the math domain.

cs.CL↗

arXiv2024

Solving Inverse Problems with Latent Diffusion Models via Hard Data Consistency

Diffusion models have recently emerged as powerful generative priors for solving inverse problems. However, training diffusion models in the pixel space are both data-intensive and computationally demanding, which restricts their applicability as priors for high-dimensional real-world data such as medical images. Latent diffusion models, which operate in a much lower-dimensional space, offer a solution to these challenges. However, incorporating latent diffusion models to solve inverse problems remains a challenging problem due to the nonlinearity of the encoder and decoder. To address these issues, we propose \textit{ReSample}, an algorithm that can solve general inverse problems with pre-trained latent diffusion models. Our algorithm incorporates data consistency by solving an optimization problem during the reverse sampling process, a concept that we term as hard data consistency. Upon solving this optimization problem, we propose a novel resampling scheme to map the measurement-consistent sample back onto the noisy data manifold and theoretically demonstrate its benefits. Lastly, we apply our algorithm to solve a wide range of linear and nonlinear inverse problems in both natural and medical images, demonstrating that our approach outperforms existing state-of-the-art approaches, including those based on pixel-space diffusion models.

cs.CV↗

arXiv2024

Error-Robust Retrieval for Chinese Spelling Check

Chinese Spelling Check (CSC) aims to detect and correct error tokens in Chinese contexts, which has a wide range of applications. However, it is confronted with the challenges of insufficient annotated data and the issue that previous methods may actually not fully leverage the existing datasets. In this paper, we introduce our plug-and-play retrieval method with error-robust information for Chinese Spelling Check (RERIC), which can be directly applied to existing CSC models. The datastore for retrieval is built completely based on the training data, with elaborate designs according to the characteristics of CSC. Specifically, we employ multimodal representations that fuse phonetic, morphologic, and contextual information in the calculation of query and key during retrieval to enhance robustness against potential errors. Furthermore, in order to better judge the retrieved candidates, the n-gram surrounding the token to be checked is regarded as the value and utilized for specific reranking. The experiment results on the SIGHAN benchmarks demonstrate that our proposed method achieves substantial improvements over existing work.

cs.CL↗

arXiv2023

A Survey of Knowledge Enhanced Pre-trained Models

Pre-trained language models learn informative word representations on a large-scale text corpus through self-supervised learning, which has achieved promising performance in fields of natural language processing (NLP) after fine-tuning. These models, however, suffer from poor robustness and lack of interpretability. We refer to pre-trained language models with knowledge injection as knowledge-enhanced pre-trained language models (KEPLMs). These models demonstrate deep understanding and logical reasoning and introduce interpretability. In this survey, we provide a comprehensive overview of KEPLMs in NLP. We first discuss the advancements in pre-trained language models and knowledge representation learning. Then we systematically categorize existing KEPLMs from three different perspectives. Finally, we outline some potential directions of KEPLMs for future research.

cs.CL↗

arXiv2023

Evoke: Evoking Critical Thinking Abilities in LLMs via Reviewer-Author Prompt Editing

Large language models (LLMs) have made impressive progress in natural language processing. These models rely on proper human instructions (or prompts) to generate suitable responses. However, the potential of LLMs are not fully harnessed by commonly-used prompting methods: many human-in-the-loop algorithms employ ad-hoc procedures for prompt selection; while auto prompt generation approaches are essentially searching all possible prompts randomly and inefficiently. We propose Evoke, an automatic prompt refinement framework. In Evoke, there are two instances of a same LLM: one as a reviewer (LLM-Reviewer), it scores the current prompt; the other as an author (LLM-Author), it edits the prompt by considering the edit history and the reviewer's feedback. Such an author-reviewer feedback loop ensures that the prompt is refined in each iteration. We further aggregate a data selection approach to Evoke, where only the hard samples are exposed to the LLM. The hard samples are more important because the LLM can develop deeper understanding of the tasks out of them, while the model may already know how to solve the easier cases. Experimental results show that Evoke significantly outperforms existing methods. For instance, in the challenging task of logical fallacy detection, Evoke scores above 80, while all other baseline methods struggle to reach 20.

cs.CL↗

arXiv2023

DeepTagger: Knowledge Enhanced Named Entity Recognition for Web-Based Ads Queries

Named entity recognition (NER) is a crucial task for online advertisement. State-of-the-art solutions leverage pre-trained language models for this task. However, three major challenges remain unresolved: web queries differ from natural language, on which pre-trained models are trained; web queries are short and lack contextual information; and labeled data for NER is scarce. We propose DeepTagger, a knowledge-enhanced NER model for web-based ads queries. The proposed knowledge enhancement framework leverages both model-free and model-based approaches. For model-free enhancement, we collect unlabeled web queries to augment domain knowledge; and we collect web search results to enrich the information of ads queries. We further leverage effective prompting methods to automatically generate labels using large language models such as ChatGPT. Additionally, we adopt a model-based knowledge enhancement method based on adversarial data augmentation. We employ a three-stage training framework to train DeepTagger models. Empirical results in various NER tasks demonstrate the effectiveness of the proposed framework.

cs.IR↗

arXiv2023

Two Ramsey-Turán numbers involving triangles

Given integers $p, q\ge2$, we say that a graph $G$ is $(K_p,K_q)$-free if there exists a red/blue edge coloring of $G$ such that it contains neither a red $K_p$ nor a blue $K_q$. Fix a function $f( n )$, the Ramsey-Turán number $RT( {n,p,q,f( n ))} $ is the maximum number of edges in an $n$-vertex $(K_p,K_q)$-free graph with independence number at most $f( n )$. For any $δ>0$, let $ρ(p, q,δ) = \mathop {\lim }\limits_{n \to \infty } \frac{RT(n,p, q,δn)}{n^2}$. We always call $ρ(p, q):= \mathop {\lim }\limits_{δ\to 0}ρ(p, q,δ)$ the Ramsey-Turán density of $K_p$ and $K_q$. In 1993, Erdős, Hajnal, Simonovits, Sós and Szemerédi proposed to determine the value of $ρ(3,q)$ for $q\ge3$, and they conjectured that for $q \ge 2$, $ρ\left( {3,2q - 1} \right) = \frac{1}{2}(1 - \frac{1}{r(3,q) - 1})$. Recently, Kim, Kim and Liu (2019) conjectured that for $q \ge 2$, $ρ( {3,2q } ) = \frac{1}{2}( 1 - \frac{1}{r( {3,q} )})$. Erdős et al. (1993) determined $ρ(3,q)$ for $q=3,4,5$ and $ρ(4,4)$. There is no progress on the Ramsey-Turán density $ρ(p, q)$ in the past thirty years. In this paper, we obtain $ρ(3,6)=\frac{5}{12}$ and $ρ(3,7)=\frac{7}{16}$. Moreover, we show that the corresponding asymptotically extremal structures are weakly stable, which answers a problem of Erdős et al. (1993) for the two cases.

math.CO↗

arXiv2022

DeeprETA: An ETA Post-processing System at Scale

Estimated Time of Arrival (ETA) plays an important role in delivery and ride-hailing platforms. For example, Uber uses ETAs to calculate fares, estimate pickup times, match riders to drivers, plan deliveries, and more. Commonly used route planning algorithms predict an ETA conditioned on the best available route, but such ETA estimates can be unreliable when the actual route taken is not known in advance. In this paper, we describe an ETA post-processing system in which a deep residual ETA network (DeeprETA) refines naive ETAs produced by a route planning algorithm. Offline experiments and online tests demonstrate that post-processing by DeeprETA significantly improves upon the accuracy of naive ETAs as measured by mean and median absolute error. We further show that post-processing by DeeprETA attains lower error than competitive baseline regression models.

cs.LG↗

arXiv2021

5th Place Solution for VSPW 2021 Challenge

In this article, we introduce the solution we used in the VSPW 2021 Challenge. Our experiments are based on two baseline models, Swin Transformer and MaskFormer. To further boost performance, we adopt stochastic weight averaging technique and design hierarchical ensemble strategy. Without using any external semantic segmentation dataset, our solution ranked the 5th place in the private leaderboard. Besides, we have some interesting attempts to tackle long-tail recognition and overfitting issues, which achieves improvement on val subset. Maybe due to distribution difference, these attempts don't work on test subset. We will also introduce these attempts and hope to inspire other researchers.

cs.CV↗

arXiv2021

When does loss-based prioritization fail?

Not all examples are created equal, but standard deep neural network training protocols treat each training point uniformly. Each example is propagated forward and backward through the network the same amount of times, independent of how much the example contributes to the learning protocol. Recent work has proposed ways to accelerate training by deviating from this uniform treatment. Popular methods entail up-weighting examples that contribute more to the loss with the intuition that examples with low loss have already been learned by the model, so their marginal value to the training procedure should be lower. This view assumes that updating the model with high loss examples will be beneficial to the model. However, this may not hold for noisy, real world data. In this paper, we theorize and then empirically demonstrate that loss-based acceleration methods degrade in scenarios with noisy and corrupted data. Our work suggests measures of example difficulty need to correctly separate out noise from other types of challenging examples.

cs.LG↗

arXiv2020

Personalized Policy Learning using Longitudinal Mobile Health Data

We address the personalized policy learning problem using longitudinal mobile health application usage data. Personalized policy represents a paradigm shift from developing a single policy that may prescribe personalized decisions by tailoring. Specifically, we aim to develop the best policy, one per user, based on estimating random effects under generalized linear mixed model. With many random effects, we consider new estimation method and penalized objective to circumvent high-dimension integrals for marginal likelihood approximation. We establish consistency and optimality of our method with endogenous app usage. We apply our method to develop personalized push ("prompt") schedules in 294 app users, with a goal to maximize the prompt response rate given past app usage and other contextual factors. We found the best push schedule given the same covariates varied among the users, thus calling for personalized policies. Using the estimated personalized policies would have achieved a mean prompt response rate of 23% in these users at 16 weeks or later: this is a remarkable improvement on the observed rate (11%), while the literature suggests 3%-15% user engagement at 3 months after download. The proposed method compares favorably to existing estimation methods including using the R function "glmer" in a simulation study.

stat.ME↗

arXiv2019

Applying SVGD to Bayesian Neural Networks for Cyclical Time-Series Prediction and Inference

A regression-based BNN model is proposed to predict spatiotemporal quantities like hourly rider demand with calibrated uncertainties. The main contributions of this paper are (i) A feed-forward deterministic neural network (DetNN) architecture that predicts cyclical time series data with sensitivity to anomalous forecasting events; (ii) A Bayesian framework applying SVGD to train large neural networks for such tasks, capable of producing time series predictions as well as measures of uncertainty surrounding the predictions. Experiments show that the proposed BNN reduces average estimation error by 10% across 8 U.S. cities compared to a fine-tuned multilayer perceptron (MLP), and 4% better than the same network architecture trained without SVGD.

cs.LG↗