Search arXiv⌕ Search

arXiv subjects

Yu Gu

Publications and source records attributed to Yu Gu.

At least 199 records · Page 11Linked to original sources
arXiv2023

Few-shot In-context Learning for Knowledge Base Question Answering

Question answering over knowledge bases is considered a difficult problem due to the challenge of generalizing to a wide variety of possible natural language questions. Additionally, the heterogeneity of knowledge base schema items between different knowledge bases often necessitates specialized training for different knowledge base question-answering (KBQA) datasets. To handle questions over diverse KBQA datasets with a unified training-free framework, we propose KB-BINDER, which for the first time enables few-shot in-context learning over KBQA tasks. Firstly, KB-BINDER leverages large language models like Codex to generate logical forms as the draft for a specific question by imitating a few demonstrations. Secondly, KB-BINDER grounds on the knowledge base to bind the generated draft to an executable one with BM25 score matching. The experimental results on four public heterogeneous KBQA datasets show that KB-BINDER can achieve a strong performance with only a few in-context demonstrations. Especially on GraphQA and 3-hop MetaQA, KB-BINDER can even outperform the state-of-the-art trained models. On GrailQA and WebQSP, our model is also on par with other fully-trained models. We believe KB-BINDER can serve as an important baseline for future research. Our code is available at https://github.com/ltl3A87/KB-BINDER.

cs.CL↗
arXiv2023

Don't Generate, Discriminate: A Proposal for Grounding Language Models to Real-World Environments

A key missing capacity of current language models (LMs) is grounding to real-world environments. Most existing work for grounded language understanding uses LMs to directly generate plans that can be executed in the environment to achieve the desired effects. It thereby casts the burden of ensuring grammaticality, faithfulness, and controllability all on the LMs. We propose Pangu, a generic framework for grounded language understanding that capitalizes on the discriminative ability of LMs instead of their generative ability. Pangu consists of a symbolic agent and a neural LM working in a concerted fashion: The agent explores the environment to incrementally construct valid plans, and the LM evaluates the plausibility of the candidate plans to guide the search process. A case study on the challenging problem of knowledge base question answering (KBQA), which features a massive environment, demonstrates the remarkable effectiveness and flexibility of Pangu: A BERT-base LM is sufficient for setting a new record on standard KBQA datasets, and larger LMs further bring substantial gains. Pangu also enables, for the first time, effective few-shot in-context learning for KBQA with large LMs such as Codex.

cs.CL↗
arXiv2023

Weak Decays of Antitriplet Charmed Baryons from the Perspective of Flavor Symmetry

In this work, we carry out a global fitting for the two-body weak decays of antitriplet charmed baryons in both SU(3) respected and broken scenarios incorporating all the available data up to date. In the SU(3) irreducible representation approach (IRA), more amplitudes for irreducible representation terms are taken into account and the ranges for their coefficients in each scenarios are predicted. By a comparison among various fitting schemes in this work, experimental data prefer the SU(3) symmetry breaking scenario. Observables of interest, branching fractions and decay asymmetries of all the Cabibbo-favored (CF), singly Cabibbo-suppresed (SCS) and doubly Cabibbo-suppressed (DCS) channels are calculated in the chosen fitting scheme. Most of our predictions are consistent well with experimental data. We further propose more ways to explore the SU(3) symmetry in charmed baryon decays. An improved measurement for $α(Λ_c^+\to p K_S)$ is called for since predictions from both fittings, including the one in current work and earlier works, and the pole model calculation prefer an opposite sign from previous experimental value. Given branching fractions of most of the CF modes and part of the SCS modes being measured, predictions for the remaining channels, including the DCS modes as well as more decay asymmetries, are anticipated to be checked by the upcoming experiments in BESIII, Belle/Belle-II and LHCb.

hep-ph↗
arXiv2023

Localization length of the $1+1$ continuum directed random polymer

In this paper, we study the localization length of the $1+1$ continuum directed polymer, defined as the distance between the endpoints of two paths sampled independently from the quenched polymer measure. We show that the localization length converges in distribution in the thermodynamic limit, and derive an explicit density formula of the limiting distribution. As a consequence, we prove the $\tfrac32$-power law decay of the density, confirming the physics prediction of Hwa-Fisher \cite{fisher}. Our proof uses the recent result of Das-Zhu \cite{daszhu}.

math.PR↗
arXiv2023

High temperature behaviors of the directed polymer on a cylinder

In this paper, we study the free energy of the directed polymer on a cylinder of radius $L$ with the inverse temperature $β$. Assuming the random environment is given by a Gaussian process that is white in time and smooth in space, with an arbitrary compactly supported spatial covariance function, we obtain precise scaling behaviors of the limiting free energy for high temperatures $β\ll1$, followed by large $L\gg1$, in all dimensions. Our approach is based on a perturbative expansion of the PDE hierarchy satisfied by the multipoint correlation function of the polymer endpoint distribution. For the random environment given by the $1+1$ spacetime white noise, we derive an explicit expression of the limiting free energy, confirming the result obtained through the replica method in [12].

math.PR↗
arXiv2022

21st Century Global and Regional Surface Temperature Projections

Many regions across the globe broke their surface temperature records in recent years, further sparking concerns about the impending arrival of "tipping points" later in the 21st century. This study analyzes observed global surface temperature trends in three target latitudinal regions: the Arctic Circle, the Tropics, and the Antarctic Circle. We show that global warming is accelerating unevenly across the planet, with the Arctic warming at approximately three times the average rate of our world. We further analyzed the reliability of latitude-dependent surface temperature simulations from a suite of Coupled Model Intercomparison Project Phase 6 models and their multi-model mean. We found that GISS-E2-1-G and FGOALS-g3 were the best-performing models based on their statistical abilities to reproduce observational, latitude-dependent data. Surface temperatures were projected from ensemble simulations of the Shared Socioeconomic Pathway 2-4.5 (SSP2-4.5). We estimate when the climate will warm by 1.5, 2.0, and 2.5 degrees C relative to the preindustrial period, globally and regionally. GISS-E2-1-G projects that global surface temperature anomalies would reach 1.5, 2.0, and 2.5 degrees C in 2024 (+/-1.34), 2039 (+/-2.83), and 2057 (+/-5.03) respectively, while FGOALS-g3 predicts these "tipping points" would arrive in 2024 (+/-2.50), 2054 (+/-7.90), and 2087 (+/-10.55) respectively. Our results reaffirm a dramatic, upward trend in projected climate warming acceleration, with upward concavity in 21st century projections of the Arctic, which could lead to catastrophic consequences across the Earth. Further studies are necessary to determine the most efficient solutions to reduce global warming acceleration and maintain a low SSP, both globally and regionally.

physics.ao-ph↗
arXiv2022

Long-time behavior for a nonlocal model from directed polymers

We consider the long time behavior of solutions to a nonlocal reaction diffusion equation that arises in the study of directed polymers. The model is characterized by convolution with a kernel $R$ and an $L^2$ inner product. In one spatial dimension, we extend a previous result of the authors [arXiv:2002.02799], where only the case $R =δ$ was considered; in particular, we show that solutions spread according to a $2/3$ power law consistent with the KPZ scaling conjectured for direct polymers. In the special case when $R = δ$, we find the exact profile of the solution in the rescaled coordinates. We also consider the behavior in higher dimensions. When the dimension is three or larger, we show that the long-time behavior is the same as the heat equation in the sense that the solution converges to a standard Gaussian. In contrast, when the dimension is two, we construct a non-Gaussian self-similar solution.

math.AP↗
arXiv2022

Another look at the Balázs-Quastel-Seppäläinen theorem

We study the KPZ equation with a $1+1-$dimensional spacetime white noise, started at equilibrium, and give a different proof of the main result of \cite{bqs}, i.e., the variance of the solution at time $t$ is of order $t^{2/3}$. Instead of using a discrete approximation through the exclusion process and the second class particle, we utilize the connection to directed polymers in a random environment. Along the way, we show the annealed density of the stationary continuum directed polymer equals to the two-point covariance function of the stationary stochastic Burgers equation, confirming the physics prediction in \cite{MT}.

math.PR↗
arXiv2022

ArcaneQA: Dynamic Program Induction and Contextualized Encoding for Knowledge Base Question Answering

Question answering on knowledge bases (KBQA) poses a unique challenge for semantic parsing research due to two intertwined challenges: large search space and ambiguities in schema linking. Conventional ranking-based KBQA models, which rely on a candidate enumeration step to reduce the search space, struggle with flexibility in predicting complicated queries and have impractical running time. In this paper, we present ArcaneQA, a novel generation-based model that addresses both the large search space and the schema linking challenges in a unified framework with two mutually boosting ingredients: dynamic program induction for tackling the large search space and dynamic contextualized encoding for schema linking. Experimental results on multiple popular KBQA datasets demonstrate the highly competitive performance of ArcaneQA in both effectiveness and efficiency.

cs.CL↗
arXiv2022

Knowledge Base Question Answering: A Semantic Parsing Perspective

Recent advances in deep learning have greatly propelled the research on semantic parsing. Improvement has since been made in many downstream tasks, including natural language interface to web APIs, text-to-SQL generation, among others. However, despite the close connection shared with these tasks, research on question answering over knowledge bases (KBQA) has comparatively been progressing slowly. We identify and attribute this to two unique challenges of KBQA, schema-level complexity and fact-level complexity. In this survey, we situate KBQA in the broader literature of semantic parsing and give a comprehensive account of how existing KBQA approaches attempt to address the unique challenges. Regardless of the unique challenges, we argue that we can still take much inspiration from the literature of semantic parsing, which has been overlooked by existing research on KBQA. Based on our discussion, we can better understand the bottleneck of current KBQA research and shed light on promising directions for KBQA to keep up with the literature of semantic parsing, particularly in the era of pre-trained language models.

cs.CL↗
arXiv2022

Dimension Reduction for Efficient Dense Retrieval via Conditional Autoencoder

Dense retrievers encode queries and documents and map them in an embedding space using pre-trained language models. These embeddings need to be high-dimensional to fit training signals and guarantee the retrieval effectiveness of dense retrievers. However, these high-dimensional embeddings lead to larger index storage and higher retrieval latency. To reduce the embedding dimensions of dense retrieval, this paper proposes a Conditional Autoencoder (ConAE) to compress the high-dimensional embeddings to maintain the same embedding distribution and better recover the ranking features. Our experiments show that ConAE is effective in compressing embeddings by achieving comparable ranking performance with its teacher model and making the retrieval system more efficient. Our further analyses show that ConAE can alleviate the redundancy of the embeddings of dense retrieval with only one linear layer. All codes of this work are available at https://github.com/NEUIR/ConAE.

cs.IR↗
arXiv2022

Maximum Likelihood Estimation for Semiparametric Regression Models with Interval-Censored Multi-State Data

Interval-censored multi-state data arise in many studies of chronic diseases, where the health status of a subject can be characterized by a finite number of disease states and the transition between any two states is only known to occur over a broad time interval. We formulate the effects of potentially time-dependent covariates on multi-state processes through semiparametric proportional intensity models with random effects. We adopt nonparametric maximum likelihood estimation (NPMLE) under general interval censoring and develop a stable expectation-maximization (EM) algorithm. We show that the resulting parameter estimators are consistent and that the finite-dimensional components are asymptotically normal with a covariance matrix that attains the semiparametric efficiency bound and can be consistently estimated through profile likelihood. In addition, we demonstrate through extensive simulation studies that the proposed numerical and inferential procedures perform well in realistic settings. Finally, we provide an application to a major epidemiologic cohort study.

stat.ME↗
arXiv2022

Event-Triggered Extended State Observer Based Distributed Control of Nonlinear Vehicle Platoons

We study the platoon control of vehicles with third-order nonlinear dynamics under the constant spacing policy. We consider a vehicle model with parameter uncertainties and external disturbances and propose a distributed control law based on an event-triggered extended state observer (ESO). First, an event-triggered ESO is designed to estimate the unmodeled dynamics in the vehicle model. Then based on the estimate of the unmodeled dynamics, a distributed control law is designed by using a modified dynamic surface control method. The control law of each follower vehicle only uses the information obtained by on-board sensors, including its own velocity, acceleration, the velocity of the preceding vehicle and the inter-vehicle distance. Finally, we give the range of the control parameters to ensure the stability of the vehicle platoon system. It is shown that the control parameters can be properly designed to make the observation errors of the ESOs bounded and ensure the string stability and closed-loop stability. We prove that the Zeno behavior is avoided under the designed event-triggered mechanism. The joint simulations of CarSim and MATLAB are given to demonstrate the effectiveness of the proposed control law.

eess.SY↗
arXiv2022

Combining deep learning and crowdsourcing geo-images to predict housing quality in rural China

Housing quality is an essential proxy for regional wealth, security and health. Understanding the distribution of housing quality is crucial for unveiling rural development status and providing political proposals. However,present rural house quality data highly depends on a top-down, time-consuming survey at the national or provincial level but fails to unpack the housing quality at the village level. To fill the gap between accurately depicting rural housing quality conditions and deficient data,we collect massive rural images and invite users to assess their housing quality at scale. Furthermore, a deep learning framework is proposed to automatically and efficiently predict housing quality based on crowd-sourcing rural images.

cs.LG↗
arXiv2022

Proprioceptive Slip Detection for Planetary Rovers in Perceptually Degraded Extraterrestrial Environments

Slip detection is of fundamental importance for the safety and efficiency of rovers driving on the surface of extraterrestrial bodies. Current planetary rover slip detection systems rely on visual perception on the assumption that sufficient visual features can be acquired in the environment. However, visual-based methods are prone to suffer in perceptually degraded planetary environments with dominant low terrain features such as regolith, glacial terrain, salt-evaporites, and poor lighting conditions such as dark caves and permanently shadowed regions. Relying only on visual sensors for slip detection also requires additional computational power and reduces the rover traversal rate. This paper answers the question of how to detect wheel slippage of a planetary rover without depending on visual perception. In this respect, we propose a slip detection system that obtains its information from a proprioceptive localization framework that is capable of providing reliable, continuous, and computationally efficient state estimation over hundreds of meters. This is accomplished by using zero velocity update, zero angular rate update, and non-holonomic constraints as pseudo-measurement updates on an inertial navigation system framework. The proposed method is evaluated on actual hardware and field-tested in a planetary-analog environment. The method achieves greater than 92% slip detection accuracy for distances around 150 m using only an IMU and wheel encoders.

cs.RO↗
arXiv2022

Gaussian fluctuations of replica overlap in directed polymers

In this short note, we prove a central limit theorem for a type of replica overlap of the Brownian directed polymer in a Gaussian random environment, in the low temperature regime and in all dimensions. The proof relies on a superconcentration result for the KPZ equation driven by a spatially mollified noise, which is inspired by the recent work of Chatterjee \cite{C1}.

math.PR↗