Search

In this paper, we explore a principal way to enhance the quality of widely pre-existing coarse masks, enabling them to serve as reliable training data for segmentation models to reduce the annotation cost. In contrast to prior refinement techniques that are tailored to specific models or tasks in a close-world manner, we propose SAMRefiner, a universal and efficient approach by adapting SAM to the mask refinement task. The core technique of our model is the noise-tolerant prompting scheme. Specifically, we introduce a multi-prompt excavation strategy to mine diverse input prompts for SAM (i.e., distance-guided points, context-aware elastic bounding boxes, and Gaussian-style masks) from initial coarse masks. These prompts can collaborate with each other to mitigate the effect of defects in coarse masks. In particular, considering the difficulty of SAM to handle the multi-object case in semantic segmentation, we introduce a split-then-merge (STM) pipeline. Additionally, we extend our method to SAMRefiner++ by introducing an additional IoU adaption step to further boost the performance of the generic SAMRefiner on the target dataset. This step is self-boosted and requires no additional annotation. The proposed framework is versatile and can flexibly cooperate with existing segmentation methods. We evaluate our mask framework on a wide range of benchmarks under different settings, demonstrating better accuracy and efficiency. SAMRefiner holds significant potential to expedite the evolution of refinement tools. Our code is available at https://github.com/linyq2117/SAMRefiner., Comment: Accepted to ICLR 2025

How to interact with LLMs through \emph{instructions} has been widely studied by researchers. However, previous studies have treated the emergence of instructions and the training of LLMs on task data as separate processes, overlooking the inherent unity between the two. This paper proposes a neural network framework, VaiBot, that integrates VAE and VIB, designed to uniformly model, learn, and infer both deduction and induction tasks under LLMs. Through experiments, we demonstrate that VaiBot performs on par with existing baseline methods in terms of deductive capabilities while significantly surpassing them in inductive capabilities. We also find that VaiBot can scale up using general instruction-following data and exhibits excellent one-shot induction abilities. We finally synergistically integrate the deductive and inductive processes of VaiBot. Through T-SNE dimensionality reduction, we observe that its inductive-deductive process significantly improves the distribution of training parameters, enabling it to outperform baseline methods in inductive reasoning tasks. The code and data for this paper can be found at https://anonymous.4open.science/r/VaiBot-021F.

Superconductivity emerges in both La$_3$Ni$_2$O$_7$ and La$_4$Ni$_3$O$_{10}$ under high pressure by suppressing their density-wave transitions, but critical temperature (Tc) differs significantly between these two compounds. To gain deeper insights into the distinct superconducting states, it is essential to unravel the nature of the density-wave states at ambient pressure, a topic that remains largely unexplored. Here, using scanning tunneling microscopy/spectroscopy (STM/STS), we report the direct visualization of an incommensurate unidirectional charge density wave (CDW) in La$_4$Ni$_3$O$_{10}$ in real space. The density of states (DOS) is strongly depleted near $E_F$, indicating the opening of a CDW gap of $2{\Delta} \approx 71$ meV, which is unfavorable for the formation of superconductivity at ambient pressure. We propose that the CDW arises from Fermi surface nesting, and is likely a subsidiary phase of a spin density wave. Compared to La$_3$Ni$_2$O$_7$, the weaker electronic correlation in La$_4$Ni$_3$O$_{10}$ is likely one reason for the lower $T_c$., Comment: 11 pages, 4 figures

The discovery of superconductivity in pressurized bilayer and trilayer nickelates has generated significant interest. However, their superconducting properties are often dependent on sample quality and pressure conditions, complicating the interpretation of the underlying physics. Finding new systems with optimized bulk superconducting properties is therefore important for advancing our understanding of these materials. Unlike cupates, where trilayer compounds typically exhibit the highest transition temperature (Tc), the bilayer nickelate La3Ni2O7 has thus far outperformed the trilayer La4Ni3O10 in reported Tc. Whether the trilayer nickelates have achieved the optimal Tc remains unclear, with various scenarios suggesting different possibilities. Here, we report the discovery of bulk superconductivity in pressurized Pr4Ni3O10 single crystals, achieving a maximum onset Tc of 40.5 K at 80.1 GPa, significantly exceeding the 30 K observed in La4Ni3O10. The bulk nature of superconductivity is confirmed by zero resistance and a strong diamagnetic response below Tc with a superconducting volume fraction exceeding 80%. These findings establish trilayer nickelates as genuine bulk high-temperature superconductors, provide new insights into the mechanisms driving superconductivity, and point to a promising route toward further enhancing superconducting properties in nickelates., Comment: 16 pages, 4 figures

The classical Moment-Sum Of Squares hierarchy allows to approximate a global minimum of a polynomial optimization problem through semidefinite relaxations of increasing size. However, for many optimization instances, solving higher order relaxations becomes impractical or even impossible due to the substantial computational demands they impose. To address this, existing methods often exploit intrinsic problem properties, such as symmetries or sparsity. Here, we present a complementary approach, which enhances the accuracy of computationally more efficient low-order relaxations by leveraging Christoffel-Darboux kernels. Such strengthened relaxations often yield significantly improved bounds or even facilitate minimizer extraction. We illustrate the efficiency of our approach on several classes of important quadratically constrained quadratic Programs., Comment: 25 pages, 4 algorithms, 5 figures, 5 tables

Since the emergence of generative AI, creative workers have spoken up about the career-based harms they have experienced arising from this new technology. A common theme in these accounts of harm is that generative AI models are trained on workers' creative output without their consent and without giving credit or compensation to the original creators. This paper reports findings from 20 interviews with creative workers in three domains: visual art and design, writing, and programming. We investigate the gaps between current AI governance strategies, what creative workers want out of generative AI governance, and the nuanced role of creative workers' consent, compensation and credit for training AI models on their work. Finally, we make recommendations for how generative AI can be governed and how operators of generative AI systems might more ethically train models on creative output in the future., Comment: Conditionally accepted at the ACM Conference on Human Factors in Computing Systems (CHI 2025)

This article investigates the adaptive resource allocation scheme for digital twin (DT) synchronization optimization over dynamic wireless networks. In our considered model, a base station (BS) continuously collects factory physical object state data from wireless devices to build a real-time virtual DT system for factory event analysis. Due to continuous data transmission, maintaining DT synchronization must use extensive wireless resources. To address this issue, a subset of devices is selected to transmit their sensing data, and resource block (RB) allocation is optimized. This problem is formulated as a constrained Markov process (CMDP) problem that minimizes the long-term mismatch between the physical and virtual systems. To solve this CMDP, we first transform the problem into a dual problem that refines RB constraint impacts on device scheduling strategies. We then propose a continual reinforcement learning (CRL) algorithm to solve the dual problem. The CRL algorithm learns a stable policy across historical experiences for quick adaptation to dynamics in physical states and network capacity. Simulation results show that the CRL can adapt quickly to network capacity changes and reduce normalized root mean square error (NRMSE) between physical and virtual states by up to 55.2%, using the same RB number as traditional methods., Comment: 14 pages, 11 figures

Semantic communication is a new paradigm that aims at providing more efficient communication for the next-generation wireless network. It focuses on transmitting extracted, meaningful information instead of the raw data. However, deep learning-enabled image semantic communication models often require a significant amount of time and energy for training, which is unacceptable, especially for mobile devices. To solve this challenge, our paper first introduces a distributed image semantic communication system where the base station and local devices will collaboratively train the models for uplink communication. Furthermore, we formulate a joint optimization problem to balance time and energy consumption on the local devices during training while ensuring effective model performance. An adaptable resource allocation algorithm is proposed to meet requirements under different scenarios, and its time complexity, solution quality, and convergence are thoroughly analyzed. Experimental results demonstrate the superiority of our algorithm in resource allocation optimization against existing benchmarks and discuss its impact on the performance of image semantic communication systems., Comment: Accepted by IEEE Transactions on Wireless Communications

Materials manifesting the Kitaev model, characterized by bond-dependent interactions on a honeycomb lattice, can host exotic phenomena like quantum spin liquid states and topological magnetic excitations. However, finding such materials remains a formidable challenge. Here, we report high-resolution inelastic neutron scattering measurements performed on VI3, a van der Waals ferromagnetic Mott insulator, covering a wide range of reciprocal space. Our measurements unveil highly anisotropic magnetic excitations in momentum space. Through a comprehensive comparative analysis of various models that incorporate diverse symmetry-allowed magnetic interactions, we find the observed excitations are well captured by a model with a large bond-dependent Kitaev interaction. These results not only help to understand the intriguing properties of VI3, such as the pronounced anomalous thermal Hall effects and strong pressure/structure dependence of magnetism, but also open a new avenue for exploring Kitaev physics., Comment: 17 pages, 4 figures

Fusing small resource states into a larger, fully connected graph-state is essential for scalable photonic quantum computing. Theoretical analysis reveals that this can only be achieved when the success probability of the fusion gate surpasses a specific percolation threshold of 58.98% by using three-photon GHZ states as resource states. However, such an implementation of a fusion gate has never been experimentally realized before. Here, we successfully demonstrate a boosted fusion gate with a theoretical success probability of 75%, using deterministically generated auxiliary states. The success probability is experimentally measured to be 71.0(7)%. We further demonstrate the effectiveness of the boosted fusion gate by fusing two Bell states with a fidelity of 67(2)%. Our work paves a crucial path toward scalable linear optical quantum computing., Comment: 5 pages, 4 figures

Despite the significant progress made by existing retrieval augmented language models (RALMs) in providing trustworthy responses and grounding in reliable sources, they often overlook effective alignment with human preferences. In the alignment process, reward models (RMs) act as a crucial proxy for human values to guide optimization. However, it remains unclear how to evaluate and select a reliable RM for preference alignment in RALMs. To this end, we propose RAG-RewardBench, the first benchmark for evaluating RMs in RAG settings. First, we design four crucial and challenging RAG-specific scenarios to assess RMs, including multi-hop reasoning, fine-grained citation, appropriate abstain, and conflict robustness. Then, we incorporate 18 RAG subsets, six retrievers, and 24 RALMs to increase the diversity of data sources. Finally, we adopt an LLM-as-a-judge approach to improve preference annotation efficiency and effectiveness, exhibiting a strong correlation with human annotations. Based on the RAG-RewardBench, we conduct a comprehensive evaluation of 45 RMs and uncover their limitations in RAG scenarios. Additionally, we also reveal that existing trained RALMs show almost no improvement in preference alignment, highlighting the need for a shift towards preference-aligned training.We release our benchmark and code publicly at https://huggingface.co/datasets/jinzhuoran/RAG-RewardBench/ for future work., Comment: 26 pages, 12 figures, 6 tables

Elucidating spin correlations in the parent compounds of high-temperature superconductors is crucial for understanding superconductivity. We used neutron scattering to study spin correlations in Li$_{1-x}$Fe$_x$ODFeSe, an insulating material with reduced electron carriers compared to its superconducting counterpart ($T_c$ = 41 K), serving as the undoped parent compound. Our findings show a reduced total fluctuating moment in this insulator relative to FeSe and 122 iron pnictides, likely due to increased interlayer distances from intercalation, which enhance fluctuations and reduce the intensity of spin excitations. Moreover, we observed a V-shaped spin wave-like excitation dispersion, contrasting with the twisted hourglass pattern in the superconducting counterpart. Electron doping shifts spin excitation from ($\pi$, 0) point to an incommensurate position towards ($\pi$, $\pi$) direction below 65 meV. This transition from V-shaped to hourglass-like dispersion, akin to behaviors in hole-doped cuprates, suggests a potential shared mechanism in magnetism and superconductivity across these diverse systems., Comment: Accepted for publication in PRL

Transformer models have revolutionized AI, enabling applications like content generation and sentiment analysis. However, their use in Machine Learning as a Service (MLaaS) raises significant privacy concerns, as centralized servers process sensitive user data. Private Transformer Inference (PTI) addresses these issues using cryptographic techniques such as Secure Multi-Party Computation (MPC) and Homomorphic Encryption (HE), enabling secure model inference without exposing inputs or models. This paper reviews recent advancements in PTI, analyzing state-of-the-art solutions, their challenges, and potential improvements. We also propose evaluation guidelines to assess resource efficiency and privacy guarantees, aiming to bridge the gap between high-performance inference and data privacy., Comment: The manuscript is still being revised and will be continuously updated in the future

The subtlety of emotional expressions makes implicit emotion analysis (IEA) particularly sensitive to user-specific characteristics. Current studies personalize emotion analysis by focusing on the author but neglect the impact of the intended reader on implicit emotional feedback. In this paper, we introduce Personalized IEA (PIEA) and present the RAPPIE model, which addresses subjective variability by incorporating reader feedback. In particular, (1) we create reader agents based on large language models to simulate reader feedback, overcoming the issue of ``spiral of silence effect'' and data incompleteness of real reader reaction. (2) We develop a role-aware multi-view graph learning to model the emotion interactive propagation process in scenarios with sparse reader information. (3) We construct two new PIEA datasets covering English and Chinese social media with detailed user metadata, addressing the text-centric limitation of existing datasets. Extensive experiments show that RAPPIE significantly outperforms state-of-the-art baselines, demonstrating the value of incorporating reader feedback in PIEA.

In communication networks, optimization is essential in enhancing performance metrics, e.g., network utility. These optimization problems often involve sum-of-products (or ratios) terms, which are typically non-convex and NP-hard, posing challenges in their solution. Recent studies have introduced transformative techniques, mainly through quadratic and parametric convex transformations, to solve these problems efficiently. Based on them, this paper introduces novel decoupling techniques and bounds for handling multiplicative and fractional terms involving any number of coupled functions by utilizing the harmonic mean (HM), geometric mean (GM), arithmetic mean (AM), and quadratic mean (QM) inequalities. We derive closed-form expressions for these bounds. Focusing on the AM upper bound, we thoroughly examine its convexity and convergence properties. Under certain conditions, we propose a novel successive convex approximation (SCA) algorithm with the AM upper bound to achieve stationary point solutions in optimizations involving general multiplicative terms. Comprehensive proofs are provided to substantiate these claims. Furthermore, we illustrate the versatility of the AM upper bound by applying it to both optimization functions and constraints, as demonstrated in case studies involving the optimization of transmission energy and quantum source positioning. Numerical results are presented to show the effectiveness of our proposed SCA method., Comment: Submitted to one journal

Language Agent could be endowed with different mechanisms for autonomous task accomplishment. Current agents typically rely on fixed mechanisms or a set of mechanisms activated in a predefined order, limiting their adaptation to varied potential task solution structures. To this end, this paper proposes \textbf{A}daptive \textbf{L}anguage \textbf{A}gent \textbf{M}echanism \textbf{A}ctivation Learning with Self-Exploration (\textbf{ALAMA}), which focuses on optimizing mechanism activation adaptability without reliance on expert models. Initially, it builds a harmonized agent framework (\textbf{UniAct}) to \textbf{Uni}fy different mechanisms via \textbf{Act}ions. Then it leverages a training-efficient optimization method based on self-exploration to enable the UniAct to adaptively activate the appropriate mechanisms according to the potential characteristics of the task. Experimental results demonstrate significant improvements in downstream agent tasks, affirming the effectiveness of our approach in facilitating more dynamic and context-sensitive mechanism activation., Comment: COLING2025

Large language models (LLMs) have learned vast amounts of factual knowledge through self-supervised pre-training on large-scale corpora. Meanwhile, LLMs have also demonstrated excellent multilingual capabilities, which can express the learned knowledge in multiple languages. However, the knowledge storage mechanism in LLMs still remains mysterious. Some researchers attempt to demystify the factual knowledge in LLMs from the perspective of knowledge neurons, and subsequently discover language-agnostic knowledge neurons that store factual knowledge in a form that transcends language barriers. However, the preliminary finding suffers from two limitations: 1) High Uncertainty in Localization Results. Existing study only uses a prompt-based probe to localize knowledge neurons for each fact, while LLMs cannot provide consistent answers for semantically equivalent queries. Thus, it leads to inaccurate localization results with high uncertainty. 2) Lack of Analysis in More Languages. The study only analyzes language-agnostic knowledge neurons on English and Chinese data, without exploring more language families and languages. Naturally, it limits the generalizability of the findings. To address aforementioned problems, we first construct a new benchmark called Rephrased Multilingual LAMA (RML-LAMA), which contains high-quality cloze-style multilingual parallel queries for each fact. Then, we propose a novel method named Multilingual Integrated Gradients with Uncertainty Estimation (MATRICE), which quantifies the uncertainty across queries and languages during knowledge localization. Extensive experiments show that our method can accurately localize language-agnostic knowledge neurons. We also further investigate the role of language-agnostic knowledge neurons in cross-lingual knowledge editing, knowledge enhancement and new knowledge injection.

In the rapidly evolving landscape of the Metaverse, enhanced by blockchain technology, the efficient processing of data has emerged as a critical challenge, especially in wireless communication systems. Addressing this need, our paper introduces the innovative concept of data processing efficiency (DPE), aiming to maximize processed bits per unit of resource consumption in blockchain-empowered Metaverse environments. To achieve this, we propose the DPE-Aware User Association and Resource Allocation (DAUR) algorithm, a tailored solution for these complex systems. The DAUR algorithm transforms the challenging task of optimizing the sum of DPE ratios into a solvable convex optimization problem. It uniquely alternates the optimization of key variables like user association, work offloading ratios, task-specific computing resource distribution, bandwidth allocation, user power usage ratios, and server computing resource allocation ratios. Our extensive numerical results demonstrate the DAUR algorithm's effectiveness in DPE., Comment: This is the full version of the conference paper published in the Twenty-fifth International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc 2024). DOI: https://doi.org/10.1145/3641512.3686376. arXiv admin note: text overlap with arXiv:2406.13602

Symmetric or asymmetric state transfer along a path encircling an exceptional point (EP) is one of the extraordinary phenomena in non-Hermitian (NH) systems. However, the application of this property in both symmetric and asymmetric entangled state transfers, within systems experiencing multiple types of dissipation, remains to be fully explored. In this work, we demonstrate efficient symmetric and asymmetric Bell-state transfers, by modulating system parameters within a Jaynes-Cummings model and considering atomic spontaneous emission and cavity decay. The effective suppression of nonadiabatic transitions facilitates a symmetric exchange of Bell states regardless of the encircling direction. Additionally, we present a counterintuitive finding, suggests that the presence of an EP may not be indispensable for implementation of asymmetric state transfers in NH systems. We further achieve perfect asymmetric Bell-state transfers even in the absence of an EP, while dynamically orbiting around an approximate EP. Our work presents an approach to effectively and reliably manipulate entangled states with both symmetric and asymmetric characteristics, through the dissipation engineering in NH systems., Comment: 7 pages, 6 figures

Structured data, such as tables, graphs, and databases, play a critical role in plentiful NLP tasks such as question answering and dialogue system. Recently, inspired by Vision-Language Models, Graph Neutral Networks (GNNs) have been introduced as an additional modality into the input of Large Language Models (LLMs) to improve their performance on Structured Knowledge Grounding (SKG) tasks. However, those GNN-enhanced LLMs have the following limitations: (1) They employ diverse GNNs to model varying types of structured data, rendering them unable to uniformly process various forms of structured data. (2) The pretraining of GNNs is coupled with specific LLMs, which prevents GNNs from fully aligning with the textual space and limits their adaptability to other LLMs. To address these issues, we propose \textbf{L}arge \textbf{L}anguage and \textbf{S}tructured Data \textbf{A}ssistant (LLaSA), a general framework for enhancing LLMs' ability to handle structured data. Specifically, we represent various types of structured data in a unified hypergraph format, and use self-supervised learning to pretrain a hypergraph encoder, and a G-Former compressing encoded hypergraph representations with cross-attention. The compressed hypergraph representations are appended to the serialized inputs during training and inference stages of LLMs. Experimental results on multiple SKG tasks show that our pretrained hypergraph encoder can adapt to various LLMs and enhance their ability to process different types of structured data. Besides, LLaSA, with LoRA fine-tuning, outperforms previous SOTA method using full parameters tuning., Comment: NAACL 2025 Main

The rapid proliferation of online news has posed significant challenges in tracking the continuous development of news topics. Traditional timeline summarization constructs a chronological summary of the events but often lacks the flexibility to meet the diverse granularity needs. To overcome this limitation, we introduce a new paradigm, Dynamic-granularity TimELine Summarization, (DTELS), which aims to construct adaptive timelines based on user instructions or requirements. This paper establishes a comprehensive benchmark for DTLES that includes: (1) an evaluation framework grounded in journalistic standards to assess the timeline quality across four dimensions: Informativeness, Granular Consistency, Factuality, and Coherence; (2) a large-scale, multi-source dataset with multiple granularity timeline annotations based on a consensus process to facilitate authority; (3) extensive experiments and analysis with two proposed solutions based on Large Language Models (LLMs) and existing state-of-the-art TLS methods. The experimental results demonstrate the effectiveness of LLM-based solutions. However, even the most advanced LLMs struggle to consistently generate timelines that are both informative and granularly consistent, highlighting the challenges of the DTELS task., Comment: Under review

Audio-driven portrait animation has made significant advances with diffusion-based models, improving video quality and lipsync accuracy. However, the increasing complexity of these models has led to inefficiencies in training and inference, as well as constraints on video length and inter-frame continuity. In this paper, we propose JoyVASA, a diffusion-based method for generating facial dynamics and head motion in audio-driven facial animation. Specifically, in the first stage, we introduce a decoupled facial representation framework that separates dynamic facial expressions from static 3D facial representations. This decoupling allows the system to generate longer videos by combining any static 3D facial representation with dynamic motion sequences. Then, in the second stage, a diffusion transformer is trained to generate motion sequences directly from audio cues, independent of character identity. Finally, a generator trained in the first stage uses the 3D facial representation and the generated motion sequences as inputs to render high-quality animations. With the decoupled facial representation and the identity-independent motion generation process, JoyVASA extends beyond human portraits to animate animal faces seamlessly. The model is trained on a hybrid dataset of private Chinese and public English data, enabling multilingual support. Experimental results validate the effectiveness of our approach. Future work will focus on improving real-time performance and refining expression control, further expanding the applications in portrait animation. The code is available at: https://github.com/jdh-algo/JoyVASA.

Nontrivial spectral properties of non-Hermitian systems can give rise to intriguing effects that lack counterparts in Hermitian systems. For instance, when dynamically varying system parameters along a path enclosing an exceptional point (EP), chiral mode conversion occurs. A recent study [Phys. Rev. Lett. 133, 113802 (2024)] demonstrates the achievability of pure adiabatic state transfer by specifically selecting a trajectory in the system parameter space where the corresponding evolution operator exhibits a real spectrum while winding around an EP. However, the intended adiabatic state transfer becomes fragile when taking into account the effect of the nonadiabatic transition. In this work, we propose a scheme for achieving robust and rapid adiabatic state transfer in time-modulated two-level non-Hermitian systems by appropriately modulating system Hamiltonian and time-evolution trajectory. Numerical simulations confirm that complete adiabatic transfer can always be achieved even under nonadiabatic conditions after one period for different initialized adiabatic states, and the scheme remains insensitive to moderate fluctuations in control parameters. Therefore, this scheme offers alternative approaches for quantum-state engineering in non-Hermitian systems., Comment: 7 pages, 6 figures

Knowledge editing technology is crucial for maintaining the accuracy and timeliness of large language models (LLMs) . However, the setting of this task overlooks a significant portion of commonsense knowledge based on free-text in the real world, characterized by broad knowledge scope, long content and non instantiation. The editing objects of previous methods (e.g., MEMIT) were single token or entity, which were not suitable for commonsense knowledge in free-text form. To address the aforementioned challenges, we conducted experiments from two perspectives: knowledge localization and knowledge editing. Firstly, we introduced Knowledge Localization for Free-Text(KLFT) method, revealing the challenges associated with the distribution of commonsense knowledge in MLP and Attention layers, as well as in decentralized distribution. Next, we propose a Dynamics-aware Editing Method(DEM), which utilizes a Dynamics-aware Module to locate the parameter positions corresponding to commonsense knowledge, and uses Knowledge Editing Module to update knowledge. The DEM method fully explores the potential of the MLP and Attention layers, and successfully edits commonsense knowledge based on free-text. The experimental results indicate that the DEM can achieve excellent editing performance., Comment: 11 pages, 8 figures

Early detection through imaging and accurate diagnosis is crucial in mitigating the high mortality rate associated with breast cancer. However, locating tumors from low-resolution and high-noise medical images is extremely challenging. Therefore, this paper proposes a novel PGDiffSeg (Prior-Guided Diffusion Denoising Model with Parameter-Shared Attention) that applies diffusion denoising methods to breast cancer medical image segmentation, accurately recovering the affected areas from Gaussian noise. Firstly, we design a parallel pipeline for noise processing and semantic information processing and propose a parameter-shared attention module (PSA) in multi-layer that seamlessly integrates these two pipelines. This integration empowers PGDiffSeg to incorporate semantic details at multiple levels during the denoising process, producing highly accurate segmentation maps. Secondly, we introduce a guided strategy that leverages prior knowledge to simulate the decision-making process of medical professionals, thereby enhancing the model's ability to locate tumor positions precisely. Finally, we provide the first-ever discussion on the interpretability of the generative diffusion model in the context of breast cancer segmentation. Extensive experiments have demonstrated the superiority of our model over the current state-of-the-art approaches, confirming its effectiveness as a flexible diffusion denoising method suitable for medical image research. Our code will be publicly available later.

With the development of large language models, they are widely used as agents in various fields. A key component of agents is memory, which stores vital information but is susceptible to jailbreak attacks. Existing research mainly focuses on single-agent attacks and shared memory attacks. However, real-world scenarios often involve independent memory. In this paper, we propose the Troublemaker Makes Chaos in Honest Town (TMCHT) task, a large-scale, multi-agent, multi-topology text-based attack evaluation framework. TMCHT involves one attacker agent attempting to mislead an entire society of agents. We identify two major challenges in multi-agent attacks: (1) Non-complete graph structure, (2) Large-scale systems. We attribute these challenges to a phenomenon we term toxicity disappearing. To address these issues, we propose an Adversarial Replication Contagious Jailbreak (ARCJ) method, which optimizes the retrieval suffix to make poisoned samples more easily retrieved and optimizes the replication suffix to make poisoned samples have contagious ability. We demonstrate the superiority of our approach in TMCHT, with 23.51%, 18.95%, and 52.93% improvements in line topology, star topology, and 100-agent settings. Encourage community attention to the security of multi-agent systems.

Large language models (LLMs) sometimes demonstrate poor performance on knowledge-intensive tasks, commonsense reasoning is one of them. Researchers typically address these issues by retrieving related knowledge from knowledge graphs or employing self-enhancement methods to elicit knowledge in LLMs. However, noisy knowledge and invalid reasoning issues hamper their ability to answer questions accurately. To this end, we propose a novel method named eliciting, filtering and integrating knowledge in large language model (LINKED). In it, we design a reward model to filter out the noisy knowledge and take the marginal consistent reasoning module to reduce invalid reasoning. With our comprehensive experiments on two complex commonsense reasoning benchmarks, our method outperforms SOTA baselines (up to 9.0% improvement of accuracy). Besides, to measure the positive and negative impact of the injected knowledge, we propose a new metric called effectiveness-preservation score for the knowledge enhancement works. Finally, through extensive experiments, we conduct an in-depth analysis and find many meaningful conclusions about LLMs in commonsense reasoning tasks., Comment: Accepted by EMNLP 2024 Findings

Inductive reasoning is an essential capability for large language models (LLMs) to achieve higher intelligence, which requires the model to generalize rules from observed facts and then apply them to unseen examples. We present {\scshape Mirage}, a synthetic dataset that addresses the limitations of previous work, specifically the lack of comprehensive evaluation and flexible test data. In it, we evaluate LLMs' capabilities in both the inductive and deductive stages, allowing for flexible variation in input distribution, task scenario, and task difficulty to analyze the factors influencing LLMs' inductive reasoning. Based on these multi-faceted evaluations, we demonstrate that the LLM is a poor rule-based reasoner. In many cases, when conducting inductive reasoning, they do not rely on a correct rule to answer the unseen case. From the perspectives of different prompting methods, observation numbers, and task forms, models tend to consistently conduct correct deduction without correct inductive rules. Besides, we find that LLMs are good neighbor-based reasoners. In the inductive reasoning process, the model tends to focus on observed facts that are close to the current test example in feature space. By leveraging these similar examples, the model maintains strong inductive capabilities within a localized region, significantly improving its deductive performance., Comment: 25 pages,9 figures, under review

The discovery of superconductivity in La3Ni2O7-${\delta}$ under high pressure,with an onset critical temperature (Tc) around 80 K, has sparked significant interest in the superconducting phases of Ruddlesden-Popper nickelates, Lan+1NinO3n+1 (n = 2,3). While La4Ni3O10 exhibits nearly 100% superconductivity with Tc~30 K under high pressure, magnetic susceptibility studies on La3Ni2O7-${\delta}$, however, reveal a more complex picture, indicating either filamentary superconductivity or that approximately 50% of crystal phase becomes superconducting in polycrystalline samples. In this study, we employed scattering-type scanning near-field optical microscopy (SNOM) to visualize nanoscale structural phase separation in La3Ni2O7-${\delta}$, identifying enhanced optical conductivity with stripes approximately 183 nm wide. These stripes run diagonally with respect to the Ni-O-Ni bond directions in the a-b plane, ruling out the possibility that they arise from impurity phases, like the '1313', '214' or '4310' structures. Our findings suggest this phase separation corresponds to coexisting orthorhombic Amam and Fmmm structures,exhibiting optical conductivities ~ 22% and 29% of gold's, respectively. Additionally, we find that the Fmmm structure constitutes about 38% of the total field of view, while the remainder consists of Amam structure and the transitional region between Fmmm and Amam structures. In contrast, La4Ni3O10 exhibits uniform and higher optical conductivity with no observable evidence of phase separation. Thus, our study represents a pioneering effort to directly image nanoscale phase separation in Lan+1NinO3n+1 (n=2,3) nickelates. This observation could provide crucial insights into the factors that limit the superconducting volume fraction of La3Ni2O7-${\delta}$, highlighting SNOM as a powerful probe for exploring nanoscale low-energy physics in correlated quantum materials., Comment: 13 pages, 4 figures

We introduce DA-Code, a code generation benchmark specifically designed to assess LLMs on agent-based data science tasks. This benchmark features three core elements: First, the tasks within DA-Code are inherently challenging, setting them apart from traditional code generation tasks and demanding advanced coding skills in grounding and planning. Second, examples in DA-Code are all based on real and diverse data, covering a wide range of complex data wrangling and analytics tasks. Third, to solve the tasks, the models must utilize complex data science programming languages, to perform intricate data processing and derive the answers. We set up the benchmark in a controllable and executable environment that aligns with real-world data analysis scenarios and is scalable. The annotators meticulously design the evaluation suite to ensure the accuracy and robustness of the evaluation. We develop the DA-Agent baseline. Experiments show that although the baseline performs better than other existing frameworks, using the current best LLMs achieves only 30.5% accuracy, leaving ample room for improvement. We release our benchmark at https://da-code-bench.github.io., Comment: EMNLP 2024

Despite the success of Transformers, handling long contexts remains challenging due to the limited length generalization and quadratic complexity of self-attention. Thus Transformers often require post-training with a larger attention window, significantly increasing computational and memory costs. In this paper, we propose a novel attention mechanism based on dynamic context, Grouped Cross Attention (GCA), which can generalize to 1000 times the pre-training context length while maintaining the ability to access distant information with a constant attention window size. For a given input sequence, we split it into chunks and use each chunk to retrieve top-k relevant past chunks for subsequent text generation. Specifically, unlike most previous works that use an off-the-shelf retriever, our key innovation allows the retriever to learn how to retrieve past chunks that better minimize the auto-regressive loss of subsequent tokens in an end-to-end manner. Such a mechanism accommodates retrieved chunks with a fixed-size attention window to achieve long-range information access, significantly reducing computational and memory costs during training and inference. Experiments show that GCA-based models achieve near-perfect accuracy in passkey retrieval for 16M context lengths, which is 1000 times the training length., Comment: preprint

Thermodynamics of black holes offers a promising avenue for exploring the quantum nature of black holes and quantum gravity. In this Letter, we investigate the thermodynamic properties of dyonic black holes in the five-dimensional Einstein-Maxwell-Chern-Simons theory, obtained from IIB supergravity. We demonstrate that the standard form of the first law of thermodynamics is inconsistent with the quantum statistical relation widely adopted in black hole physics. By employing the on-shell variation of the Euclidean action and the Iyer-Wald formalism, we resolve this discrepancy and derive both the standard form of the first law and Smarr formula for the dyonic black holes. Furthermore, our findings are corroborated by numerical tests and are consistent with general hydrodynamic expectations., Comment: 5 pages, 2 figures, and supplementary information

AI foundation models have recently demonstrated impressive capabilities across a wide range of tasks. Fine-tuning (FT) is a method of customizing a pre-trained AI foundation model by further training it on a smaller, targeted dataset. In this paper, we initiate the study of the Privacy-Preserving Parameter-Efficient FT (P3EFT) framework, which can be viewed as the intersection of Parameter-Efficient FT (PEFT) and Privacy-Preserving FT (PPFT). PEFT modifies only a small subset of the model's parameters to achieve FT (i.e., adapting a pre-trained model to a specific dataset), while PPFT uses privacy-preserving technologies to protect the confidentiality of the model during the FT process. There have been many studies on PEFT or PPFT but very few on their fusion, which motivates our work on P3EFT to achieve both parameter efficiency and model privacy. To exemplify our P3EFT, we present the PrivTuner scheme, which incorporates Fully Homomorphic Encryption (FHE) enabled privacy protection into LoRA (short for ``Low-Rank Adapter''). Intuitively speaking, PrivTuner allows the model owner and the external data owners to collaboratively implement PEFT with encrypted data. After describing PrivTuner in detail, we further investigate its energy consumption and privacy protection. Then, we consider a PrivTuner system over wireless communications and formulate a joint optimization problem to adaptively minimize energy while maximizing privacy protection, with the optimization variables including FDMA bandwidth allocation, wireless transmission power, computational resource allocation, and privacy protection. A resource allocation algorithm is devised to solve the problem. Experiments demonstrate that our algorithm can significantly reduce energy consumption while adapting to different privacy requirements.

As mobile devices increasingly become focal points for advanced applications, edge computing presents a viable solution to their inherent computational limitations, particularly in deploying large language models (LLMs). However, despite the advancements in edge computing, significant challenges remain in efficient training and deploying LLMs due to the computational demands and data privacy concerns associated with these models. This paper explores a collaborative training framework that integrates mobile users with edge servers to optimize resource allocation, thereby enhancing both performance and efficiency. Our approach leverages parameter-efficient fine-tuning (PEFT) methods, allowing mobile users to adjust the initial layers of the LLM while edge servers handle the more demanding latter layers. Specifically, we formulate a multi-objective optimization problem to minimize the total energy consumption and delay during training. We also address the common issue of instability in model performance by incorporating stability enhancements into our objective function. Through novel fractional programming technique, we achieve a stationary point for the formulated problem. Simulations demonstrate that our method reduces the energy consumption as well as the latency, and increases the reliability of LLMs across various mobile settings., Comment: This paper appears in the 2024 International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc)

Generating images with accurately represented text, especially in non-Latin languages, poses a significant challenge for diffusion models. Existing approaches, such as the integration of hint condition diagrams via auxiliary networks (e.g., ControlNet), have made strides towards addressing this issue. However, diffusion models often fall short in tasks requiring controlled text generation, such as specifying particular fonts or producing text in small fonts. In this paper, we introduce a novel approach for multilingual visual text creation, named JoyType, designed to maintain the font style of text during the image generation process. Our methodology begins with assembling a training dataset, JoyType-1M, comprising 1 million pairs of data. Each pair includes an image, its description, and glyph instructions corresponding to the font style within the image. We then developed a text control network, Font ControlNet, tasked with extracting font style information to steer the image generation. To further enhance our model's ability to maintain font style, notably in generating small-font text, we incorporated a multi-layer OCR-aware loss into the diffusion process. This enhancement allows JoyType to direct text rendering using low-level descriptors. Our evaluations, based on both visual and accuracy metrics, demonstrate that JoyType significantly outperforms existing state-of-the-art methods. Additionally, JoyType can function as a plugin, facilitating the creation of varied image styles in conjunction with other stable diffusion models on HuggingFace and CivitAI. Our project is open-sourced on https://jdh-algo.github.io/JoyType/., Comment: Under Review at AAAI 2025

In audio-driven video generation, creating Mandarin videos presents significant challenges. Collecting comprehensive Mandarin datasets is difficult, and the complex lip movements in Mandarin further complicate model training compared to English. In this study, we collected 29 hours of Mandarin speech video from JD Health International Inc. employees, resulting in the jdh-Hallo dataset. This dataset includes a diverse range of ages and speaking styles, encompassing both conversational and specialized medical topics. To adapt the JoyHallo model for Mandarin, we employed the Chinese wav2vec2 model for audio feature embedding. A semi-decoupled structure is proposed to capture inter-feature relationships among lip, expression, and pose features. This integration not only improves information utilization efficiency but also accelerates inference speed by 14.3%. Notably, JoyHallo maintains its strong ability to generate English videos, demonstrating excellent cross-language generation capabilities. The code and models are available at https://jdh-algo.github.io/JoyHallo.

In this paper, we propose $\textbf{Ne}$ural-$\textbf{Sy}$mbolic $\textbf{C}$ollaborative $\textbf{D}$istillation ($\textbf{NesyCD}$), a novel knowledge distillation method for learning the complex reasoning abilities of Large Language Models (LLMs, e.g., \textgreater 13B). We argue that complex reasoning tasks are difficult for Small Language Models (SLMs, e.g., $\leq$ 7B), as these tasks demand not only general cognitive abilities but also specialized knowledge, which is often sparse and difficult for these neural-based SLMs to effectively capture. Therefore, NesyCD distills the general capabilities and specialized knowledge in LLMs using different manners. On the one hand, we distill only general abilities from teacher LLMs into the student SLMs of parameterized neural networks. On the other hand, for the specialized abilities and uncommon knowledge of a complex reasoning task, we employ a symbolic knowledge distillation approach to obtain and store the specialized knowledge within a symbolic knowledge base (KB). By decoupling general and specialized capabilities, the proposed NesyCD can achieve superior performance cost-effectively, utilizing smaller models and blending parameterized neural networks with symbolic KB. Moreover, the specialized KB generalizes well and is comprehended and manipulated by humans. Our experiments show that NesyCD significantly boosts SLMs' complex reasoning performance on in-domain (BBH, GSM8K) and out-of-domain (AGIEval, ARC) datasets. Notably, our approach enabled the LLaMA3-8B and Qwen2-7B to surpass GPT-3.5-turbo in performance and come close to matching LLaMA3-70B, despite the latter having nine times more parameters. Our code will be available at https://github.com/Xnhyacinth/NesyCD., Comment: Accepted to AAAI 2025

Tool learning enables the Large Language Models (LLMs) to interact with the external environment by invoking tools, enriching the accuracy and capability scope of LLMs. However, previous works predominantly focus on improving model's tool-utilizing accuracy and the ability to generalize to new, unseen tools, excessively forcing LLMs to adjust specific tool-invoking pattern without considering the harm to model's general performance. This deviates from the actual applications and original intention of integrating tools to enhance model. To tackle this problem, we dissect the capability trade-offs by examining the hidden representation changes and the gradient-based importance score of model's components. Based on the analysis result, we propose a Component Importance-based Tool-utilizing ability Injection method (CITI). According to the gradient-based importance score of different components, it alleviates the capability conflicts caused by fine-tuning process by applying distinct training strategies to different components. CITI applies Mixture-Of-LoRA (MOLoRA) for important components. Meanwhile, it fine-tunes the parameters of few components deemed less important in the backbone of the LLM, while keeping other parameters frozen. CITI can effectively enhance the model's tool-utilizing capability without excessively compromising its general performance. Experimental results demonstrate that our approach achieves outstanding performance across a range of evaluation metrics.

Small Language Models (SLMs) are attracting attention due to the high computational demands and privacy concerns of Large Language Models (LLMs). Some studies fine-tune SLMs using Chains of Thought (CoT) data distilled from LLMs, aiming to enhance their reasoning ability. Furthermore, Some CoT distillation methods introduce external symbolic knowledge into the generation process to improve the limited knowledge memory, reasoning ability and out-of-domain (OOD) generalization of SLMs. However, the introduction of symbolic knowledge increases computational overhead and introduces potential noise. In this paper, we introduce $\textit{SKIntern}$, an innovative approach that empowers SLMs to internalize symbolic knowledge and few-shot examples gradually through a progressive fine-tuning process, guided by a predefined linear decay schedule under curriculum learning. By efficiently internalizing knowledge, $\textit{SKIntern}$ reduces computational overhead and speeds up the reasoning process by focusing solely on the question during inference. It outperforms state-of-the-art baselines by over 5\%, while reducing inference costs (measured in FLOPs) by up to $4\times$ across a wide range of SLMs in both in-domain (ID) and out-of-domain (OOD) tasks. Our code will be available at \url{https://github.com/Xnhyacinth/SKIntern}., Comment: Accepted to COLING 2025

Video-based physiology, exemplified by remote photoplethysmography (rPPG), extracts physiological signals such as pulse and respiration by analyzing subtle changes in video recordings. This non-contact, real-time monitoring method holds great potential for home settings. Despite the valuable contributions of public benchmark datasets to this technology, there is currently no dataset specifically designed for passive home monitoring. Existing datasets are often limited to close-up, static, frontal recordings and typically include only 1-2 physiological signals. To advance video-based physiology in real home settings, we introduce the MHAD dataset. It comprises 1,440 videos from 40 subjects, capturing 6 typical activities from 3 angles in a real home environment. Additionally, 5 physiological signals were recorded, making it a comprehensive video-based physiology dataset. MHAD is compatible with the rPPG-toolbox and has been validated using several unsupervised and supervised methods. Our dataset is publicly available at https://github.com/jdh-algo/MHAD-Dataset.

Large language models encapsulate knowledge and have demonstrated superior performance on various natural language processing tasks. Recent studies have localized this knowledge to specific model parameters, such as the MLP weights in intermediate layers. This study investigates the differences between entity and relational knowledge through knowledge editing. Our findings reveal that entity and relational knowledge cannot be directly transferred or mapped to each other. This result is unexpected, as logically, modifying the entity or the relation within the same knowledge triplet should yield equivalent outcomes. To further elucidate the differences between entity and relational knowledge, we employ causal analysis to investigate how relational knowledge is stored in pre-trained models. Contrary to prior research suggesting that knowledge is stored in MLP weights, our experiments demonstrate that relational knowledge is also significantly encoded in attention modules. This insight highlights the multifaceted nature of knowledge storage in language models, underscoring the complexity of manipulating specific types of knowledge within these models., Comment: CIKM 2024

Pretrained language models like BERT and T5 serve as crucial backbone encoders for dense retrieval. However, these models often exhibit limited generalization capabilities and face challenges in improving in domain accuracy. Recent research has explored using large language models (LLMs) as retrievers, achieving SOTA performance across various tasks. Despite these advancements, the specific benefits of LLMs over traditional retrievers and the impact of different LLM configurations, such as parameter sizes, pretraining duration, and alignment processes on retrieval tasks remain unclear. In this work, we conduct a comprehensive empirical study on a wide range of retrieval tasks, including in domain accuracy, data efficiency, zero shot generalization, lengthy retrieval, instruction based retrieval, and multi task learning. We evaluate over 15 different backbone LLMs and non LLMs. Our findings reveal that larger models and extensive pretraining consistently enhance in domain accuracy and data efficiency. Additionally, larger models demonstrate significant potential in zero shot generalization, lengthy retrieval, instruction based retrieval, and multi task learning. These results underscore the advantages of LLMs as versatile and effective backbone encoders in dense retrieval, providing valuable insights for future research and development in this field., Comment: Submitted to EMNLP24

LLM have achieved success in many fields but still troubled by problematic content in the training corpora. LLM unlearning aims at reducing their influence and avoid undesirable behaviours. However, existing unlearning methods remain vulnerable to adversarial queries and the unlearned knowledge resurfaces after the manually designed attack queries. As part of a red-team effort to proactively assess the vulnerabilities of unlearned models, we design Dynamic Unlearning Attack (DUA), a dynamic and automated framework to attack these models and evaluate their robustness. It optimizes adversarial suffixes to reintroduce the unlearned knowledge in various scenarios. We find that unlearned knowledge can be recovered in $55.2\%$ of the questions, even without revealing the unlearned model's parameters. In response to this vulnerability, we propose Latent Adversarial Unlearning (LAU), a universal framework that effectively enhances the robustness of the unlearned process. It formulates the unlearning process as a min-max optimization problem and resolves it through two stages: an attack stage, where perturbation vectors are trained and added to the latent space of LLMs to recover the unlearned knowledge, and a defense stage, where previously trained perturbation vectors are used to enhance unlearned model's robustness. With our LAU framework, we obtain two robust unlearning methods, AdvGA and AdvNPO. We conduct extensive experiments across multiple unlearning benchmarks and various models, and demonstrate that they improve the unlearning effectiveness by over $53.5\%$, cause only less than a $11.6\%$ reduction in neighboring knowledge, and have almost no impact on the model's general capabilities., Comment: 13 pages

Knowledge editing aims to update outdated or incorrect knowledge in large language models (LLMs). However, current knowledge editing methods have limited scalability for lifelong editing. This study explores the fundamental reason why knowledge editing fails in lifelong editing. We begin with the closed-form solution derived from linear associative memory, which underpins state-of-the-art knowledge editing methods. We extend the solution from single editing to lifelong editing, and through rigorous mathematical derivation, identify an interference term in the final solution, suggesting that editing knowledge may impact irrelevant knowledge. Further analysis of the interference term reveals a close relationship with superposition between knowledge representations. When knowledge superposition does not exist in language models, the interference term vanishes, allowing for lossless knowledge editing. Experiments across numerous language models reveal that knowledge superposition is universal, exhibiting high kurtosis, zero mean, and heavy-tailed distributions with clear scaling laws. Ultimately, by combining theory and experiments, we demonstrate that knowledge superposition is the fundamental reason for the failure of lifelong editing. Moreover, this is the first study to investigate knowledge editing from the perspective of superposition and provides a comprehensive observation of superposition across numerous real-world language models. Code available at https://github.com/ChenhuiHu/knowledge_in_superposition., Comment: To be published in AAAI 2025

We report a comprehensive investigation of the lattice and magnetic structure in van der Waals antiferromagnet VBr3, characterized by a BiI3-type structure at room temperature. Neutron diffraction experiments were performed on both polycrystalline and single-crystalline VBr3 samples, revealing clear magnetic Bragg peaks emerging below the N\'eel temperature of TN = 26.5 K. These magnetic Bragg peaks can be indexed by k = (0, 0.5, 1) in hexagonal notation. Our refinement analysis suggests that the antiferromagnetic order in VBr3 manifests as a zigzag structure. Moreover, we observed peak splitting for nuclear Bragg peaks in the HK-plane below the structure transition temperature of Ts = 94 K, indicating the breaking of 3-fold symmetry within the ab-plane., Comment: 13 pages, 4 figures

Task-oriented dialogue (TOD) systems aim to efficiently handle task-oriented conversations, including information collection. How to utilize TOD accurately, efficiently and effectively for information collection has always been a critical and challenging task. Recent studies have demonstrated that Large Language Models (LLMs) excel in dialogue, instruction generation, and reasoning, and can significantly enhance the performance of TOD through fine-tuning. However, current datasets primarily cater to user-led systems and are limited to predefined specific scenarios and slots, thereby necessitating improvements in the proactiveness, diversity, and capabilities of TOD. In this study, we present a detailed multi-domain task-oriented data construction process for conversations, and a Chinese dialogue dataset generated based on this process, TransferTOD, which authentically simulates human-computer dialogues in 30 popular life service scenarios. Leveraging this dataset, we trained a model called TransferTOD-7B using full-parameter fine-tuning, showcasing notable abilities in slot filling and questioning. Our work has demonstrated its strong generalization capabilities in various downstream scenarios, significantly enhancing both data utilization efficiency and system performance. The data is released in https://github.com/KongLongGeFDU/TransferTOD.

Type-II multiferroicity, where electric polarization is induced by specific spin patterns, is crucial in fundamental physics and advanced spintronics. However, the spin model and magnetoelectric coupling mechanisms in prototypical type-II multiferroic CuFeO$_2$ and Al-doped CuFeO$_2$ remain unclear. Here, by considering both spin and alloy degrees of freedom, we develop a magnetic cluster expansion method, which considers all symmetry allowed interactions. Applying such method, we not only obtain realistic spin model that can correctly reproduce observations for both CuFeO$_2$ and CuFe$_{1-x}$Al$_x$O$_2$, but also revisit well-known theories of the original spin-current (SC) model and $p$-$d$ hybridization model. Specifically, we find that (i) a previously overlooked biquadratic interaction is critical to reproduce the $\uparrow\uparrow\downarrow\downarrow$ ground state and excited states of CuFeO$_2$; (ii) the combination of absent biquadratic interaction and increased magnetic frustration around Al dopants stabilizes the proper screw state; and (iii) it is the generalized spin-current (GSC) model that can correctly characterize the multiferroicity of CuFeO$_2$. These findings have broader implications for understanding novel magnetoelectric couplings in, e.g., monolayer multiferroic NiI$_2$.

Although Large Language Models (LLMs) have demonstrated strong ability, they are further supposed to be controlled and guided by in real-world scenarios to be safe, accurate, and intelligent. This demands the possession of capability of LLMs. However, no prior work has made a clear evaluation of the inferential rule-following capability of LLMs. Previous studies that try to evaluate the inferential rule-following capability of LLMs fail to distinguish the inferential rule-following scenarios from the instruction-following scenarios. Therefore, this paper first clarifies the concept of inferential rule-following and proposes a comprehensive benchmark, RuleBench, to evaluate a diversified range of inferential rule-following abilities. Our experimental results on a variety of LLMs show that they are still limited in following rules. Our analysis based on the evaluation results provides insights into the improvements for LLMs toward a better inferential rule-following intelligent agent. We further propose Inferential Rule-Following Tuning (IRFT). The experimental results show that through IRFT, LLMs can learn abstract rule-following abilities from purely synthetic data and then generalize to RuleBench. The data and code can be found at: https://anonymous.4open.science/r/llm-rule-following-B3E3/

With the rapid advancement of large models and mobile edge computing, transfer learning, particularly through fine-tuning, has become crucial for adapting models to downstream tasks. Traditionally, this requires users to share their data with model owners for fine-tuning, which is not only costly but also raises significant privacy concerns. Furthermore, fine-tuning large-scale models is computationally intensive and often impractical for many users. To tackle these challenges, we introduce a system that combines offsite-tuning with physical-layer security, which provides local data owners with a lightweight adapter and a compressed emulator. Data owners then fine-tune the adapter locally and securely send it back to the model owners through a confidential channel for integration, ensuring privacy and resource conservation. Our paper focuses on optimizing computational resource allocation among data owners and the large model owner deployed on edge, and on the compression ratio of adapters. We incorporate a secrecy uplink channel to maximize the utility that we defined while minimizing system costs like energy consumption and delay. The optimization uses the Dinkelbach algorithm, fractional programming, successive convex approximation and alternating optimization. Experiments demonstrate our algorithm's superiority over existing methods.

In non-Hermitian crystal systems under open boundary condition (OBC), it is generally believed that the OBC modes with frequencies containing positive imaginary parts, when excited by external driving, will experience exponential growth in population, thereby leading to instability. However, our work challenges this conventional understanding. In such a system, we find an anomalous response that grows exponentially with the frequency aligned with those of saddle points. The frequencies of these saddle points on the complex plane are below the maximum imaginary part of OBC spectrum, but they can lie within or beyond the OBC spectrum. We derive general formulas of excitation-response relationships and find that this anomalous response can occur because the excitation of OBC modes eventually evolve toward these saddle points at long times. Only when the frequencies of all these saddle points are below the real axis do the non-Hermitian crystal systems remain stable under periodic excitation. Thus our results also provide new insights on the stability criterion of non-Hermitian crystal systems., Comment: 5 pages, 4 figures