Search

We develop a 3D Eulerian model to study the transport and distribution of microplastics in the global ocean. Among other benefits that will be discussed in the paper, one unique feature of our model is that it takes into consideration the effect of properties of particles (size and density, the former for the first time) to their vertical terminal velocity. With ocean current velocity taken from ECCOv4r4, a dataset generated from a data-assimilated MITgcm reanalysis, our model is integrated for 26 years for particles of different properties with their stationary patterns studied. We find that only low-density particles with sufficient size (e.g. density $900kg/m^3$ with size $\gtrsim 10 \mu m$) aggregate in the five subtropical gyres observed in previous studies. In contrast, particles of smaller size ($\sim 1 \mu m$), irrespective of their density, behave like neutrally buoyant particles with a weaker pattern on the surface and a deeper penetration into depth (up to about 1km deep). In addition, we observe seasonal variations of floating particle concentration on the ocean surface, which reasonably agree with the satellite observation by Cyclone Global Navigation Satellite System (CYGNSS) in terms of the phase of the variation. We find that the seasonal variation of the surface particle concentration correlates well with the variation of the mixing layer (ML) depth globally, due to an almost uniform vertical distribution of particles in the ML with total amount of particles conserved., Comment: 22 pages, 13 figures. Results are presented at the APS-DFD 77th annual meeting

We introduce BackdoorMBTI, the first backdoor learning toolkit and benchmark designed for multimodal evaluation across three representative modalities from eleven commonly used datasets. BackdoorMBTI provides a systematic backdoor learning pipeline, encompassing data processing, data poisoning, backdoor training, and evaluation. The generated poison datasets and backdoor models enable detailed evaluation of backdoor defense methods. Given the diversity of modalities, BackdoorMBTI facilitates systematic evaluation across different data types. Furthermore, BackdoorMBTI offers a standardized approach to handling practical factors in backdoor learning, such as issues related to data quality and erroneous labels. We anticipate that BackdoorMBTI will expedite future research in backdoor defense methods within a multimodal context. Code is available at https://anonymous.4open.science/r/BackdoorMBTI-D6A1/README.md.

A unified definition for the rotation angle and rotation angular speed of general beams, including those with orbital angular momentum (OAM), has been lacking until now. The rotation of a general beam is characterized by observing the rotational behavior of the directions of the extreme spot sizes during propagation. We introduce the beam quality $M^2(\psi)$ factor to characterize the unique beam quality of a general beam across all directions, not limited to the $x$- or $y$-axes. Besides that, we present the beam center $s_{\psi}(\psi,z)$, spot size $w_{\psi}(\psi,z)$, waist position, waist radius, and divergence angle along the direction that forms an angle $\psi$ with the $x$-axis in the plane perpendicular to the $z$-axis for the general beam. Furthermore, this paper presents rapid calculation formulas for these parameters, utilizing the mode expansion method (MEM). Subsequently, we prove that only two extreme spot sizes exist in a given detection plane and the angle between the maximum and minimum spot angles is consistently $90^{\circ}$ during the propagation. We also prove the spot rotation angles converge as $z$ approaches either positive or negative infinity. We first show the extreme spot sizes, spot rotation angle, and angular speed for the vortex beam. Our formulas efficiently differentiate between vortex OAM beams and asymmetry OAM beams., Comment: 22 pages, 6 figures

Thermal broadening of the quasi-particle peak in the spectral function is an important physical feature in many statistical systems, but difficult to calculate. Within the projective truncation approximation (PTA) of Green's function equation of motion for classical systems, we produce the spectral function with thermal broadened quasi-particles peak using an $H$-expanded basis. We demonstrate this method on two model systems, the one-variable anharmonic oscillator model and the one-dimensional classical $\phi^4$ lattice model. Comparison with exact spectral function and the molecular dynamics simulation results shows that the method is semi-quantitatively accurate. Extension of the $H$-expanded basis to PTA for quantum system is possible., Comment: 20 pages, 14 figures

Cell-free massive multiple-input-multiple-output (CF-mMIMO) is regarded as one of the promising technologies for next-generation wireless networks. However, due to its distributed architecture, geographically separated access points (APs) jointly serve a large number of user-equipments (UEs), there will inevitably be a discrepancies in the arrival time of transmitted signals. In this paper, we investigate millimeter-wave (mmWave) CF-mMIMO orthogonal frequency division multiplexing (OFDM) systems with asynchronous reception in a wide area coverage scenario, where asynchronous timing offsets may extend far beyond the cyclic prefix (CP) range. A comprehensive asynchronous beam-domain signal transmission model is presented for mmWave CF-mMIMO-OFDM systems in both downlink and uplink, incorporating phase offset, inter-carrier interference (ICI) and inter-symbol interference (ISI). To address the issue of asynchronous reception, we propose a novel per-beam timing advance (PBTA) hybrid precoding architecture and analyze the spectral efficiency (SE) in the beam domain for downlink and uplink asynchronous receptions. Both scalable centralized and distributed implementations are taken into account, and the asynchronous delay phase is utilized to design precoding/combining vectors. Furthermore, we formulate the sum rate maximization problem and develop two low-complexity joint beam selection and UE association algorithms considering the impact of asynchronous timing offset exceeding the CP range. Simulation results demonstrate that the performance will be severely limited by ICI and ISI, and our proposed PBTA hybrid precoding architecture effectively mitigates asynchronous interference compared to the nearest AAU/UE-based timing-advance scheme. Additionally, numerical results show that our proposed low-complexity joint beam selection and UE association algorithms achieve superior SE performance.

Space-based gravitational wave observatories, such as LISA, Taiji, and TianQin, employ long-baseline laser interferometry, necessitating displacement measurement sensitivity at 1 pm/$\sqrt{Hz}$ level. A significant challenge in achieving this precision is the coupling noise arising from far-field wavefront errors (WFE) and laser pointing jitter. This paper presents a comprehensive noise model that incorporates three critical factors: transmitted WFE, static pointing angle, and laser beam jitter. Utilizing the Nijboer-Zernike diffraction theory, we derive an approximate expression for far-field WFE, ensuring minimal error and efficient computational performance. The approximate expression has convincing physical interpretability and reveals how various Zernike aberrations and their coupling impact far-field WFE. Furthermore, the study identifies that correcting optical axis deviations induced by $Z_3^{\pm1}$ through beam tilt exacerbates far-field WFE, underscoring the necessity for active suppression of $Z_3^{\pm1}$. The proposed model facilitates detailed system simulations of the laser link, evaluates Tilt-to-Length (TTL) noise, and offers theoretical insights for system optimization., Comment: 25 pages, 13 figures

While large language models (LLMs) showcase unprecedented capabilities, they also exhibit certain inherent limitations when facing seemingly trivial tasks. A prime example is the recently debated "reversal curse", which surfaces when models, having been trained on the fact "A is B", struggle to generalize this knowledge to infer that "B is A". In this paper, we examine the manifestation of the reversal curse across various tasks and delve into both the generalization abilities and the problem-solving mechanisms of LLMs. This investigation leads to a series of significant insights: (1) LLMs are able to generalize to "B is A" when both A and B are presented in the context as in the case of a multiple-choice question. (2) This generalization ability is highly correlated to the structure of the fact "A is B" in the training documents. For example, this generalization only applies to biographies structured in "[Name] is [Description]" but not to "[Description] is [Name]". (3) We propose and verify the hypothesis that LLMs possess an inherent bias in fact recalling during knowledge application, which explains and underscores the importance of the document structure to successful learning. (4) The negative impact of this bias on the downstream performance of LLMs can hardly be mitigated through training alone. These findings offer a novel perspective on interpreting LLMs' generalization through their intrinsic mechanisms and provide insights for developing more effective learning methods. Our code and data are available at https://github.com/alibaba/thinking_bias.git., Comment: Accepted at NeurIPS 2024. Our code and data are available at https://github.com/alibaba/thinking_bias.git

We propose a novel approach to scaling LLM inference for code generation. We frame code generation as a black box optimization problem within the code space, and employ optimization-inspired techniques to enhance exploration. Specifically, we introduce Scattered Forest Search to enhance solution diversity while searching for solutions. Our theoretical analysis illustrates how these methods avoid local optima during optimization. Extensive experiments on HumanEval, MBPP, APPS, CodeContests, and Leetcode reveal significant performance improvements. For instance, our method achieves a pass@1 rate of 67.1% on HumanEval+ and 87.2% on HumanEval with GPT-3.5, marking improvements of 8.6% and 4.3% over the state-of-the-art, while also halving the iterations needed to find the correct solution. Furthermore, our method scales more efficiently than existing search techniques, including tree search, line search, and repeated sampling.

Inference-time alignment enhances the performance of large language models without requiring additional training or fine-tuning but presents challenges due to balancing computational efficiency with high-quality output. Best-of-N (BoN) sampling, as a simple yet powerful approach, generates multiple responses and selects the best one, achieving improved performance but with a high computational cost. We propose TreeBoN, a novel framework that integrates a speculative tree-search strategy into Best-of-N (BoN) Sampling. TreeBoN maintains a set of parent nodes, iteratively branching and pruning low-quality responses, thereby reducing computational overhead while maintaining high output quality. Our approach also leverages token-level rewards from Direct Preference Optimization (DPO) to guide tree expansion and prune low-quality paths. We evaluate TreeBoN using AlpacaFarm, HH-RLHF, UltraFeedback, GSM8K, and TutorEval datasets, demonstrating consistent improvements. Specifically, TreeBoN achieves the highest win rate of 65% on TutorEval and around 60% win rates across other different datasets, outperforming standard BoN with the same computational cost and showcasing its scalability and alignment efficacy.

Reinforcement Learning from Human Feedback (RLHF) has become the predominant approach for language model (LM) alignment. At its core, RLHF uses a margin-based loss for preference optimization, specifying ideal LM behavior only by the difference between preferred and dispreferred responses. In this paper, we identify a common pitfall of margin-based methods -- the under-specification of ideal LM behavior on preferred and dispreferred responses individually, which leads to two unintended consequences as the margin increases: (1) The probability of dispreferred (e.g., unsafe) responses may increase, resulting in potential safety alignment failures. (2) The probability of preferred responses may decrease, even when those responses are ideal. We demystify the reasons behind these problematic behaviors: margin-based losses couple the change in the preferred probability to the gradient of the dispreferred one, and vice versa, often preventing the preferred probability from increasing while the dispreferred one decreases, and thus causing a synchronized increase or decrease in both probabilities. We term this effect, inherent in margin-based objectives, gradient entanglement. Formally, we derive conditions for general margin-based alignment objectives under which gradient entanglement becomes concerning: the inner product of the gradients of preferred and dispreferred log-probabilities is large relative to the individual gradient norms. We theoretically investigate why such inner products can be large when aligning language models and empirically validate our findings. Empirical implications of our framework extend to explaining important differences in the training dynamics of various preference optimization algorithms, and suggesting potential algorithm designs to mitigate the under-specification issue of margin-based methods and thereby improving language model alignment.

White dwarf, one of the compact objects in the universe, is not only pivotal for astrophysical inquiries but also furnishes a dynamic arena for nuclear physics exploration. In this work, we extend the relativistic mean field approach using a Walecka-type quantum hadrodynamics model to encapsulate the intricate structure of white dwarfs. This methodological advancement facilitates a cohesive analysis of white dwarfs, neutron stars, and the nuclear pasta within a unified theoretical framework. By meticulously calibrating the model parameters to nuclear matter properties, we successfully replicate the structures of various nuclei, such as \(^{4}\rm He\), isotopes of \(\rm C\), and \(^{16}\rm O\). Subsequently, we predict the properties of white dwarfs composed of atom-like units and the gravitational waves stemming from binary white dwarf inspirals incorporating tidal deformability contributions up to the 2.5 post-Newtonian order. These results shed light on the structure of white dwarfs and offer more clues for future gravitational wave detection.

Quantum error correction (QEC) is a critical component of FTQC; the QEC decoder is an important part of Classical Computing for Quantum or C4Q. Recent years have seen fast development in real-time QEC decoders. Existing efforts to build real-time decoders have yet to achieve a critical milestone: decoding dynamic logical circuits with error-corrected readout and feed forward. Achieving this requires significant engineering effort to adapt and reconfigure the decoders during runtime, depending on the branching of the logical circuit. We present a QEC decoder architecture called LEGO, with the ambitious goal of supporting dynamic logical operations. LEGO employs a novel abstraction called the decoding block to describe the decoding problem of a dynamic logical circuit. Moreover, decoding blocks can be combined with three other ideas to improve the efficiency, accuracy and latency of the decoder. First, they provide data and task parallelisms when combined with fusion-based decoding. Second, they can exploit the pipeline parallelism inside multi-stage decoders. Finally, they serve as basic units of work for computational resource management. Using decoding blocks, LEGO can be easily reconfigured to support all QEC settings and to easily accommodate innovations in three interdependent fields: code, logical operations and qubit hardware. In contrast, existing decoders are highly specialized to a specific QEC setting, which leads to redundant research and engineering efforts, slows down innovation, and further fragments the nascent quantum computing industry.

As a manner to augment pre-trained large language models (LLM), knowledge injection is critical to develop vertical domain large models and has been widely studied. Although most current approaches, including parameter-efficient fine-tuning (PEFT) and block expansion methods, uniformly apply knowledge across all LLM layers, it raises the question: are all layers equally crucial for knowledge injection? We begin by evaluating the importance of each layer in finding the optimal layer range for knowledge injection. Intuitively, the more important layers should play a more critical role in knowledge injection and deserve a denser injection. We observe performance dips in question-answering benchmarks after the removal or expansion of the shallow layers, and the degradation shrinks as the layer gets deeper, indicating that the shallow layers hold the key to knowledge injection. This insight leads us to propose the S strategy, a post-pretraining strategy of selectively enhancing shallow layers while pruning the less effective deep ones. Based on this strategy, we introduce Llama Slayer-8B and Llama Slayer-8B-Instruct. We experimented on the corpus of code $\&$ math and demonstrated the effectiveness of our strategy. Further experiments across different LLM, Mistral-7B, and a legal corpus confirmed the general applicability of the approach, underscoring its wide-ranging efficacy. Our code is available at: \https://github.com/txchen-USTC/Llama-Slayer

Modeling human preferences is crucial for aligning foundation models with human values. Traditional reward modeling methods, such as the Bradley-Terry (BT) reward model, fall short in expressiveness, particularly in addressing intransitive preferences. Although supervised pair preference models (PairPM) can express general preferences, their implementation is highly ad-hoc and cannot guarantee a consistent preference probability of compared pairs. Additionally, they impose high computational costs due to their quadratic query complexity when comparing multiple responses. In this paper, we introduce preference representation learning, an approach that embeds responses into a latent space to capture intricate preference structures efficiently, achieving linear query complexity. Additionally, we propose preference score-based General Preference Optimization (GPO), which generalizes reward-based reinforcement learning from human feedback. Experimental results show that our General Preference representation model (GPM) outperforms the BT reward model on the RewardBench benchmark with a margin of up to 5.6% and effectively models cyclic preferences where any BT reward model behaves like a random guess. Furthermore, evaluations on downstream tasks such as AlpacaEval2.0 and MT-Bench, following the language model post-training with GPO and our general preference model, reveal substantial performance improvements with margins up to 9.3%. These findings indicate that our method may enhance the alignment of foundation models with nuanced human values. The code is available at https://github.com/general-preference/general-preference-model., Comment: 34 pages

For AI agents to be helpful to humans, they should be able to follow natural language instructions to complete everyday cooperative tasks in human environments. However, real human instructions inherently possess ambiguity, because the human speakers assume sufficient prior knowledge about their hidden goals and intentions. Standard language grounding and planning methods fail to address such ambiguities because they do not model human internal goals as additional partially observable factors in the environment. We propose a new framework, Follow Instructions with Social and Embodied Reasoning (FISER), aiming for better natural language instruction following in collaborative embodied tasks. Our framework makes explicit inferences about human goals and intentions as intermediate reasoning steps. We implement a set of Transformer-based models and evaluate them over a challenging benchmark, HandMeThat. We empirically demonstrate that using social reasoning to explicitly infer human intentions before making action plans surpasses purely end-to-end approaches. We also compare our implementation with strong baselines, including Chain of Thought prompting on the largest available pre-trained language models, and find that FISER provides better performance on the embodied social reasoning tasks under investigation, reaching the state-of-the-art on HandMeThat.

Existing 3D mask learning methods encounter performance bottlenecks under limited data, and our objective is to overcome this limitation. In this paper, we introduce a triple point masking scheme, named TPM, which serves as a scalable framework for pre-training of masked autoencoders to achieve multi-mask learning for 3D point clouds. Specifically, we augment the baselines with two additional mask choices (i.e., medium mask and low mask) as our core insight is that the recovery process of an object can manifest in diverse ways. Previous high-masking schemes focus on capturing the global representation but lack the fine-grained recovery capability, so that the generated pre-trained weights tend to play a limited role in the fine-tuning process. With the support of the proposed TPM, available methods can exhibit more flexible and accurate completion capabilities, enabling the potential autoencoder in the pre-training stage to consider multiple representations of a single 3D object. In addition, an SVM-guided weight selection module is proposed to fill the encoder parameters for downstream networks with the optimal weight during the fine-tuning stage, maximizing linear accuracy and facilitating the acquisition of intricate representations for new objects. Extensive experiments show that the four baselines equipped with the proposed TPM achieve comprehensive performance improvements on various downstream tasks. Our code and models are available at https://github.com/liujia99/TPM.

The rapid advancement of machine learning technologies raises questions about the security of machine learning models, with respect to both training-time (poisoning) and test-time (evasion, impersonation, and inversion) attacks. Models performing image-related tasks, e.g. detection, and classification, are vulnerable to adversarial attacks that can degrade their performance and produce undesirable outcomes. This paper introduces a novel technique for anomaly detection in images called 2DSig-Detect, which uses a 2D-signature-embedded semi-supervised framework rooted in rough path theory. We demonstrate our method in adversarial settings for training-time and test-time attacks, and benchmark our framework against other state of the art methods. Using 2DSig-Detect for anomaly detection, we show both superior performance and a reduction in the computation time to detect the presence of adversarial perturbations in images.

In this paper, we study well-posedness of random periodic solutions of stochastic differential equations (SDEs) of McKean-Vlasov type driven by a two-sided Brownian motion, where the random periodic behaviour is characterised by the equations' long-time behaviour. Given the well-known connection between McKean-Vlasov SDEs and interacting particle systems, we show propagation of chaos and that the key properties of the interacting particle systems recover those of the McKean-Vlasov SDEs in the particle limit. All results in the present work are shown under two settings: fully and partially dissipative case. Each setting has its challenges and limitations. For instance, weakening full dissipativity to partial dissipativity demands stronger structural assumptions on the equations' dynamics and yields random periodic behaviour in the weak sense instead of pathwise sense (as in the full dissipativity case). The proof mechanisms are close but fundamentally different., Comment: 24 pages, no figures

Optical coherence tomography (OCT) has become critical for diagnosing retinal diseases as it enables 3D images of the retina and optic nerve. OCT acquisition is fast, non-invasive, affordable, and scalable. Due to its broad applicability, massive numbers of OCT images have been accumulated in routine exams, making it possible to train large-scale foundation models that can generalize to various diagnostic tasks using OCT images. Nevertheless, existing foundation models for OCT only consider 2D image slices, overlooking the rich 3D structure. Here, we present OCTCube, a 3D foundation model pre-trained on 26,605 3D OCT volumes encompassing 1.62 million 2D OCT images. OCTCube is developed based on 3D masked autoencoders and exploits FlashAttention to reduce the larger GPU memory usage caused by modeling 3D volumes. OCTCube outperforms 2D models when predicting 8 retinal diseases in both inductive and cross-dataset settings, indicating that utilizing the 3D structure in the model instead of 2D data results in significant improvement. OCTCube further shows superior performance on cross-device prediction and when predicting systemic diseases, such as diabetes and hypertension, further demonstrating its strong generalizability. Finally, we propose a contrastive-self-supervised-learning-based OCT-IR pre-training framework (COIP) for cross-modality analysis on OCT and infrared retinal (IR) images, where the OCT volumes are embedded using OCTCube. We demonstrate that COIP enables accurate alignment between OCT and IR en face images. Collectively, OCTCube, a 3D OCT foundation model, demonstrates significantly better performance against 2D models on 27 out of 29 tasks and comparable performance on the other two tasks, paving the way for AI-based retinal disease diagnosis.

We report the first experimental observation of dissipation-driven coherent quantum many-body oscillation, and this oscillation is manifested as the coherent exchange of atoms between the thermal and the condensate components in a three-dimensional partially condensed Bose gas. Firstly, we observe that the dissipation leads to two different atom loss rates between the thermal and the condensate components, such that the thermal fraction increases as dissipation time increases. Therefore, this dissipation process serves as a tool to uniformly ramp up the system's temperature without introducing extra density excitation. Subsequently, a coherent pair exchange of atoms between the thermal and the condensate components occurs, resulting in coherent oscillation of atom numbers in both components. This oscillation, permanently embedded in the atom loss process, is revealed clearly when we inset a duration of dissipation-free evolution into the entire dynamics, manifested as an oscillation of total atom number at the end. Finally, we also present a theoretical calculation to support this physical mechanism, which simultaneously includes dissipation, interaction, finite temperature, and harmonic trap effects. Our work introduces a highly controllable dissipation as a new tool to control quantum many-body dynamics., Comment: 11 pages, 5 figures, 1 table

This paper introduces a pioneering methodology, termed StructTuning, to efficiently transform foundation Large Language Models (LLMs) into domain specialists. It significantly reduces the training corpus requirement to a mere 0.3%, while achieving an impressive 50% of traditional knowledge injection performance. Our method is inspired by the educational processes of human students, particularly how structured domain knowledge from textbooks is assimilated and subsequently applied to tackle real-world challenges through specific exercises. Based on this, we propose a novel two-stage strategy for knowledge injection and alignment: Structure-aware Continual Pre-Training (SCPT) and Structure-aware Supervised Fine-Tuning (SSFT). In the SCPT phase, we automatically extract the domain knowledge taxonomy and reorganize the training corpora, enabling LLMs to effectively link textual segments to targeted knowledge points within the taxonomy. In the SSFT phase, we explicitly prompt models to elucidate the underlying knowledge structure in their outputs, leveraging the structured domain insight to address practical problems. Our ultimate method has undergone extensive evaluations across model architectures and scales, using closed-book question-answering tasks on LongBench and MMedBench datasets. Remarkably, our method demonstrates the potential of comparable improvement against the state-of-the-art MMedLM2 on MMedBench, while significantly reducing the training costs to 5%. This breakthrough paves the way for scaling up our StructTuning for stronger domain-specific LLMs with comprehensive data utilization. Code is available at https://github.com/alibaba/struxgpt., Comment: Preprint. Code is available at https://github.com/alibaba/struxgpt

When reading long-form text, human cognition is complex and structurized. While large language models (LLMs) process input contexts through a causal and sequential perspective, this approach can potentially limit their ability to handle intricate and complex inputs effectively. To enhance LLM's cognition capability, this paper presents a novel concept of context structurization. Specifically, we transform the plain, unordered contextual sentences into well-ordered and hierarchically structurized elements. By doing so, LLMs can better grasp intricate and extended contexts through precise attention and information-seeking along the organized structures. Extensive evaluations are conducted across various model architectures and sizes (including a series of auto-regressive LLMs as well as BERT-like masking models) on a diverse set of NLP tasks (e.g., context-based question-answering, exhaustive hallucination evaluation, and passage-level dense retrieval). Empirical results show consistent and significant performance gains afforded by a single-round structurization. In particular, we boost the open-sourced LLaMA2-70B model to achieve comparable performance against GPT-3.5-Turbo as the hallucination evaluator. Besides, we show the feasibility of distilling advanced LLMs' language processing abilities to a smaller yet effective StruXGPT-7B to execute structurization, addressing the practicality of our approach. Code is available at https://github.com/alibaba/struxgpt., Comment: This paper has been accepted by NeurIPS 2024. Code is available at https://github.com/alibaba/struxgpt

Text-to-video (T2V) generation models have advanced significantly, yet their ability to compose different objects, attributes, actions, and motions into a video remains unexplored. Previous text-to-video benchmarks also neglect this important ability for evaluation. In this work, we conduct the first systematic study on compositional text-to-video generation. We propose T2V-CompBench, the first benchmark tailored for compositional text-to-video generation. T2V-CompBench encompasses diverse aspects of compositionality, including consistent attribute binding, dynamic attribute binding, spatial relationships, motion binding, action binding, object interactions, and generative numeracy. We further carefully design evaluation metrics of MLLM-based metrics, detection-based metrics, and tracking-based metrics, which can better reflect the compositional text-to-video generation quality of seven proposed categories with 700 text prompts. The effectiveness of the proposed metrics is verified by correlation with human evaluations. We also benchmark various text-to-video generative models and conduct in-depth analysis across different models and different compositional categories. We find that compositional text-to-video generation is highly challenging for current models, and we hope that our attempt will shed light on future research in this direction., Comment: 13 pages (30 in total), project page: https://t2v-compbench.github.io/

A self-consistent gravitational quantum field theory, with gravitational force treated on the same footing as the other three fundamental interactions, was established recently. The gravidynamics predicted by such a theory could lead to important implications, and the comparisons with experimental results may provide us opportunities to test such new approach of gravity based on the framework of the quantum field theory of gauge interactions. In this work, we start with the effective field equation of the gravitational quantum field theory, and then solve the perturbative gravigauge field order by order up to the 1st post-Newtonian level under the assumption of a simplified energy-momentum tensor of perfect fluids. Having the constraints on the related post-Newtonian parameters from the most up-to-date observational data, the new bound on the combined coupling in the gravitational quantum field theory $|\gamma_G(\alpha_G-\alpha_W/2)| \leq (2.4\pm30)\times10^{-6}$ is obtained. Under such bound, we found that the new gravitational quantum field theory successfully passed and found no conflict with the contemporary keynote Solar system experiments of gravity., Comment: 7 pages

Financial market simulation (FMS) serves as a promising tool for understanding market anomalies and the underlying trading behaviors. To ensure high-fidelity simulations, it is crucial to calibrate the FMS model for generating data closely resembling the observed market data. Previous efforts primarily focused on calibrating the mid-price data, leading to essential information loss of the market activities and thus biasing the calibrated model. The Limit Order Book (LOB) data is the fundamental data fully capturing the market micro-structure and is adopted by worldwide exchanges. However, LOB is not applicable to existing calibration objective functions due to its tabular structure not suitable for the vectorized input requirement. This paper proposes to explicitly learn the vectorized representations of LOB with a Transformer-based autoencoder. Then the latent vector, which captures the major information of LOB, can be applied for calibration. Extensive experiments show that the learned latent representation not only preserves the non-linear auto-correlation in the temporal axis, but the precedence between successive price levels of LOB. Besides, it is verified that the performance of the representation learning stage is consistent with the downstream calibration tasks. Thus, this work also progresses the FMS on LOB data, for the first time.

The present work introduces and investigates an explicit time discretization scheme, called the projected Euler method, to numerically approximate random periodic solutions of semi-linear SDEs under non-globally Lipschitz conditions. The existence of the random periodic solution is demonstrated as the limit of the pull-back of the discretized SDE. Without relying on a priori high-order moment bounds of the numerical approximations, the mean square convergence rate is proved to be order 0.5 for SDEs with multiplicative noise and order 1 for SDEs with additive noise. Numerical examples are also provided to validate our theoretical findings., Comment: 25 pages,5 figures

The rapid advancement of Large Language Models (LLMs) has ushered in an era where AI-generated text is increasingly indistinguishable from human-generated content. Detecting AI-generated text has become imperative to combat misinformation, ensure content authenticity, and safeguard against malicious uses of AI. In this paper, we propose a novel hybrid approach that combines traditional TF-IDF techniques with advanced machine learning models, including Bayesian classifiers, Stochastic Gradient Descent (SGD), Categorical Gradient Boosting (CatBoost), and 12 instances of Deberta-v3-large models. Our approach aims to address the challenges associated with detecting AI-generated text by leveraging the strengths of both traditional feature extraction methods and state-of-the-art deep learning models. Through extensive experiments on a comprehensive dataset, we demonstrate the effectiveness of our proposed method in accurately distinguishing between human and AI-generated text. Our approach achieves superior performance compared to existing methods. This research contributes to the advancement of AI-generated text detection techniques and lays the foundation for developing robust solutions to mitigate the challenges posed by AI-generated content.

The intermediate-mass-ratio inspirals (IMRIs) may be surrounded by dark matter (DM) minispikes. The dynamical friction from these DM minispike structures can affect the dynamics and the gravitational wave (GW) emission of the IMRIs. We analyze the effects of general dynamical friction, with a particular contribution from DM particles moving faster than the stellar-mass black hole in an eccentric IMRI. The results show that the dynamical friction caused by these DM particles tends to eccentricify the orbit, and the general dynamical friction is able to increase the eccentricity. We also analyze the effects of general dynamical friction on the GW characteristic strain. The results indicate that the peak value of the characteristic strain occurs at higher frequencies as the power law index of DM minispike $\gamma_\mathrm{sp}$ increases. For the first time, a general analytical relation between the frequency peak value of characteristic strain of GWs and $\gamma_\mathrm{sp}$ is established. Using the analytical relation, the presence of DM and its halo density may be determined potentially from future GW data., Comment: 7 pages, 4 figures, submitted to PRX

Federated learning is highly valued due to its high-performance computing in distributed environments while safeguarding data privacy. To address resource heterogeneity, researchers have proposed a semi-asynchronous federated learning (SAFL) architecture. However, the performance gap between different aggregation targets in SAFL remain unexplored. In this paper, we systematically compare the performance between two algorithm modes, FedSGD and FedAvg that correspond to aggregating gradients and models, respectively. Our results across various task scenarios indicate these two modes exhibit a substantial performance gap. Specifically, FedSGD achieves higher accuracy and faster convergence but experiences more severe fluctuates in accuracy, whereas FedAvg excels in handling straggler issues but converges slower with reduced accuracy.

Diffusion models have achieved great success in image generation, with the backbone evolving from U-Net to Vision Transformers. However, the computational cost of Transformers is quadratic to the number of tokens, leading to significant challenges when dealing with high-resolution images. In this work, we propose Diffusion Mamba (DiM), which combines the efficiency of Mamba, a sequence model based on State Space Models (SSM), with the expressive power of diffusion models for efficient high-resolution image synthesis. To address the challenge that Mamba cannot generalize to 2D signals, we make several architecture designs including multi-directional scans, learnable padding tokens at the end of each row and column, and lightweight local feature enhancement. Our DiM architecture achieves inference-time efficiency for high-resolution images. In addition, to further improve training efficiency for high-resolution image generation with DiM, we investigate "weak-to-strong" training strategy that pretrains DiM on low-resolution images ($256\times 256$) and then finetune it on high-resolution images ($512 \times 512$). We further explore training-free upsampling strategies to enable the model to generate higher-resolution images (e.g., $1024\times 1024$ and $1536\times 1536$) without further fine-tuning. Experiments demonstrate the effectiveness and efficiency of our DiM. The code of our work is available here: {\url{https://github.com/tyshiwo1/DiM-DiffusionMamba/}}., Comment: The code of our work is available here: {\url{https://github.com/tyshiwo1/DiM-DiffusionMamba/}}

Standard reinforcement learning from human feedback (RLHF) approaches relying on parametric models like the Bradley-Terry model fall short in capturing the intransitivity and irrationality in human preferences. Recent advancements suggest that directly working with preference probabilities can yield a more accurate reflection of human preferences, enabling more flexible and accurate language model alignment. In this paper, we propose a self-play-based method for language model alignment, which treats the problem as a constant-sum two-player game aimed at identifying the Nash equilibrium policy. Our approach, dubbed Self-Play Preference Optimization (SPPO), utilizes iterative policy updates to provably approximate the Nash equilibrium. Additionally, we propose a new SPPO objective which is both strongly motivated by theory and is simple and effective in practice. In our experiments, using only 60k prompts (without responses) from the UltraFeedback dataset and without any prompt augmentation, by leveraging a pre-trained preference model PairRM with only 0.4B parameters, SPPO can obtain a model from fine-tuning Mistral-7B-Instruct-v0.2 that achieves the state-of-the-art length-controlled win-rate of 28.53% against GPT-4-Turbo on AlpacaEval 2.0. It also outperforms the (iterative) DPO and IPO on MT-Bench, Arena-Hard, and the Open LLM Leaderboard. Starting from a stronger base model Llama-3-8B-Instruct, we are able to achieve a length-controlled win rate of 38.77%. Notably, the strong performance of SPPO is achieved without additional external supervision (e.g., responses, preferences, etc.) from GPT-4 or other stronger language models. Codes are available at https://github.com/uclaml/SPPO., Comment: 27 pages, 4 figures, 5 tables

Floquet engineering is a powerful manipulation method in modern quantum technology. However, unwanted heating is the main challenge of Floquet engineering, therefore the Floquet thermalization has attracting considerable attentions recently. In this work, we investigate thermalization of periodically driven many-body systems through the lens of Krylov complexity, and find a topological origin of different thermalization behaviors. We demonstrate that If the topology of the Krylov chain is nontrivial, a periodically driven system will reach a state with finite temperature. When the Krylov chain is topologically trivial, the system will be heated to infinite temperature. We further show that the prethermalization can be understood as the tunnelling process of a quasi-edge mode through the local gap on Krylov chain. This picture provides a systematically method to obtain the effective prethermal Hamiltonian., Comment: 6 pages, 5 figures

For space-based gravitational wave (GW) detection, the main noise source for massive black hole binaries (MBHBs) is attributed to approximately $10^7$ double white dwarf binaries in the foreground. For a GW source, the amplitude of the detector response, recorded by a space-based gravitational wave detector, exhibits a modulation effect with a year period when observing the source from various orbital positions. Under the adverse conditions mentioned above, where there is a strong foreground noise and annual modulation in the signals, we employed the wavelet transform and the strong-amplitude relevant orbital position search methods, which allows the weak MBHB sources to achieve higher locating accuracy. In detail, for two MBHB sources of lower intensity, the precision of luminosity distance, represented by the ratio $\Delta D_L / D_L$ at the 95$\%$ confidence level, is enhanced by factors of $\sim$ 2. And the angular resolutions, denoted by $\Delta \Omega_s$, are enhanced by a factor of $\sim$ 20. These improvements increase the number of detectable GW sources, facilitates multi-messenger follow-up observations and provides constraints on the cosmological constant., Comment: 11 pages, 6 figures, publication in the SCIENCE CHINA Physics, Mechanics & Astronomy (SCPMA)

We propose an intuitive LLM prompting framework (AgentKit) for multifunctional agents. AgentKit offers a unified framework for explicitly constructing a complex "thought process" from simple natural language prompts. The basic building block in AgentKit is a node, containing a natural language prompt for a specific subtask. The user then puts together chains of nodes, like stacking LEGO pieces. The chains of nodes can be designed to explicitly enforce a naturally structured "thought process". For example, for the task of writing a paper, one may start with the thought process of 1) identify a core message, 2) identify prior research gaps, etc. The nodes in AgentKit can be designed and combined in different ways to implement multiple advanced capabilities including on-the-fly hierarchical planning, reflection, and learning from interactions. In addition, due to the modular nature and the intuitive design to simulate explicit human thought process, a basic agent could be implemented as simple as a list of prompts for the subtasks and therefore could be designed and tuned by someone without any programming experience. Quantitatively, we show that agents designed through AgentKit achieve SOTA performance on WebShop and Crafter. These advances underscore AgentKit's potential in making LLM agents effective and accessible for a wider range of applications. https://github.com/holmeswww/AgentKit

We propose a framework for automatic colorization that allows for iterative editing and modifications. The core of our framework lies in an imagination module: by understanding the content within a grayscale image, we utilize a pre-trained image generation model to generate multiple images that contain the same content. These images serve as references for coloring, mimicking the process of human experts. As the synthesized images can be imperfect or different from the original grayscale image, we propose a Reference Refinement Module to select the optimal reference composition. Unlike most previous end-to-end automatic colorization algorithms, our framework allows for iterative and localized modifications of the colorization results because we explicitly model the coloring samples. Extensive experiments demonstrate the superiority of our framework over existing automatic colorization algorithms in editability and flexibility. Project page: https://xy-cong.github.io/imagine-colorization., Comment: CVPR 2024. Project page: https://xy-cong.github.io/imagine-colorization

We revisit the unified two-timescale Q-learning algorithm as initially introduced by Angiuli et al. \cite{angiuli2022unified}. This algorithm demonstrates efficacy in solving mean field game (MFG) and mean field control (MFC) problems, simply by tuning the ratio of two learning rates for mean field distribution and the Q-functions respectively. In this paper, we provide a comprehensive theoretical explanation of the algorithm's bifurcated numerical outcomes under fixed learning rates. We achieve this by establishing a diagram that correlates continuous-time mean field problems to their discrete-time Q-function counterparts, forming the basis of the algorithm. Our key contribution lies in the construction of a Lyapunov function integrating both mean field distribution and Q-function iterates. This Lyapunov function facilitates a unified convergence of the algorithm across the entire spectrum of learning rates, thus providing a cohesive framework for analysis., Comment: 34 pages. Updated version for submission. We added more numerical results and fixed several minor mistakes

Ultralight dark matter (ULDM) is one of the leading well-motivated dark matter candidates, predicted in many theories beyond the standard model of particle physics and cosmology. There have been increasing interests in searching for ULDM in physical and astronomical experiments, mostly assuming there are additional interactions other than gravity between ULDM and normal matter. Here we demonstrate that even if ULDM has only gravitational interaction, it shall induce gravitational perturbations in solar system that may be large enough to cause detectable signals in future gravitational-wave (GW) laser interferometers in space. We investigate the sensitivities of Michelson time-delay interferometer to ULDM of various spins, and show vector ULDM with mass $m\lesssim 10^{-18}~$eV can be probed by space-based GW detectors aiming at $\mu$Hz frequencies. Our findings exhibit that GW detectors may directly probe ULDM in some mass ranges that otherwise are challenging to examine., Comment: 7 pages + supplementary material, 3 figures

Preference alignment is pivotal for empowering large language models (LLMs) to generate helpful and harmless responses. However, the performance of preference alignment is highly sensitive to the prevalent noise in the preference data. Recent efforts for this problem either marginally alleviate the impact of noise without the ability to actually reduce its presence, or rely on costly teacher LLMs prone to reward misgeneralization. To address these challenges, we propose the RObust Preference Optimization (ROPO) framework, an iterative alignment approach that integrates noise-tolerance and filtering of noisy samples without the aid of external models. Specifically, ROPO iteratively solves a constrained optimization problem, where we dynamically assign a quality-aware weight for each sample and constrain the sum of the weights to the number of samples we intend to retain. For noise-tolerant training and effective noise identification, we derive a robust loss by suppressing the gradients of samples with high uncertainty. We demonstrate both empirically and theoretically that the derived loss is critical for distinguishing noisy samples from clean ones. Furthermore, inspired by our derived loss, we propose a robustness-guided rejection sampling technique to compensate for the potential important information in discarded queries. Experiments on three widely-used datasets with Mistral-7B and Llama-2-7B demonstrate that ROPO significantly outperforms existing preference alignment methods, with its superiority growing as the noise rate increases.