Search

3D Gaussian Splatting (3D-GS) has made a notable advancement in the field of neural rendering, 3D scene reconstruction, and novel view synthesis. Nevertheless, 3D-GS encounters the main challenge when it comes to accurately representing physical reflections, especially in the case of total reflection and semi-reflection that are commonly found in real-world scenes. This limitation causes reflections to be mistakenly treated as independent elements with physical presence, leading to imprecise reconstructions. Herein, to tackle this challenge, we propose RefGaussian to disentangle reflections from 3D-GS for realistically modeling reflections. Specifically, we propose to split a scene into transmitted and reflected components and represent these components using two Spherical Harmonics (SH). Given that this decomposition is not fully determined, we employ local regularization techniques to ensure local smoothness for both the transmitted and reflected components, thereby achieving more plausible decomposition outcomes than 3D-GS. Experimental results demonstrate that our approach achieves superior novel view synthesis and accurate depth estimation outcomes. Furthermore, it enables the utilization of scene editing applications, ensuring both high-quality results and physical coherence.

Time series self-supervised learning (SSL) aims to exploit unlabeled data for pre-training to mitigate the reliance on labels. Despite the great success in recent years, there is limited discussion on the potential noise in the time series, which can severely impair the performance of existing SSL methods. To mitigate the noise, the de facto strategy is to apply conventional denoising methods before model training. However, this pre-processing approach may not fully eliminate the effect of noise in SSL for two reasons: (i) the diverse types of noise in time series make it difficult to automatically determine suitable denoising methods; (ii) noise can be amplified after mapping raw data into latent space. In this paper, we propose denoising-aware contrastive learning (DECL), which uses contrastive learning objectives to mitigate the noise in the representation and automatically selects suitable denoising methods for every sample. Extensive experiments on various datasets verify the effectiveness of our method. The code is open-sourced., Comment: Accepted to 33rd International Joint Conference on Artificial Intelligence (IJCAI 2024)

Overfitting in RL has become one of the main obstacles to applications in reinforcement learning(RL). Existing methods do not provide explicit semantic constrain for the feature extractor, hindering the agent from learning a unified cross-domain representation and resulting in performance degradation on unseen domains. Besides, abundant data from multiple domains are needed. To address these issues, in this work, we propose prompt-based visual alignment (PVA), a robust framework to mitigate the detrimental domain bias in the image for zero-shot policy transfer. Inspired that Visual-Language Model (VLM) can serve as a bridge to connect both text space and image space, we leverage the semantic information contained in a text sequence as an explicit constraint to train a visual aligner. Thus, the visual aligner can map images from multiple domains to a unified domain and achieve good generalization performance. To better depict semantic information, prompt tuning is applied to learn a sequence of learnable tokens. With explicit constraints of semantic information, PVA can learn unified cross-domain representation under limited access to cross-domain data and achieves great zero-shot generalization ability in unseen domains. We verify PVA on a vision-based autonomous driving task with CARLA simulator. Experiments show that the agent generalizes well on unseen domains under limited access to multi-domain data., Comment: This paper has been accepted by ICML2024

Spatiotemporal data is prevalent in a wide range of edge devices, such as those used in personal communication and financial transactions. Recent advancements have sparked a growing interest in integrating spatiotemporal analysis with large-scale language models. However, spatiotemporal data often contains sensitive information, making it unsuitable for open third-party access. To address this challenge, we propose a Graph-GAN-based model for generating privacy-protected spatiotemporal data. Our approach incorporates spatial and temporal attention blocks in the discriminator and a spatiotemporal deconvolution structure in the generator. These enhancements enable efficient training under Gaussian noise to achieve differential privacy. Extensive experiments conducted on three real-world spatiotemporal datasets validate the efficacy of our model. Our method provides a privacy guarantee while maintaining the data utility. The prediction model trained on our generated data maintains a competitive performance compared to the model trained on the original data.

We propose a theoretical scheme to enhance the sensitivity of a quantum fiber-optical gyroscope (QFOG) via a non-Gaussian-state probe based on quadrature measurements of the optical field. The non-Gaussian-state probe utilizes the product state comprising a photon-added coherent state (PACS) with photon excitations and a coherent state CS. We study the sensitivity of the QFOG, and find that it can be significantly enhanced through increasing the photon excitations in the PACS probe. We investigate the influence of photon loss on the performance of QFOG and demonstrate that the PACS probe exhibits robust resistance to photon loss. Furthermore, we compare the performance of the QFOG using the PACS probe against two Gaussian-state probes: the CS probe and the squeezed state (SS) probe and indicate that the PACS probe offers a significant advantage in terms of sensitivity, regardless of photon loss, under the constraint condition of the same total number of input photons. Particularly, it is found that the sensitivity of the PACS probe can be three orders of magnitude higher than that of two Gaussian-state probes for certain values of the measured parameter. The capabilities of the non-Gaussian state probe on enhancing the sensitivity and resisting photon loss could have a wide-ranging impact on future high-performance QFOGs., Comment: 10 pages, 8 figures

Addressing the challenge of effectively processing long contexts has become a critical issue for Large Language Models (LLMs). Two common strategies have emerged: 1) reducing the input length, such as retrieving relevant chunks by Retrieval-Augmented Generation (RAG), and 2) expanding the context window limit of LLMs. However, both strategies have drawbacks: input reduction has no guarantee of covering the part with needed information, while window extension struggles with focusing on the pertinent information for solving the task. To mitigate these limitations, we propose Chain-of-Agents (CoA), a novel framework that harnesses multi-agent collaboration through natural language to enable information aggregation and context reasoning across various LLMs over long-context tasks. CoA consists of multiple worker agents who sequentially communicate to handle different segmented portions of the text, followed by a manager agent who synthesizes these contributions into a coherent final output. CoA processes the entire input by interleaving reading and reasoning, and it mitigates long context focus issues by assigning each agent a short context. We perform comprehensive evaluation of CoA on a wide range of long-context tasks in question answering, summarization, and code completion, demonstrating significant improvements by up to 10% over strong baselines of RAG, Full-Context, and multi-agent LLMs., Comment: 19 pages, 6 figures

Self-correction is an approach to improving responses from large language models (LLMs) by refining the responses using LLMs during inference. Prior work has proposed various self-correction frameworks using different sources of feedback, including self-evaluation and external feedback. However, there is still no consensus on the question of when LLMs can correct their own mistakes, as recent studies also report negative results. In this work, we critically survey broad papers and discuss the conditions required for successful self-correction. We first find that prior studies often do not define their research questions in detail and involve impractical frameworks or unfair evaluations that over-evaluate self-correction. To tackle these issues, we categorize research questions in self-correction research and provide a checklist for designing appropriate experiments. Our critical survey based on the newly categorized research questions shows that (1) no prior work demonstrates successful self-correction with feedback from prompted LLMs, except for studies in tasks that are exceptionally suited for self-correction, (2) self-correction works well in tasks that can use reliable external feedback, and (3) large-scale fine-tuning enables self-correction., Comment: Accepted for publication in Transactions of the Association for Computational Linguistics (TACL), 2024. Author's final version

We present a direct lattice QCD calculation of the $x$-dependence of the pion distribution amplitude (DA), which is performed using the quasi-DA in large momentum effective theory on a domain-wall fermion ensemble at physical quark masses and spacing $a\approx 0.084$ fm. The bare quais-DA matrix elements are renormalized in the hybrid scheme and matched to $\overline{\rm MS}$ with a subtraction of the leading renormalon in the Wilson-line mass. For the first time, we include threshold resummation in the perturbative matching onto the light-cone DA, which resums the large logarithms in the soft gluon limit at next-to-next-to-leading log. The resummed results show controlled scale-variation uncertainty within the range of momentum fraction $x\in[0.25,0.75]$ at the largest pion momentum $P_z\approx 1.85$~GeV. In addition, we apply the same analysis to quasi-DAs from a highly-improved-staggered-quark ensemble at physical pion mass and $a=0.076$ fm. By comparison we find with $2\sigma$ confidence level that the DA obtained from chiral fermions is flatter and lower near $x=0.5$., Comment: update to match the version published in journal

Optical intelligent reflecting surface (OIRS) offers a new and effective approach to resolving the line-of-sight blockage issue in visible light communication (VLC) by enabling redirection of light to bypass obstacles, thereby dramatically enhancing indoor VLC coverage and reliability. This article provides a comprehensive overview of OIRS for VLC, including channel modeling, design techniques, and open issues. First, we present the characteristics of OIRS-reflected channels and introduce two practical models, namely, optics model and association model, which are then compared in terms of applicable conditions, configuration methods, and channel parameters. Next, under the more practically appealing association model, we discuss the main design techniques for OIRS-aided VLC systems, including beam alignment, channel estimation, and OIRS reflection optimization. Finally, open issues are identified to stimulate future research in this area.

Video recognition remains an open challenge, requiring the identification of diverse content categories within videos. Mainstream approaches often perform flat classification, overlooking the intrinsic hierarchical structure relating categories. To address this, we formalize the novel task of hierarchical video recognition, and propose a video-language learning framework tailored for hierarchical recognition. Specifically, our framework encodes dependencies between hierarchical category levels, and applies a top-down constraint to filter recognition predictions. We further construct a new fine-grained dataset based on medical assessments for rehabilitation of stroke patients, serving as a challenging benchmark for hierarchical recognition. Through extensive experiments, we demonstrate the efficacy of our approach for hierarchical recognition, significantly outperforming conventional methods, especially for fine-grained subcategories. The proposed framework paves the way for hierarchical modeling in video understanding tasks, moving beyond flat categorization.

Biological systems, particularly the brain, are frequently analyzed as networks, conveying mechanistic insights into their function and pathophysiology. This is the first study of a functional network of cardiac tissue. We use calcium imaging to obtain two functional networks in a subsidiary but essential pacemaker of the heart, the atrioventricular node (AVN). The AVN is a small cellular structure with dual functions: a) to delay the pacemaker signal passing from the sinoatrial node (SAN) to the ventricles, and b) to serve as a back-up pacemaker should the primary SAN pacemaker fail. Failure of the AVN can lead to syncope and death. We found that the shortest path lengths and clustering coefficients of the AVN are remarkably similar to those of the brain. The network is ``small-world," thus optimized for energy use vs transmission efficiency. We further study the network properties of AVN tissue with knock-out of the sodium-calcium exchange transporter. In this case, the average shortest path-lengths remained nearly unchanged showing network resilience, while the clustering coefficient was somewhat reduced, similar to schizophrenia in brain networks. When we removed the global action potential using principal component analysis (PCA) in wild-type model, the network lost its ``small-world" characteristics with less information-passing efficiency due to longer shortest path lengths but more robust signal propagation resulting from higher clustering. These two wild-type networks (with and without global action potential) may correspond to fast and slow conduction pathways. Laslty, a one-parameter non-linear preferential attachment model is a good fit to all three AVN networks., Comment: 14 pages, 8 figures at the end of the manuscript, two videos and three datasets in the Source folder

Prompt tuning based on Context Optimization (CoOp) effectively adapts visual-language models (VLMs) to downstream tasks by inferring additional learnable prompt tokens. However, these tokens are less discriminative as they are independent of the pre-trained tokens and fail to capture input-specific knowledge, such as class-aware textual or instance-aware visual knowledge. Leveraging the discriminative and generalization capabilities inherent in pre-trained tokens, we introduce a novel approach named Self-Enhanced Prompt Tuning (SEP). The core principle of SEP involves adapting the learnable prompt tokens at each encoder layer from the corresponding self-pretrained tokens, thereby explicitly incorporating discriminative prior knowledge to enhance both textual-level and visual-level embeddings. Furthermore, SEP's self-enhanced tokens not only boost discrimination but also mitigate domain shifts in unseen domains, enhancing generalization. In practice, SEP selects several representative tokens from all pre-trained tokens for each input data at every layer of the text/visual encoders. Subsequently, a Token Fusion Module (TFM) is introduced to generate a self-enhanced token by merging these representative tokens with the learnable tokens using a cross-attention mechanism. This self-enhanced token is then concatenated with all pre-trained tokens, serving as input for subsequent encoder layers to produce the relevant embeddings. Comprehensive evaluations across various benchmarks and tasks confirm SEP's efficacy in prompt tuning. Code: \href{Code}{https://github.com/htyao89/SEP}.

Building open agents has always been the ultimate goal in AI research, and creative agents are the more enticing. Existing LLM agents excel at long-horizon tasks with well-defined goals (e.g., `mine diamonds' in Minecraft). However, they encounter difficulties on creative tasks with open goals and abstract criteria due to the inability to bridge the gap between them, thus lacking feedback for self-improvement in solving the task. In this work, we introduce autonomous embodied verification techniques for agents to fill the gap, laying the groundwork for creative tasks. Specifically, we propose the Luban agent target creative building tasks in Minecraft, which equips with two-level autonomous embodied verification inspired by human design practices: (1) visual verification of 3D structural speculates, which comes from agent synthesized CAD modeling programs; (2) pragmatic verification of the creation by generating and verifying environment-relevant functionality programs based on the abstract criteria. Extensive multi-dimensional human studies and Elo ratings show that the Luban completes diverse creative building tasks in our proposed benchmark and outperforms other baselines ($33\%$ to $100\%$) in both visualization and pragmatism. Additional demos on the real-world robotic arm show the creation potential of the Luban in the physical world.

We propose a new simple and explicit numerical scheme for time-homogeneous stochastic differential equations. The scheme is based on sampling increments at each time step from a skew-symmetric probability distribution, with the level of skewness determined by the drift and volatility of the underlying process. We show that as the step-size decreases the scheme converges weakly to the diffusion of interest, and also prove path-wise accuracy in a particular setting. We then consider the problem of simulating from the limiting distribution of an ergodic diffusion process using the numerical scheme with a fixed step-size. We establish conditions under which the numerical scheme converges to equilibrium at a geometric rate, and quantify the bias between the equilibrium distributions of the scheme and of the true diffusion process. Notably, our results do not require a global Lipschitz assumption on the drift, in contrast to those required for the Euler--Maruyama scheme for long-time simulation at fixed step-sizes. Our weak convergence result relies on an extension of the theory of Milstein \& Tretyakov to stochastic differential equations with non-Lipschitz drift, which could also be of independent interest. We support our theoretical results with numerical simulations., Comment: 48 pages, 3 figures Keywords: Stochastic differential equations, Skew-symmetric distributions, Sampling algorithms, Markov Chain Monte Carlo

In this paper, we study an IRS-assisted coverage enhancement problem for a given region, aiming to optimize the passive reflection of the IRS for improving the average communication performance in the region by accounting for both deterministic and random channels in the environment. To this end, we first derive the closed-form expression of the average received signal power in terms of the deterministic base station (BS)-IRS-user cascaded channels over all user locations, and propose an IRS-aided coverage enhancement framework to facilitate the estimation of such deterministic channels for IRS passive reflection design. Specifically, to avoid the exorbitant overhead of estimating the cascaded channels at all possible user locations, a location selection method is first proposed to select only a set of typical user locations for channel estimation by exploiting the channel spatial correlation in the region. To estimate the deterministic cascaded channels at the selected user locations, conventional IRS channel estimation methods require additional pilot signals, which not only results in high system training overhead but also may not be compatible with the existing communication protocols. To overcome this issue, we further propose a single-layer neural network (NN)-enabled IRS channel estimation method in this paper, based on only the average received signal power measurements at each selected location corresponding to different IRS random training reflections, which can be offline implemented in current wireless systems. Numerical results demonstrate that our proposed scheme can significantly improve the coverage performance of the target region and outperform the existing power-measurement-based IRS reflection designs., Comment: arXiv admin note: text overlap with arXiv:2309.08275

Large language models (LLM) have demonstrated remarkable capabilities in various biomedical natural language processing (NLP) tasks, leveraging the demonstration within the input context to adapt to new tasks. However, LLM is sensitive to the selection of demonstrations. To address the hallucination issue inherent in LLM, retrieval-augmented LLM (RAL) offers a solution by retrieving pertinent information from an established database. Nonetheless, existing research work lacks rigorous evaluation of the impact of retrieval-augmented large language models on different biomedical NLP tasks. This deficiency makes it challenging to ascertain the capabilities of RAL within the biomedical domain. Moreover, the outputs from RAL are affected by retrieving the unlabeled, counterfactual, or diverse knowledge that is not well studied in the biomedical domain. However, such knowledge is common in the real world. Finally, exploring the self-awareness ability is also crucial for the RAL system. So, in this paper, we systematically investigate the impact of RALs on 5 different biomedical tasks (triple extraction, link prediction, classification, question answering, and natural language inference). We analyze the performance of RALs in four fundamental abilities, including unlabeled robustness, counterfactual robustness, diverse robustness, and negative awareness. To this end, we proposed an evaluation framework to assess the RALs' performance on different biomedical NLP tasks and establish four different testbeds based on the aforementioned fundamental abilities. Then, we evaluate 3 representative LLMs with 3 different retrievers on 5 tasks over 9 datasets.

Six-dimensional movable antenna (6DMA) is an effective solution for enhancing wireless network capacity through the adjustment of both 3D positions and 3D rotations of distributed antennas/antenna surfaces. Although freely positioning/rotating 6DMA surfaces offers the greatest flexibility and thus highest capacity improvement, its implementation may be challenging in practice due to the drastic architecture change required for existing base stations (BSs), which predominantly adopt fixed-position antenna (FPA) arrays (e.g., sector antenna arrays). Thus, we introduce in this letter a new BS architecture called hybrid fixed and movable antennas (HFMA), which consists of both conventional FPA arrays and position/rotation-adjustable 6DMA surfaces. For ease of implementation, we consider that all 6DMA surfaces can rotate along a circular track above the FPA arrays. We aim to maximize the network capacity via optimizing the rotation angles of all 6DMA surfaces based on the users' spatial distribution. Since this problem is combinatorial and its optimal solution requires prohibitively high computational complexity via exhaustive search, we propose an alternative adaptive Markov Chain Monte Carlo based method to solve it more efficiently. Finally, we present simulation results that show significant performance gains achieved by our proposed design over various benchmark schemes.

Electronic countermeasure (ECM) technology plays a critical role in modern electronic warfare, which can interfere with enemy radar detection systems by noise or deceptive signals. However, the conventional active jamming strategy incurs additional hardware and power costs and has the potential threat of exposing the target itself. To tackle the above challenges, we propose a new intelligent reflecting surface (IRS)-aided radar spoofing strategy in this letter, where IRS is deployed on the surface of a target to help eliminate the signals reflected towards the hostile radar to shield the target, while simultaneously redirecting its reflected signal towards a surrounding clutter to generate deceptive angle-of-arrival (AoA) sensing information for the radar. We optimize the IRS's reflection to maximize the received signal power at the radar from the direction of the selected clutter subject to the constraint that its received power from the direction of the target is lower than a given detection threshold. We first solve this non-convex optimization problem using the semidefinite relaxation (SDR) method and further propose a lower-complexity solution for real-time implementation. Simulation results validate the efficacy of our proposed IRS-aided spoofing system as compared to various benchmark schemes., Comment: 5 pages, 4 figures

Personalized recommendation stands as a ubiquitous channel for users to explore information or items aligned with their interests. Nevertheless, prevailing recommendation models predominantly rely on unique IDs and categorical features for user-item matching. While this ID-centric approach has witnessed considerable success, it falls short in comprehensively grasping the essence of raw item contents across diverse modalities, such as text, image, audio, and video. This underutilization of multimodal data poses a limitation to recommender systems, particularly in the realm of multimedia services like news, music, and short-video platforms. The recent surge in pretraining and generation techniques presents both opportunities and challenges in the development of multimodal recommender systems. This tutorial seeks to provide a thorough exploration of the latest advancements and future trajectories in multimodal pretraining and generation techniques within the realm of recommender systems. The tutorial comprises three parts: multimodal pretraining, multimodal generation, and industrial applications and open challenges in the field of recommendation. Our target audience encompasses scholars, practitioners, and other parties interested in this domain. By providing a succinct overview of the field, we aspire to facilitate a swift understanding of multimodal recommendation and foster meaningful discussions on the future development of this evolving landscape., Comment: Published in WWW 2024 Tutorial. Find the tutorial materials at https://mmrec.github.io/tutorial/www2024/

Large Language Models (LLMs) have exhibited remarkable proficiency across a wide array of NLP tasks. However, the escalation in model size also engenders substantial deployment costs. While few efforts have explored model pruning techniques to reduce the size of LLMs, they mainly center on general or task-specific weights. This leads to suboptimal performance due to lacking specificity on the target domain or generality on different tasks when applied to domain-specific challenges. This work introduces an innovative unstructured dual-pruning methodology, D-Pruner, for domain-specific compression on LLM. It extracts a compressed, domain-specific, and task-agnostic LLM by identifying LLM weights that are pivotal for general capabilities, like linguistic capability and multi-task solving, and domain-specific knowledge. More specifically, we first assess general weight importance by quantifying the error incurred upon their removal with the help of an open-domain calibration dataset. Then, we utilize this general weight importance to refine the training loss, so that it preserves generality when fitting into a specific domain. Moreover, by efficiently approximating weight importance with the refined training loss on a domain-specific calibration dataset, we obtain a pruned model emphasizing generality and specificity. Our comprehensive experiments across various tasks in healthcare and legal domains show the effectiveness of D-Pruner in domain-specific compression. Our code is available at https://github.com/psunlpgroup/D-Pruner., Comment: NAACL 2024 Findings

In this paper, we propose a new wireless sensing system equipped with the movable-antenna (MA) array, which can flexibly adjust the positions of antenna elements for improving the sensing performance over conventional antenna arrays with fixed-position antennas (FPAs). First, we show that the angle estimation performance in wireless sensing is fundamentally determined by the array geometry, where the Cramer-Rao bound (CRB) of the mean square error (MSE) for angle of arrival (AoA) estimation is derived as a function of the antennas' positions for both one-dimensional (1D) and two-dimensional (2D) MA arrays. Then, for the case of 1D MA array, we obtain a globally optimal solution for the MAs' positions in closed form to minimize the CRB of AoA estimation MSE. While in the case of 2D MA array, we aim to achieve the minimum of maximum (min-max) CRBs of estimation MSE for the two AoAs with respect to the horizontal and vertical axes, respectively. In particular, for the special case of circular antenna movement region, an optimal solution for the MAs' positions is derived under certain numbers of MAs and circle radii. Thereby, both the lower- and upper-bounds of the min-max CRB are obtained for the antenna movement region with arbitrary shapes. Moreover, we develop an efficient alternating optimization algorithm to obtain a locally optimal solution for MAs' positions by iteratively optimizing one between their horizontal and vertical coordinates with the other being fixed. Numerical results demonstrate that our proposed 1D/2D MA arrays can significantly decrease the CRB of AoA estimation MSE as well as the actual MSE compared to conventional uniform linear arrays (ULAs)/uniform planar arrays (UPAs) with different values of uniform inter-antenna spacing., Comment: 13 pages, 13 figures. We propose a new wireless sensing system equipped with the movable-antenna (MA) array, which can flexibly adjust the positions of antenna elements for improving the sensing performance over conventional antenna arrays with fixed-position antennas (FPAs)

Regression-based sEMG prosthetic hands are widely used for their ability to provide continuous kinetic and kinematic parameters. However, establishing these models requires complex sensors systems to collect corresponding kinetic and kinematic data in synchronization with sEMG, which is cumbersome and user-unfriendly. This paper proposes a kinetic and kinematic sensors-free approach for controlling sEMG prosthetic hands, enabling continuous decoding and execution of three hand movements: individual finger flexion/extension, multiple finger flexion/extension, and fist opening/closing. This approach utilizes only two data points (-1 and 1), representing maximal finger flexion force label and extension force label respectively, and their corresponding sEMG data to establish a near-linear model based on sEMG data and labels. The model's output labels values are used to control the direction and magnitude of fingers forces, enabling the estimation of continuous gestures. To validate this approach, we conducted offline and online experiments using four models: Dendritic Net (DD), Linear Net (LN), Multi-Layer Perceptron (MLP), and Convolutional Neural Network (CNN). The offline analysis assessed each model's ability to classify finger force direction and interpolate intermediate force values, while online experiments evaluated real-time control performance in controlling gestures and accurately adjusting forces. Our results demonstrate that the DD and LN models provide excellent real-time control of finger forces and gestures, highlighting the practical potential of this sensors-free approach for prosthetic applications. This study significantly reduces the complexity of collecting kinetic and kinematic parameters in sEMG-based regression prosthetics, thus enhancing the usability and convenience of prosthetic hands., Comment: 17 pages

Large Language Models (LLMs) have swiftly emerged as vital resources for different applications in the biomedical and healthcare domains; however, these models encounter issues such as generating inaccurate information or hallucinations. Retrieval-augmented generation provided a solution for these models to update knowledge and enhance their performance. In contrast to previous retrieval-augmented LMs, which utilize specialized cross-attention mechanisms to help LLM encode retrieved text, BiomedRAG adopts a simpler approach by directly inputting the retrieved chunk-based documents into the LLM. This straightforward design is easily applicable to existing retrieval and language models, effectively bypassing noise information in retrieved documents, particularly in noise-intensive tasks. Moreover, we demonstrate the potential for utilizing the LLM to supervise the retrieval model in the biomedical domain, enabling it to retrieve the document that assists the LM in improving its predictions. Our experiments reveal that with the tuned scorer,\textsc{ BiomedRAG} attains superior performance across 5 biomedical NLP tasks, encompassing information extraction (triple extraction, relation extraction), text classification, link prediction, and question-answering, leveraging over 9 datasets. For instance, in the triple extraction task, \textsc{BiomedRAG} outperforms other triple extraction systems with micro-F1 scores of 81.42 and 88.83 on GIT and ChemProt corpora, respectively.

Like electrons, Cooper pairs can carry a monopole charge if the pairing electrons come from two or more Fermi surfaces with different Chern numbers. In such an instance, a superconductor is necessarily nodal due to an inherent topological pairing obstruction. In this work, we show that a similar obstruction is also possible when there is only one Fermi surface involved in the pairing process. By developing a Chern-vorticity theorem, we have identified a class of Fermi surfaces with a quantized dipolar Berry flux pattern, where all intra-Fermi-surface Cooper pairings are ``dipole-obstructed" and nodal. As a real-world application, we find that the dipole obstruction plays a crucial role in stabilizing the superconducting nodal structure for $j=3/2$ half-Heusler compounds., Comment: 6+9 pages, 3+5 figures

Diffusion based text-to-image models are trained on large datasets scraped from the Internet, potentially containing unacceptable concepts (e.g., copyright infringing or unsafe). We need concept removal techniques (CRTs) which are effective in preventing the generation of images with unacceptable concepts, utility-preserving on acceptable concepts, and robust against evasion with adversarial prompts. None of the prior CRTs satisfy all these requirements simultaneously. We introduce Espresso, the first robust concept filter based on Contrastive Language-Image Pre-Training (CLIP). We configure CLIP to identify unacceptable concepts in generated images using the distance of their embeddings to the text embeddings of both unacceptable and acceptable concepts. This lets us fine-tune for robustness by separating the text embeddings of unacceptable and acceptable concepts while preserving their pairing with image embeddings for utility. We present a pipeline to evaluate various CRTs, attacks against them, and show that Espresso, is more effective and robust than prior CRTs, while retaining utility.

Due to the ultra-dense constellation, efficient beam coverage and interference mitigation are crucial to low-earth orbit (LEO) satellite communication systems, while the conventional directional antennas and fixed-position antenna (FPA) arrays both have limited degrees of freedom (DoFs) in beamforming to adapt to the time-varying coverage requirement of terrestrial users. To address this challenge, we propose in this paper utilizing movable antenna (MA) arrays to enhance the satellite beam coverage and interference mitigation. Specifically, given the satellite orbit and the coverage requirement within a specific time interval, the antenna position vector (APV) and antenna weight vector (AWV) of the satellite-mounted MA array are jointly optimized over time to minimize the average signal leakage power to the interference area of the satellite, subject to the constraints of the minimum beamforming gain over the coverage area, the continuous movement of MAs, and the constant modulus of AWV. The corresponding continuous-time decision process for the APV and AWV is first transformed into a more tractable discrete-time optimization problem. Then, an alternating optimization (AO)-based algorithm is developed by iteratively optimizing the APV and AWV, where the successive convex approximation (SCA) technique is utilized to obtain locally optimal solutions during the iterations. Moreover, to further reduce the antenna movement overhead, a low-complexity MA scheme is proposed by using an optimized common APV over all time slots. Simulation results validate that the proposed MA array-aided beam coverage schemes can significantly decrease the interference leakage of the satellite compared to conventional FPA-based schemes, while the low-complexity MA scheme can achieve a performance comparable to the continuous-movement scheme.

Mainstream backdoor attacks on large language models (LLMs) typically set a fixed trigger in the input instance and specific responses for triggered queries. However, the fixed trigger setting (e.g., unusual words) may be easily detected by human detection, limiting the effectiveness and practicality in real-world scenarios. To enhance the stealthiness of backdoor activation, we present a new poisoning paradigm against LLMs triggered by specifying generation conditions, which are commonly adopted strategies by users during model inference. The poisoned model performs normally for output under normal/other generation conditions, while becomes harmful for output under target generation conditions. To achieve this objective, we introduce BrieFool, an efficient attack framework. It leverages the characteristics of generation conditions by efficient instruction sampling and poisoning data generation, thereby influencing the behavior of LLMs under target conditions. Our attack can be generally divided into two types with different targets: Safety unalignment attack and Ability degradation attack. Our extensive experiments demonstrate that BrieFool is effective across safety domains and ability domains, achieving higher success rates than baseline methods, with 94.3 % on GPT-3.5-turbo

Optical intelligent reflecting surface (OIRS) has attracted increasing attention due to its capability of overcoming signal blockages in visible light communication (VLC), an emerging technology for the next-generation advanced transceivers. However, current works on OIRS predominantly assume known channel state information (CSI), which is essential to practical OIRS configuration. To bridge such a gap, this paper proposes a new and customized channel estimation protocol for OIRSs under the alignment-based channel model. Specifically, we first unveil OIRS spatial and temporal coherence characteristics and derive the coherence distance and the coherence time in closed form. Next, to achieve fast beam alignment over different coherence time, we propose to dynamically tune the rotational angles of the OIRS reflecting elements following a geometric optics-based non-uniform codebook. Given the above beam alignment, we propose an efficient joint space-time sampling-based algorithm to estimate the OIRS channel. In particular, we divide the OIRS into multiple subarrays based on the coherence distance and sequentially estimate their associated CSI, followed by a spacetime interpolation to retrieve full CSI for other non-aligned transceiver antennas. Numerical results validate our theoretical analyses and demonstrate the efficacy of our proposed OIRS channel estimation scheme as compared to other benchmark schemes.

Optical intelligent reflecting surface (OIRS) has been considered a promising technology for visible light communication (VLC) by constructing visual line-of-sight propagation paths to address the signal blockage issue. However, the existing works on OIRSs are mostly based on perfect channel state information (CSI), whose acquisition appears to be challenging due to the passive nature of the OIRS. To tackle this challenge, this paper proposes a customized channel estimation algorithm for OIRSs. Specifically, we first unveil the OIRS spatial coherence characteristics and derive the coherence distance in closed form. Based on this property, a spatial sampling-based algorithm is proposed to estimate the OIRS-reflected channel, by dividing the OIRS into multiple subarrays based on the coherence distance and sequentially estimating their associated CSI, followed by an interpolation to retrieve the full CSI. Simulation results validate the derived OIRS spatial coherence and demonstrate the efficacy of the proposed OIRS channel estimation algorithm.