arXiv subjects

Jia Liu

Publications and source records attributed to Jia Liu.

At least 181 records · Page 10Linked to original sources

arXiv2024

StegaINR4MIH: steganography by implicit neural representation for multi-image hiding

Multi-image hiding, which embeds multiple secret images into a cover image and is able to recover these images with high quality, has gradually become a research hotspot in the field of image steganography. However, due to the need to embed a large amount of data in a limited cover image space, issues such as contour shadowing or color distortion often arise, posing significant challenges for multi-image hiding. In this paper, we propose StegaINR4MIH, a novel implicit neural representation steganography framework that enables the hiding of multiple images within a single implicit representation function. In contrast to traditional methods that use multiple encoders to achieve multi-image embedding, our approach leverages the redundancy of implicit representation function parameters and employs magnitude-based weight selection and secret weight substitution on pre-trained cover image functions to effectively hide and independently extract multiple secret images. We conduct experiments on images with a resolution of from three different datasets: CelebA-HQ, COCO, and DIV2K. When hiding two secret images, the PSNR values of both the secret images and the stego images exceed 42. When hiding five secret images, the PSNR values of both the secret images and the stego images exceed 39. Extensive experiments demonstrate the superior performance of the proposed method in terms of visual quality and undetectability.

cs.CV↗

arXiv2024

Significance of void shape: Neutrino mass from Voronoi void halos?

Massive neutrinos suppress the growth of cosmic structure on nonlinear scales, motivating the use of information beyond the power spectrum to tighten constraints on the neutrino mass, for example by considering cosmic voids. It was recently proposed that constraints on neutrino mass from the halo mass function (HMF) can be improved by considering only the halos that reside within voids -- the void-halo mass function (VHMF). We extend this analysis, which made spherical assumptions about the shape of voids, to take into account the non-spherical nature of voids as defined by the Voronoi-tessellation-based void finder, VIDE. In turn, after accounting for one spurious non-spherical void, we find no evidence that the VHMF contains information beyond the HMF. Given this finding, we then introduce a novel summary statistic by splitting halos according to the emptiness of their individual environments, defined by the Voronoi cell volume each halo resides in, and combining the mass functions from each split. We name the corresponding statistic the VorHMF and find that it could provide information regarding neutrino mass beyond the HMF. Our work thus motivates the importance of accounting for the full shape of voids in future analyses, both in terms of removing outliers to achieve robust results and as an additional source of cosmological information.

astro-ph.CO↗

arXiv2024

EfficientRAG: Efficient Retriever for Multi-Hop Question Answering

Retrieval-augmented generation (RAG) methods encounter difficulties when addressing complex questions like multi-hop queries. While iterative retrieval methods improve performance by gathering additional information, current approaches often rely on multiple calls of large language models (LLMs). In this paper, we introduce EfficientRAG, an efficient retriever for multi-hop question answering. EfficientRAG iteratively generates new queries without the need for LLM calls at each iteration and filters out irrelevant information. Experimental results demonstrate that EfficientRAG surpasses existing RAG methods on three open-domain multi-hop question-answering datasets.

cs.CL↗

arXiv2024

AToM-Bot: Embodied Fulfillment of Unspoken Human Needs with Affective Theory of Mind

We propose AToM-Bot, a novel task generation and execution framework for proactive robot-human interaction, which leverages the human mental and physical state inference capabilities of the Vision Language Model (VLM) prompted by the Affective Theory of Mind (AToM). Without requiring explicit commands by humans, AToM-Bot proactively generates and follows feasible tasks to improve general human well-being. When around humans, AToM-Bot first detects current human needs based on inferred human states and observations of the surrounding environment. It then generates tasks to fulfill these needs, taking into account its embodied constraints. We designed 16 daily life scenarios spanning 4 common scenes and tasked the same visual stimulus to 59 human subjects and our robot. We used the similarity between human open-ended answers and robot output, and the human satisfaction scores to metric robot performance. AToM-Bot received high human evaluations in need detection (6.42/7, 91.7%), embodied solution (6.15/7, 87.8%) and task execution (6.17/7, 88.1%). We show that AToM-Bot excels in generating and executing feasible plans to fulfill unspoken human needs. Videos and code are available at https://affective-tom-bot.github.io.

cs.RO↗

arXiv2024

Strength Lies in Differences! Improving Strategy Planning for Non-collaborative Dialogues via Diversified User Simulation

We investigate non-collaborative dialogue agents, which are expected to engage in strategic conversations with diverse users, for securing a mutual agreement that leans favorably towards the system's objectives. This poses two main challenges for existing dialogue agents: 1) The inability to integrate user-specific characteristics into the strategic planning, and 2) The difficulty of training strategic planners that can be generalized to diverse users. To address these challenges, we propose Trip to enhance the capability in tailored strategic planning, incorporating a user-aware strategic planning module and a population-based training paradigm. Through experiments on benchmark non-collaborative dialogue tasks, we demonstrate the effectiveness of Trip in catering to diverse users.

cs.CL↗

arXiv2024

On the Hardness of Decentralized Multi-Agent Policy Evaluation under Byzantine Attacks

In this paper, we study a fully-decentralized multi-agent policy evaluation problem, which is an important sub-problem in cooperative multi-agent reinforcement learning, in the presence of up to $f$ faulty agents. In particular, we focus on the so-called Byzantine faulty model with model poisoning setting. In general, policy evaluation is to evaluate the value function of any given policy. In cooperative multi-agent system, the system-wide rewards are usually modeled as the uniform average of rewards from all agents. We investigate the multi-agent policy evaluation problem in the presence of Byzantine agents, particularly in the setting of heterogeneous local rewards. Ideally, the goal of the agents is to evaluate the accumulated system-wide rewards, which are uniform average of rewards of the normal agents for a given policy. It means that all agents agree upon common values (the consensus part) and furthermore, the consensus values are the value functions (the convergence part). However, we prove that this goal is not achievable. Instead, we consider a relaxed version of the problem, where the goal of the agents is to evaluate accumulated system-wide reward, which is an appropriately weighted average reward of the normal agents. We further prove that there is no correct algorithm that can guarantee that the total number of positive weights exceeds $|\mathcal{N}|-f $, where $|\mathcal{N}|$ is the number of normal agents. Towards the end, we propose a Byzantine-tolerant decentralized temporal difference algorithm that can guarantee asymptotic consensus under scalar function approximation. We then empirically test the effective of the proposed algorithm.

cs.CR↗

arXiv2024

Drift to Remember

Lifelong learning in artificial intelligence (AI) aims to mimic the biological brain's ability to continuously learn and retain knowledge, yet it faces challenges such as catastrophic forgetting. Recent neuroscience research suggests that neural activity in biological systems undergoes representational drift, where neural responses evolve over time, even with consistent inputs and tasks. We hypothesize that representational drift can alleviate catastrophic forgetting in AI during new task acquisition. To test this, we introduce DriftNet, a network designed to constantly explore various local minima in the loss landscape while dynamically retrieving relevant tasks. This approach ensures efficient integration of new information and preserves existing knowledge. Experimental studies in image classification and natural language processing demonstrate that DriftNet outperforms existing models in lifelong learning. Importantly, DriftNet is scalable in handling a sequence of tasks such as sentiment analysis and question answering using large language models (LLMs) with billions of parameters on a single Nvidia A100 GPU. DriftNet efficiently updates LLMs using only new data, avoiding the need for full dataset retraining. Tested on GPT-2 and RoBERTa, DriftNet is a robust, cost-effective solution for lifelong learning in LLMs. This study not only advances AI systems to emulate biological learning, but also provides insights into the adaptive mechanisms of biological neural systems, deepening our understanding of lifelong learning in nature.

cs.AI↗

arXiv2024

Improving Anomalous Sound Detection via Low-Rank Adaptation Fine-Tuning of Pre-Trained Audio Models

Anomalous Sound Detection (ASD) has gained significant interest through the application of various Artificial Intelligence (AI) technologies in industrial settings. Though possessing great potential, ASD systems can hardly be readily deployed in real production sites due to the generalization problem, which is primarily caused by the difficulty of data collection and the complexity of environmental factors. This paper introduces a robust ASD model that leverages audio pre-trained models. Specifically, we fine-tune these models using machine operation data, employing SpecAug as a data augmentation strategy. Additionally, we investigate the impact of utilizing Low-Rank Adaptation (LoRA) tuning instead of full fine-tuning to address the problem of limited data for fine-tuning. Our experiments on the DCASE2023 Task 2 dataset establish a new benchmark of 77.75% on the evaluation set, with a significant improvement of 6.48% compared with previous state-of-the-art (SOTA) models, including top-tier traditional convolutional networks and speech pre-trained models, which demonstrates the effectiveness of audio pre-trained models with LoRA tuning. Ablation studies are also conducted to showcase the efficacy of the proposed scheme.

cs.SD↗

arXiv2024

Hitting the mark: Optimising Marked Power Spectra for Cosmology

Marked power spectra provide a computationally efficient way to extract non-Gaussian information from the matter density field using the usual analysis tools developed for the power spectrum without the need for explicit calculation of higher-order correlators. In this work, we explore the optimal form of the mark function used for re-weighting the density field, to maximally constrain cosmology. We show that adding to the mark function or multiplying it by a constant leads to no additional information gain, which significantly reduces our search space for optimal marks. We quantify the information gain of this optimal function and compare it against mark functions previously proposed in the literature. We find that we can gain around $\sim2$ times smaller errors in $σ_8$ and $\sim4$ times smaller errors in $Ω_m$ compared to using the traditional power spectrum alone, an improvement of $\sim60\%$ compared to other proposed marks when applied to the same dataset.

astro-ph.CO↗

arXiv2024

EmoAttack: Utilizing Emotional Voice Conversion for Speech Backdoor Attacks on Deep Speech Classification Models

Deep speech classification tasks, mainly including keyword spotting and speaker verification, play a crucial role in speech-based human-computer interaction. Recently, the security of these technologies has been demonstrated to be vulnerable to backdoor attacks. Specifically speaking, speech samples are attacked by noisy disruption and component modification in present triggers. We suggest that speech backdoor attacks can strategically focus on emotion, a higher-level subjective perceptual attribute inherent in speech. Furthermore, we proposed that emotional voice conversion technology can serve as the speech backdoor attack trigger, and the method is called EmoAttack. Based on this, we conducted attack experiments on two speech classification tasks, showcasing that EmoAttack method owns impactful trigger effectiveness and its remarkable attack success rate and accuracy variance. Additionally, the ablation experiments found that speech with intensive emotion is more suitable to be targeted for attacks.

cs.SD↗

arXiv2024

Can We Theoretically Quantify the Impacts of Local Updates on the Generalization Performance of Federated Learning?

Federated Learning (FL) has gained significant popularity due to its effectiveness in training machine learning models across diverse sites without requiring direct data sharing. While various algorithms along with their optimization analyses have shown that FL with local updates is a communication-efficient distributed learning framework, the generalization performance of FL with local updates has received comparatively less attention. This lack of investigation can be attributed to the complex interplay between data heterogeneity and infrequent communication due to the local updates within the FL framework. This motivates us to investigate a fundamental question in FL: Can we quantify the impact of data heterogeneity and local updates on the generalization performance for FL as the learning process evolves? To this end, we conduct a comprehensive theoretical study of FL's generalization performance using a linear model as the first step, where the data heterogeneity is considered for both the stationary and online/non-stationary cases. By providing closed-form expressions of the model error, we rigorously quantify the impact of the number of the local updates (denoted as $K$) under three settings ($K=1$, $K<\infty$, and $K=\infty$) and show how the generalization performance evolves with the number of rounds $t$. Our investigation also provides a comprehensive understanding of how different configurations (including the number of model parameters $p$ and the number of training samples $n$) contribute to the overall generalization performance, thus shedding new insights (such as benign overfitting) for implementing FL over networks.

cs.LG↗

arXiv2024

Cosmology from HSC Y1 Weak Lensing with Combined Higher-Order Statistics and Simulation-based Inference

We present cosmological constraints from weak lensing with the Subaru Hyper Suprime-Cam (HSC) first-year (Y1) data, using a simulation-based inference (SBI) method. % We explore the performance of a set of higher-order statistics (HOS) including the Minkowski functionals, counts of peaks and minima, and the probability distribution function and compare them to the traditional two-point statistics. The HOS, also known as non-Gaussian statistics, can extract additional non-Gaussian information that is inaccessible to the two-point statistics. We use a neural network to compress the summary statistics, followed by an SBI approach to infer the posterior distribution of the cosmological parameters. We apply cuts on angular scales and redshift bins to mitigate the impact of systematic effects. Combining two-point and non-Gaussian statistics, we obtain $S_8 \equiv σ_8 \sqrt{Ω_m/0.3} = 0.804_{-0.040}^{+0.041}$ and $Ω_m = 0.344_{-0.090}^{+0.083}$, similar to that from non-Gaussian statistics alone. These results are consistent with previous HSC analyses and Planck 2018 cosmology. Our constraints from non-Gaussian statistics are $\sim 25\%$ tighter in $S_8$ than two-point statistics, where the main improvement lies in $Ω_m$, with $\sim 40$\% tighter error bar compared to using the angular power spectrum alone ($S_8 = 0.766_{-0.056}^{+0.054}$ and $Ω_m = 0.365_{-0.141}^{+0.148}$). We find that, among the non-Gaussian statistics we studied, the Minkowski functionals are the primary driver for this improvement. Our analyses confirm the SBI as a powerful approach for cosmological constraints, avoiding any assumptions about the functional form of the data's likelihood.

astro-ph.CO↗

arXiv2024

CoopASD: Cooperative Machine Anomalous Sound Detection with Privacy Concerns

Machine anomalous sound detection (ASD) has emerged as one of the most promising applications in the Industrial Internet of Things (IIoT) due to its unprecedented efficacy in mitigating risks of malfunctions and promoting production efficiency. Previous works mainly investigated the machine ASD task under centralized settings. However, developing the ASD system under decentralized settings is crucial in practice, since the machine data are dispersed in various factories and the data should not be explicitly shared due to privacy concerns. To enable these factories to cooperatively develop a scalable ASD model while preserving their privacy, we propose a novel framework named CoopASD, where each factory trains an ASD model on its local dataset, and a central server aggregates these local models periodically. We employ a pre-trained model as the backbone of the ASD model to improve its robustness and develop specialized techniques to stabilize the model under a completely non-iid and domain shift setting. Compared with previous state-of-the-art (SOTA) models trained in centralized settings, CoopASD showcases competitive results with negligible degradation of 0.08%. We also conduct extensive ablation studies to demonstrate the effectiveness of CoopASD.

cs.SD↗

arXiv2024

CARE: A Clue-guided Assistant for CSRs to Read User Manuals

It is time-saving to build a reading assistant for customer service representations (CSRs) when reading user manuals, especially information-rich ones. Current solutions don't fit the online custom service scenarios well due to the lack of attention to user questions and possible responses. Hence, we propose to develop a time-saving and careful reading assistant for CSRs, named CARE. It can help the CSRs quickly find proper responses from the user manuals via explicit clue chains. Specifically, each of the clue chains is formed by inferring over the user manuals, starting from the question clue aligned with the user question and ending at a possible response. To overcome the shortage of supervised data, we adopt the self-supervised strategy for model learning. The offline experiment shows that CARE is efficient in automatically inferring accurate responses from the user manual. The online experiment further demonstrates the superiority of CARE to reduce CSRs' reading burden and keep high service quality, in particular with >35% decrease in time spent and keeping a >0.75 ICC score.

cs.CL↗

arXiv2024

NeR-VCP: A Video Content Protection Method Based on Implicit Neural Representation

With the popularity of video applications, the security of video content has emerged as a pressing issue that demands urgent attention. Most video content protection methods mainly rely on encryption technology, which needs to be manually designed or implemented in an experience-based manner. To address this problem, we propose an automatic encryption technique for video content protection based on implicit neural representation. We design a key-controllable module, which serves as a key for encryption and decryption. NeR-VCP first pre-distributes the key-controllable module trained by the sender to the recipients, and then uses Implicit Neural Representation (INR) with a (pre-distributed) key-controllable module to encrypt plain video as an implicit neural network, and the legal recipients uses a pre-distributed key-controllable module to decrypt this cipher neural network (the corresponding implicit neural network). Under the guidance of the key-controllable design, our method can improve the security of video content and provide a novel video encryption scheme. Moreover, using model compression techniques, this method can achieve video content protection while effectively mitigating the amount of encrypted data transferred. We experimentally find that it has superior performance in terms of visual representation, imperceptibility to illegal users, and security from a cryptographic viewpoint.

cs.CR↗

arXiv2024

Geometric Pooling: maintaining more useful information

Graph Pooling technology plays an important role in graph node classification tasks. Sorting pooling technologies maintain large-value units for pooling graphs of varying sizes. However, by analyzing the statistical characteristic of activated units after pooling, we found that a large number of units dropped by sorting pooling are negative-value units that contain useful information and can contribute considerably to the final decision. To maintain more useful information, a novel pooling technology, called Geometric Pooling (GP), was proposed to contain the unique node features with negative values by measuring the similarity of all node features. We reveal the effectiveness of GP from the entropy reduction view. The experiments were conducted on TUdatasets to show the effectiveness of GP. The results showed that the proposed GP outperforms the SOTA graph pooling technologies by 1%\sim5% with fewer parameters.

cs.CV↗

arXiv2024

Muon g-2, Long-Range Muon Spin Force, and Neutrino Oscillations

Recent studies have proposed using a geocentric muon spin force to account for the $(g-2)_μ$ anomaly, with the long-range force mediator being a light axion-like particle. The mediator exhibits a CP-violating scalar coupling to nucleons and a normal derivative coupling to muons. Due to the weak symmetry, this axion inevitably couples to neutrinos, providing potential impact on neutrino oscillations. By utilizing neutrino data from BOREXINO, IceCube DeepCore, Super-Kamiokande, and SNO, we have identified that both atmospheric and solar neutrino data can impose stringent constraints on the long-range muon spin force model and the $(g-2)_μ$ parameter space. With optimized data analysis techniques and the potential from future experiments, such as JUNO, Hyper-Kamiokande, SNO+, and IceCube PINGU, there exists a promising opportunity to achieve even greater sensitivities. Indeed, neutrino oscillations offer a robust and distinctive cross-check for the model, offering stringent constraints on the $(g-2)_μ$ parameter space.

hep-ph↗

arXiv2024

Object Space is Embodied

The perceived similarity between objects has often been attributed to their physical and conceptual features, such as appearance and animacy, and the theoretical framework of object space is accordingly conceived. Here, we extend this framework by proposing that object space may also be defined by embodied features, specifically action possibilities that objects afford to an agent (i.e., affordance) and their spatial relation with the agent (i.e., situatedness). To test this proposal, we quantified the embodied features with a set of action atoms. We found that embodied features explained the subjective similarity among familiar objects along with the objects' visual features. This observation was further replicated with novel objects. Our study demonstrates that embodied features, which place objects within an ecological context, are essential in constructing object space in the human visual system, emphasizing the importance of incorporating embodiment as a fundamental dimension in our understanding of the visual world.

q-bio.NC↗