Searcharxiv⌕ Search

arXiv subjects

Xiang Liu

Publications and source records attributed to Xiang Liu.

At least 217 records · Page 12Linked to original sources
arXiv2024

3D Question Answering for City Scene Understanding

3D multimodal question answering (MQA) plays a crucial role in scene understanding by enabling intelligent agents to comprehend their surroundings in 3D environments. While existing research has primarily focused on indoor household tasks and outdoor roadside autonomous driving tasks, there has been limited exploration of city-level scene understanding tasks. Furthermore, existing research faces challenges in understanding city scenes, due to the absence of spatial semantic information and human-environment interaction information at the city level.To address these challenges, we investigate 3D MQA from both dataset and method perspectives. From the dataset perspective, we introduce a novel 3D MQA dataset named City-3DQA for city-level scene understanding, which is the first dataset to incorporate scene semantic and human-environment interactive tasks within the city. From the method perspective, we propose a Scene graph enhanced City-level Understanding method (Sg-CityU), which utilizes the scene graph to introduce the spatial semantic. A new benchmark is reported and our proposed Sg-CityU achieves accuracy of 63.94 % and 63.76 % in different settings of City-3DQA. Compared to indoor 3D MQA methods and zero-shot using advanced large language models (LLMs), Sg-CityU demonstrates state-of-the-art (SOTA) performance in robustness and generalization.

cs.CV↗
arXiv2024

Active Prompting with Chain-of-Thought for Large Language Models

The increasing scale of large language models (LLMs) brings emergent abilities to various complex tasks requiring reasoning, such as arithmetic and commonsense reasoning. It is known that the effective design of task-specific prompts is critical for LLMs' ability to produce high-quality answers. In particular, an effective approach for complex question-and-answer tasks is example-based prompting with chain-of-thought (CoT) reasoning, which significantly improves the performance of LLMs. However, current CoT methods rely on a fixed set of human-annotated exemplars, which are not necessarily the most effective examples for different tasks. This paper proposes a new method, Active-Prompt, to adapt LLMs to different tasks with task-specific example prompts (annotated with human-designed CoT reasoning). For this purpose, we propose a solution to the key problem of determining which questions are the most important and helpful ones to annotate from a pool of task-specific queries. By borrowing ideas from the related problem of uncertainty-based active learning, we introduce several metrics to characterize the uncertainty so as to select the most uncertain questions for annotation. Experimental results demonstrate the superiority of our proposed method, achieving state-of-the-art on eight complex reasoning tasks. Further analyses of different uncertainty metrics, pool sizes, zero-shot learning, and accuracy-uncertainty relationship demonstrate the effectiveness of our method. Our code will be available at https://github.com/shizhediao/active-prompt.

cs.CL↗
arXiv2024

$R^3$-NL2GQL: A Model Coordination and Knowledge Graph Alignment Approach for NL2GQL

While current tasks of converting natural language to SQL (NL2SQL) using Foundation Models have shown impressive achievements, adapting these approaches for converting natural language to Graph Query Language (NL2GQL) encounters hurdles due to the distinct nature of GQL compared to SQL, alongside the diverse forms of GQL. Moving away from traditional rule-based and slot-filling methodologies, we introduce a novel approach, $R^3$-NL2GQL, integrating both small and large Foundation Models for ranking, rewriting, and refining tasks. This method leverages the interpretative strengths of smaller models for initial ranking and rewriting stages, while capitalizing on the superior generalization and query generation prowess of larger models for the final transformation of natural language queries into GQL formats. Addressing the scarcity of datasets in this emerging field, we have developed a bilingual dataset, sourced from graph database manuals and selected open-source Knowledge Graphs (KGs). Our evaluation of this methodology on this dataset demonstrates its promising efficacy and robustness.

cs.CL↗
arXiv2024

Are LLMs Rational Investors? A Study on Detecting and Reducing the Financial Bias in LLMs

Large Language Models (LLMs) are increasingly adopted in financial analysis for interpreting complex market data and trends. However, their use is challenged by intrinsic biases (e.g., risk-preference bias) and a superficial understanding of market intricacies, necessitating a thorough assessment of their financial insight. To address these issues, we introduce Financial Bias Indicators (FBI), a framework with components like Bias Unveiler, Bias Detective, Bias Tracker, and Bias Antidote to identify, detect, analyze, and eliminate irrational biases in LLMs. By combining behavioral finance principles with bias examination, we evaluate 23 leading LLMs and propose a de-biasing method based on financial causal knowledge. Results show varying degrees of financial irrationality among models, influenced by their design and training. Models trained specifically on financial datasets may exhibit more irrationality, and even larger financial language models (FinLLMs) can show more bias than smaller, general models. We utilize four prompt-based methods incorporating causal debiasing, effectively reducing financial biases in these models. This work enhances the understanding of LLMs' bias in financial applications, laying the foundation for developing more reliable and rational financial analysis tools.

cs.CL↗
arXiv2024

Plum: Prompt Learning using Metaheuristic

Since the emergence of large language models, prompt learning has become a popular method for optimizing and customizing these models. Special prompts, such as Chain-of-Thought, have even revealed previously unknown reasoning capabilities within these models. However, the progress of discovering effective prompts has been slow, driving a desire for general prompt optimization methods. Unfortunately, few existing prompt learning methods satisfy the criteria of being truly "general", i.e., automatic, discrete, black-box, gradient-free, and interpretable all at once. In this paper, we introduce metaheuristics, a branch of discrete non-convex optimization methods with over 100 options, as a promising approach to prompt learning. Within our paradigm, we test six typical methods: hill climbing, simulated annealing, genetic algorithms with/without crossover, tabu search, and harmony search, demonstrating their effectiveness in white-box and black-box prompt learning. Furthermore, we show that these methods can be used to discover more human-understandable prompts that were previously unknown in both reasoning and image generation tasks, opening the door to a cornucopia of possibilities in prompt optimization. We release all the codes in \url{https://github.com/research4pan/Plum}.

cs.LG↗
arXiv2024

Prospects for detecting the hidden-strange pentaquarklike state $N^{*}(2080)$ in the $π^{-} p\rightarrowϕn$ reaction

In this work, the production of the hidden-strange pentaquarklike state $N^{*}(2080)$ via the $π^{-} p$ scattering process is studied by the effective Lagrangian approach. Concretely, we consider the $ρ$ meson exchange of $t$-channel and the nucleon exchange of $u$-channel, which are treated as the background terms, and take into account the contribution of the $N^{*}(2080)$ via the $s$-channel as a signal term. By global fitting of the total and differential cross sections of the $π^{-} p\rightarrowϕn$ process, it is shown that the $N^{*}(2080)$ contributes an obvious peak at the threshold energy point of the differential cross section at the forward angle ($\cosθ= 1$), which provides clues to the detection of the $N^{*}(2080)$ by the $π^{-} p$ scattering process. However, due to the limited accuracy of the experimental data, it is an objective limitation for us to determine the properties of the $N^{*}(2080)$ by the $π^{-} p$ scattering process. Therefore, more accurate experimental measurements of the $π^{-} p\rightarrowϕn$ reaction are highly desirable, and we have also proposed that correlation measurements can be performed on J-PARC, AMBER, and future HIKE and HIAF meson beam experiments.

hep-ph↗
arXiv2024

An Efficient Implicit Neural Representation Image Codec Based on Mixed Autoregressive Model for Low-Complexity Decoding

Displaying high-quality images on edge devices, such as augmented reality devices, is essential for enhancing the user experience. However, these devices often face power consumption and computing resource limitations, making it challenging to apply many deep learning-based image compression algorithms in this field. Implicit Neural Representation (INR) for image compression is an emerging technology that offers two key benefits compared to cutting-edge autoencoder models: low computational complexity and parameter-free decoding. It also outperforms many traditional and early neural compression methods in terms of quality. In this study, we introduce a new Mixed AutoRegressive Model (MARM) to significantly reduce the decoding time for the current INR codec, along with a new synthesis network to enhance reconstruction quality. MARM includes our proposed AutoRegressive Upsampler (ARU) blocks, which are highly computationally efficient, and ARM from previous work to balance decoding time and reconstruction quality. We also propose enhancing ARU's performance using a checkerboard two-stage decoding strategy. Moreover, the ratio of different modules can be adjusted to maintain a balance between quality and speed. Comprehensive experiments demonstrate that our method significantly improves computational efficiency while preserving image quality. With different parameter settings, our method can achieve over a magnitude acceleration in decoding time without industrial level optimization, or achieve state-of-the-art reconstruction quality compared with other INR codecs. To the best of our knowledge, our method is the first INR-based codec comparable with Hyperprior in both decoding speed and quality while maintaining low complexity.

eess.IV↗
arXiv2024

Gravitational Wave Emission from Close-in Strange Quark Planets Around Strange Stars with Magnetic Interactions

According to the strange quark matter hypothesis, strange planets may exist, which are planetary mass objects composed of almost equal numbers of up, down and strange quarks. A strange planet can revolve around its host strange star in a very close-in orbit. When it finally merges with the host, strong gravitational wave emissions will be generated. Here the gravitational waveforms are derived for the merging process, taking into account the effects of the strange star's magnetic field on the dynamics. Effects of the inclination angle are also considered. Templates of the gravitational waveforms are derived. It is found that the magnetic interactions significantly speed up the merging process. Coalescence events of such strange planetary systems occurring in our Galaxy as well as in local galaxies can be effectively detected by current and future gravitational experiments, which may hopefully provide a new method to test the strange quark matter hypothesis and probe the magnetic field of compact stars.

astro-ph.HE↗
arXiv2024

Pruner-Zero: Evolving Symbolic Pruning Metric from scratch for Large Language Models

Despite the remarkable capabilities, Large Language Models (LLMs) face deployment challenges due to their extensive size. Pruning methods drop a subset of weights to accelerate, but many of them require retraining, which is prohibitively expensive and computationally demanding. Recently, post-training pruning approaches introduced novel metrics, enabling the pruning of LLMs without retraining. However, these metrics require the involvement of human experts and tedious trial and error. To efficiently identify superior pruning metrics, we develop an automatic framework for searching symbolic pruning metrics using genetic programming. In particular, we devise an elaborate search space encompassing the existing pruning metrics to discover the potential symbolic pruning metric. We propose an opposing operation simplification strategy to increase the diversity of the population. In this way, Pruner-Zero allows auto-generation of symbolic pruning metrics. Based on the searched results, we explore the correlation between pruning metrics and performance after pruning and summarize some principles. Extensive experiments on LLaMA and LLaMA-2 on language modeling and zero-shot tasks demonstrate that our Pruner-Zero obtains superior performance than SOTA post-training pruning methods. Code at: \url{https://github.com/pprp/Pruner-Zero}.

cs.LG↗
arXiv2024

How higher charmonia shape the puzzling data of the $e^+e^-\to ηJ/ψ$ cross section

Recently, the BESIII collaboration performed a precise measurement of the $e^+e^-\to ηJ/ψ$ cross section. It is puzzling that the resonance parameters of the reported $Y(4230)$ show a substantial divergence from the previously measured results in both the open-charmed and hidden-charmed decay channels, and the line shape asymmetry of the data approaching 4.2 GeV also suggests that it might be difficult to characterize the details of the structure around 4.2 GeV by a single resonance. This has motivated our great curiosity about how the charmonium states are distributed in the measured energy range and how they shape the puzzling data of the $e^+e^-\to ηJ/ψ$ cross section. In this work, we use five theoretically constructed charmonia in the range of $4.0\rm{-}4.5$ $\text{GeV}$, i.e., $ψ(4040)$, $ψ(4160)$, $ψ(4220)$, $ψ(4380)$, and $ψ(4415)$, to apply a combined fit to the data, in which their calculated decay ratios into $ηJ/ψ$ via hadronic loop mechanism are taken as input. The fit results can reproduce the measured cross section data well, especially for the subtle line shape around 4.2 GeV, showing that the structure around 4.2 GeV is possible from the contribution of both $ψ(4160)$ and $ψ(4220)$.

hep-ph↗
arXiv2024

FedCAda: Adaptive Client-Side Optimization for Accelerated and Stable Federated Learning

Federated learning (FL) has emerged as a prominent approach for collaborative training of machine learning models across distributed clients while preserving data privacy. However, the quest to balance acceleration and stability becomes a significant challenge in FL, especially on the client-side. In this paper, we introduce FedCAda, an innovative federated client adaptive algorithm designed to tackle this challenge. FedCAda leverages the Adam algorithm to adjust the correction process of the first moment estimate $m$ and the second moment estimate $v$ on the client-side and aggregate adaptive algorithm parameters on the server-side, aiming to accelerate convergence speed and communication efficiency while ensuring stability and performance. Additionally, we investigate several algorithms incorporating different adjustment functions. This comparative analysis revealed that due to the limited information contained within client models from other clients during the initial stages of federated learning, more substantial constraints need to be imposed on the parameters of the adaptive algorithm. As federated learning progresses and clients gather more global information, FedCAda gradually diminishes the impact on adaptive parameters. These findings provide insights for enhancing the robustness and efficiency of algorithmic improvements. Through extensive experiments on computer vision (CV) and natural language processing (NLP) datasets, we demonstrate that FedCAda outperforms the state-of-the-art methods in terms of adaptability, convergence, stability, and overall performance. This work contributes to adaptive algorithms for federated learning, encouraging further exploration.

cs.LG↗
arXiv2024

Very Long Baseline Array Observations of Parsec-scale Radio Emission in Dual Active Galactic Nuclei

It is believed that dual active galactic nuclei (dual AGN) will form during galaxies merge. Studying dual-AGN emission can provide valuable insights into galaxy merging and evolution. To investigate parsec-scale radio emission properties, we observed eight radio components of four selected dual-AGN systems using the Very Long Baseline Array (VLBA) at 5 GHz in multiple-phase-center mode. Among them, two compact radio components, labeled J0051+0020B and J2300-0005A, were detected clearly on parsec scales for the first time. However, the radio emission of the other six components was resolved out in the high-resolution images. We provided the values or upper limits of the brightness temperature and radio emission power, and analyzed the emission origins in detail for each target. Based on their physical properties reported in this work and in the literature, we suggest the radio emission in J0051+0020B and J2300-0005A originates primarily from compact jets, while the other six sources show more complex emission mechanisms. In addition, our VLBA observations suggest the systematic X-ray deficit in our dual-AGN sample is likely attributed to the tidally induced effect and possible viewing angle effect.

astro-ph.GA↗
arXiv2024

$\frac{5}{2}$ fractional quantum Hall state in GaAs with Landau level mixing

The Landau level mixing is the key in understanding the mysterious $5/2$ fractional quantum Hall effect in GaAs quantum well. Theoretical calculations with and without Landau level mixing show striking differences. However, the way to deal with the considerable strong Landau level mixing in GaAs is still unsatisfactory. We develop a method combining the screening and the perturbation theories to study the nature of the $5/2$ fractional quantum Hall effect in GaAs efficiently. The screening which has been succeed in explaining ZnO systems integrates out the low-energy Landau levels close to the related Landau level, while the other high-energy Landau levels are integrated out by the perturbation theory. We find that the ground states still hold the quasi-triplet degeneracy which implies the Pfaffian nature of the system. Furthermore, the particle-hole symmetry is only weakly violated since the particle-hole parity is close to unity. We propose that the ground state in the finite-size calculations can be approximated as a variational superposition of the Pfaffian and anit-Pfaffian states. In the experimental environment the symmetrized Pfaffian component is dominant, corresponding a thermal conductance around $2.5$ quanta can be understood consequently.

cond-mat.str-el↗
arXiv2024

The role of electromagnetic interaction in the $X(3872)$ and its analogs

We investigate the role of the electromagnetic interaction in the formation and decay of the $X(3872)$. The binding properties of the $X(3872)$ are studied by assuming the molecular nature and considering the $S$-$D$ wave mixing, isospin breaking, and coupled channel effects, and in particular the correction from the electromagnetic interaction. The radiative decays can better reflect the difference between the charged and neutral $D\bar D^*$ components, since the electromagnetic interaction explicitly breaks the isospin symmetry. We further study the radiative decay widths with the obtained wave functions for different $D\bar D^*$ channels. We also explore other similar hidden-charm molecular states. The electromagnetic interaction can make the molecule tighter. Our result of the radiative decay width for $X(3872)\rightarrow γJ/ψ$ is in agreement with the experiment. The branching ratio $R_{γψ}$ is less than 1 in our framework, which supports the Belle and BESIII measurements.

hep-ph↗
arXiv2024

Isospin violation effect and three-body decays of the $T_{cc}^{+}$ state

In this work, we make a study of $T_{cc}^+$ state observed by the LHCb collaboration in 2021. In obtaining the effective potentials using the One-Boson-Exchange Potential Model we use an exponential form factor, and find that in the short and medium range, the contributions of the $π$, $ρ$ and $ω$ exchanges are comparable while in the long range the pion-exchange contribution is dominant. Based on the assumption that $T_{cc}^+$ is a loosely bound state of $D^*D$, we focus on its three-body decay using the meson-exchange method. Considering that the difference between the thresholds of $D^{*+}D^0$ and $D^{*0}D^+$ is even larger than the binding energy of $T_{cc}^+$, the isospin-breaking effect is amplified by the small binding energy of $T_{cc}^+$. Explicitly including such an isospin-breaking effect we obtain, by solving the Schrödinger equation, that the probability of the isoscalar component is about $91\%$ while that of the isovector component is around $9\%$ for $T_{cc}^+$. Using the experimental value of the mass of $T_{cc}^+$ as an input, we obtain the wave function of $T_{cc}^+$ and further obtain its width via the three-body hadronic as well as the radiative decays. The total width we obtain is in agreement with the experimental value of the LHCb measurement with a unitarised Breit-Wigner profile. Conversely, the current results support the conclusion that $T_{cc}^+$ is a hadronic molecule of $D^*D$.

hep-ph↗
arXiv2024

Sparse Variable Selection on High Dimensional Heterogeneous Data with Tree Structured Responses

We consider the problem of sparse variable selection on high dimension heterogeneous data sets, which has been taking on renewed interest recently due to the growth of biological and medical data sets with complex, non-i.i.d. structures and huge quantities of response variables. The heterogeneity is likely to confound the association between explanatory variables and responses, resulting in enormous false discoveries when Lasso or its variants are naïvely applied. Therefore, developing effective confounder correction methods is a growing heat point among researchers. However, ordinarily employing recent confounder correction methods will result in undesirable performance due to the ignorance of the convoluted interdependency among response variables. To fully improve current variable selection methods, we introduce a model, the tree-guided sparse linear mixed model, that can utilize the dependency information from multiple responses to explore how specifically clusters are and select the active variables from heterogeneous data. Through extensive experiments on synthetic and real data sets, we show that our proposed model outperforms the existing methods and achieves the highest ROC area.

stat.ME↗
arXiv2024

MAPO: Advancing Multilingual Reasoning through Multilingual Alignment-as-Preference Optimization

Though reasoning abilities are considered language-agnostic, existing LLMs exhibit inconsistent reasoning abilities across different languages, e.g., reasoning in the dominant language like English is superior to other languages due to the imbalance of multilingual training data. To enhance reasoning abilities in non-dominant languages, we propose a Multilingual-Alignment-as-Preference Optimization framework (MAPO), aiming to align the reasoning processes in other languages with the dominant language. Specifically, we harness an off-the-shelf translation model for the consistency between answers in non-dominant and dominant languages, which we adopt as the preference for optimization, e.g., Direct Preference Optimization (DPO) or Proximal Policy Optimization (PPO). Experiments show that MAPO stably achieves significant improvements in the multilingual reasoning of various models on all three benchmarks (MSVAMP +16.2%, MGSM +6.1%, and MNumGLUESub +13.3%), with improved reasoning consistency across languages.

cs.CL↗