arXiv subjects

Wei Ji

Publications and source records attributed to Wei Ji.

At least 145 records · Page 8Linked to original sources

arXiv2024

Described Spatial-Temporal Video Detection

Detecting visual content on language expression has become an emerging topic in the community. However, in the video domain, the existing setting, i.e., spatial-temporal video grounding (STVG), is formulated to only detect one pre-existing object in each frame, ignoring the fact that language descriptions can involve none or multiple entities within a video. In this work, we advance the STVG to a more practical setting called described spatial-temporal video detection (DSTVD) by overcoming the above limitation. To facilitate the exploration of DSTVD, we first introduce a new benchmark, namely DVD-ST. Notably, DVD-ST supports grounding from none to many objects onto the video in response to queries and encompasses a diverse range of over 150 entities, including appearance, actions, locations, and interactions. The extensive breadth and diversity of the DVD-ST dataset make it an exemplary testbed for the investigation of DSTVD. In addition to the new benchmark, we further present two baseline methods for our proposed DSTVD task by extending two representative STVG models, i.e., TubeDETR, and STCAT. These extended models capitalize on tubelet queries to localize and track referred objects across the video sequence. Besides, we adjust the training objectives of these models to optimize spatial and temporal localization accuracy and multi-class classification capabilities. Furthermore, we benchmark the baselines on the introduced DVD-ST dataset and conduct extensive experimental analysis to guide future investigation. Our code and benchmark will be publicly available.

cs.CV↗

arXiv2024

NExT-GPT: Any-to-Any Multimodal LLM

While recently Multimodal Large Language Models (MM-LLMs) have made exciting strides, they mostly fall prey to the limitation of only input-side multimodal understanding, without the ability to produce content in multiple modalities. As we humans always perceive the world and communicate with people through various modalities, developing any-to-any MM-LLMs capable of accepting and delivering content in any modality becomes essential to human-level AI. To fill the gap, we present an end-to-end general-purpose any-to-any MM-LLM system, NExT-GPT. We connect an LLM with multimodal adaptors and different diffusion decoders, enabling NExT-GPT to perceive inputs and generate outputs in arbitrary combinations of text, images, videos, and audio. By leveraging the existing well-trained highly-performing encoders and decoders, NExT-GPT is tuned with only a small amount of parameter (1%) of certain projection layers, which not only benefits low-cost training and also facilitates convenient expansion to more potential modalities. Moreover, we introduce a modality-switching instruction tuning (MosIT) and manually curate a high-quality dataset for MosIT, based on which NExT-GPT is empowered with complex cross-modal semantic understanding and content generation. Overall, our research showcases the promising possibility of building an AI agent capable of modeling universal modalities, paving the way for more human-like AI research in the community. Project page: https://next-gpt.github.io/

cs.AI↗

arXiv2024

Unusual charge density wave introduced by Janus structure in monolayer vanadium dichalcogenides

As a fundamental structural feature, the symmetry of materials determines the exotic quantum properties in transition metal dichalcogenides (TMDs) with charge density wave (CDW). Breaking the inversion symmetry, the Janus structure, an artificially constructed lattice, provides an opportunity to tune the CDW states and the related properties. However, limited by the difficulties in atomic-level fabrication and material stability, the experimental visualization of the CDW states in 2D TMDs with Janus structure is still rare. Here, using surface selenization of VTe2, we fabricated monolayer Janus VTeSe. With scanning tunneling microscopy, an unusual root13-root13 CDW state with threefold rotational symmetry breaking was observed and characterized. Combined with theoretical calculations, we find this CDW state can be attributed to the charge modulation in the Janus VTeSe, beyond the conventional electron-phonon coupling. Our findings provide a promising platform for studying the CDW states and artificially tuning the electronic properties toward the applications.

cond-mat.mtrl-sci↗

arXiv2024

Correlated electrons of the flat band in charge density wave state of 4Hb-TaSexS2-x

Many intriguing quantum states of matter, such as unconventional superconductivity, magnetic phases and fractional quantum Hall physics, emergent from the spatially-correlated localized electrons in the flat band of solid materials. By using scanning tunneling microscopy and spectroscopy (STM/STS), we report the real-space investigation of correlated electrons in the flat band of superlattice 4Hb-TaSexS2-x. In contrast with the pristine 4Hb-TaS2, the selenium (Se) substitutions significantly affect the interfacial transfer of correlated electrons between the CDW states of 1T- and 1H-TaS2 layers, and contribute a real-space fractional electron-filling configurations with the distributed electron-filled and -void SoD clusters of 1T-layer. The site-specific STS spectra directly reveal their respective prominent spectra weight above EF and symmetric Mott-like spectra. In addition, the spatial distributions of these electron-filled SoDs in the 1T-layer of 4Hb-TaSe0.7S1.3 demonstrate different local short-range patterning, clearly indicating the complex neighboring interactions among the localized electrons in the flat band of 1T-layer. Our results not only provide an in-depth insight of correlated electrons in the flat CDW band, and provide a simple platform to manipulate the electron-correlation-related quantum states.

cond-mat.mtrl-sci↗

arXiv2024

Controlled fabrication of freestanding monolayer SiC by electron irradiation

The design and preparation of novel quantum materials with atomic precision are crucial for exploring new physics and for device applications. Electron irradiation has demonstrated as an effective method for preparing novel quantum materials and quantum structures that could be challenging to obtain otherwise. It features the advantages of precise control over the patterning of such new materials and their integration with other materials with different functionalities. Here, we present a new strategy for fabricating freestanding monolayer SiC within nanopores of a graphene membrane. By regulating the energy of the incident electron beam and the in-situ heating temperature in a scanning transmission electron microscope (STEM), we can effectively control the patterning of nanopores and subsequent growth of monolayer SiC within the graphene lattice. The resultant SiC monolayers seamlessly connect with the graphene lattice, forming a planar structure distinct by a wide direct bandgap. Our in-situ STEM observations further uncover that the growth of monolayer SiC within the graphene nanopore is driven by a combination of bond rotation and atom extrusion, providing new insights into the atom-by-atom self-assembly of freestanding two-dimensional (2D) monolayers.

cond-mat.mtrl-sci↗

arXiv2024

Competing multiferroic phases in monolayer and few-layer NiI$_{2}$

A recent experiment reported type-II multiferroicity in monolayer (ML) NiI$_{2}$ based on a presumed spiral magnetic configuration (Spiral-B), which is, as we found here, under debate in the ML limit. Freestanding ML NiI$_{2}$ breaks its C$_{3}$ symmetry, as it prefers a striped antiferromagnetic order (AABB-AFM) along with an intralayer antiferroelectric (AFE) order. However, substrate confinement may preserve the C$_{3}$ symmetry and/or apply tensile strain to the ML. This leads to another spiral magnetic order (Spiral-$IV^X$), while 2L shows a different order (Spiral-$V^Y$) and Spiral-B dominates in thicker layers. Thus, three multiferroic phases, namely, Spiral-B+FE, Spiral-$IV^X$ +FE, Spiral-$V^Y$+FE, and an anti-multiferroic AABB-AFM+AFE one, show layer-thickness-dependent and geometry-dependent dominance, ascribed to competitions among thickness-dependent Kitaev, biquadratic, and Heisenberg spin-exchange interactions and single-ion magnetic anisotropy. Our theoretical results clarify the debate on the multiferroicity of ML NiI$_{2}$ and shed light on the role of layer-stacking-induced changes in noncollinear spin-exchange interactions and magnetic anisotropy in thickness-dependent magnetism.

cond-mat.mtrl-sci↗

arXiv2024

Spider: A Unified Framework for Context-dependent Concept Segmentation

Different from the context-independent (CI) concepts such as human, car, and airplane, context-dependent (CD) concepts require higher visual understanding ability, such as camouflaged object and medical lesion. Despite the rapid advance of many CD understanding tasks in respective branches, the isolated evolution leads to their limited cross-domain generalisation and repetitive technique innovation. Since there is a strong coupling relationship between foreground and background context in CD tasks, existing methods require to train separate models in their focused domains. This restricts their real-world CD concept understanding towards artificial general intelligence (AGI). We propose a unified model with a single set of parameters, Spider, which only needs to be trained once. With the help of the proposed concept filter driven by the image-mask group prompt, Spider is able to understand and distinguish diverse strong context-dependent concepts to accurately capture the Prompter's intention. Without bells and whistles, Spider significantly outperforms the state-of-the-art specialized models in 8 different context-dependent segmentation tasks, including 4 natural scenes (salient, camouflaged, and transparent objects and shadow) and 4 medical lesions (COVID-19, polyp, breast, and skin lesion with color colonoscopy, CT, ultrasound, and dermoscopy modalities). Besides, Spider shows obvious advantages in continuous learning. It can easily complete the training of new tasks by fine-tuning parameters less than 1\% and bring a tolerable performance degradation of less than 5\% for all old tasks. The source code will be publicly available at \href{https://github.com/Xiaoqi-Zhao-DLUT/Spider-UniCDSeg}{Spider-UniCDSeg}.

cs.CV↗

arXiv2024

Fine-tuning Multimodal LLMs to Follow Zero-shot Demonstrative Instructions

Recent advancements in Multimodal Large Language Models (MLLMs) have been utilizing Visual Prompt Generators (VPGs) to convert visual features into tokens that LLMs can recognize. This is achieved by training the VPGs on millions of image-caption pairs, where the VPG-generated tokens of images are fed into a frozen LLM to generate the corresponding captions. However, this image-captioning based training objective inherently biases the VPG to concentrate solely on the primary visual contents sufficient for caption generation, often neglecting other visual details. This shortcoming results in MLLMs' underperformance in comprehending demonstrative instructions consisting of multiple, interleaved, and multimodal instructions that demonstrate the required context to complete a task. To address this issue, we introduce a generic and lightweight Visual Prompt Generator Complete module (VPG-C), which can infer and complete the missing details essential for comprehending demonstrative instructions. Further, we propose a synthetic discriminative training strategy to fine-tune VPG-C, eliminating the need for supervised demonstrative instructions. As for evaluation, we build DEMON, a comprehensive benchmark for demonstrative instruction understanding. Synthetically trained with the proposed strategy, VPG-C achieves significantly stronger zero-shot performance across all tasks of DEMON. Further evaluation on the MME and OwlEval benchmarks also demonstrate the superiority of VPG-C. Our benchmark, code, and pre-trained models are available at https://github.com/DCDmllm/Cheetah.

cs.CV↗

arXiv2024

Alkaline earth metal mediated inter-molecular magnetism in perfluorocubane dimers and chains

Perfluorocubane ($C_8F_8$) was successfully synthesized and found to accept and store electrons in its internal cubic cavity to form magnetic moments. However their inter-molecule spin-exchange coupling mechanism is yet to be revealed. In this study, we found the inter-molecule magnetic groundstates of $C_8F_8$ dimer and one-dimensional (1D) chain are tunable from antiferromagnetic (AFM) to ferromagnetic (FM) by stacking orders and alkaline earth metals intercalation using first-principle calculations. The inter-molecule couplings are dominated by noncovalent halogen $C-F...C_4$ interactions. Stacking orders of dimers can regulate the relative position of the lone pairs and $σ-holes$ at the molecular interface and thus the magnetic groundstates. Alkaline earth metals M (M = Na, Mg) intercalations could form $C_4-M-C_4$ bonds and lead to FM direct exchange at the inter-molecule region. An unpaired electron donated by the intercalated atoms or electron doping can result in a local magnetic moment in dimers, exhibiting an on-off switching by the odd-even number of electron filling. Novel electronic properties such as spin gapless semiconductor and charge density wave (CDW) states emerge when $C_8F_8$ molecules self-assemble with intercalated atoms to form 1D chains. These findings manifest the roles of stacking and intercalation in modifying intermolecular magnetism and the revealed halogen bond-dominated exchange mechanisms are paramount additions to those previously established non-covalent couplings.

cond-mat.mtrl-sci↗

arXiv2024

Magnetic Properties of NH$_4$H$_2$PO$_4$ and KH$_2$PO$_4$: Emergence of Multiferroic Salts

We observe sharp step-down discontinuities in the magnetic susceptibility of NH$_4$H$_2$PO$_4$ and NH$_4$H$_2$PO$_4$-$d$$_{60}$ (60% deuterated) along the $a$ and $c$-axes occurring exactly at their antiferroelectric transition temperatures. For the case of KH$_2$PO$_4$, less pronounced discontinuities occur at the ferroelectric transition temperature. To explain this, we treat the acid protons as individual oscillators that generate current elements which translate to magnetic forces in near resonance with each other. With decreasing temperature, the resonant forces become more commensurate which amplifies a disproportionate drop off of two types of magnetic forces to eventually trigger the structural phase transitions. For the case of NH$_4$H$_2$PO$_4$, the associated internal magnetic field appears to aid the NH$_4$$^+$ to order at higher temperature. At 49 K, a shoulder-like anomaly in both NH$_4$H$_2$PO$_4$ and KH$_2$PO$_4$ is attributed to a possible onset of macroscopic quantum tunneling of protons. Our findings bring forth a new category of intrinsic multiferroic systems.

cond-mat.mtrl-sci↗

arXiv2024

Video-of-Thought: Step-by-Step Video Reasoning from Perception to Cognition

Existing research of video understanding still struggles to achieve in-depth comprehension and reasoning in complex videos, primarily due to the under-exploration of two key bottlenecks: fine-grained spatial-temporal perceptive understanding and cognitive-level video scene comprehension. This paper bridges the gap by presenting a novel solution. We first introduce a novel video Multimodal Large Language Model (MLLM), MotionEpic, which achieves fine-grained pixel-level spatial-temporal video grounding by integrating video spatial-temporal scene graph (STSG) representation. Building upon MotionEpic, we then develop a Video-of-Thought (VoT) reasoning framework. VoT inherits the Chain-of-Thought (CoT) core, breaking down a complex task into simpler and manageable sub-problems, and addressing them step-by-step from a low-level pixel perception to high-level cognitive interpretation. Extensive experiments across various complex video QA benchmarks demonstrate that our overall framework strikingly boosts existing state-of-the-art. To our knowledge, this is the first attempt at successfully implementing the CoT technique for achieving human-level video reasoning, where we show great potential in extending it to a wider range of video understanding scenarios. Project is open at https://haofei.vip/VoT

cs.AI↗

arXiv2024

I3: Intent-Introspective Retrieval Conditioned on Instructions

Recent studies indicate that dense retrieval models struggle to perform well on a wide variety of retrieval tasks that lack dedicated training data, as different retrieval tasks often entail distinct search intents. To address this challenge, in this work we leverage instructions to flexibly describe retrieval intents and introduce I3, a unified retrieval system that performs Intent-Introspective retrieval across various tasks, conditioned on Instructions without any task-specific training. I3 innovatively incorporates a pluggable introspector in a parameter-isolated manner to comprehend specific retrieval intents by jointly reasoning over the input query and instruction, and seamlessly integrates the introspected intent into the original retrieval model for intent-aware retrieval. Furthermore, we propose progressively-pruned intent learning. It utilizes extensive LLM-generated data to train I3 phase-by-phase, embodying two key designs: progressive structure pruning and drawback extrapolation-based data refinement. Extensive experiments show that in the BEIR benchmark, I3 significantly outperforms baseline methods designed with task-specific retrievers, achieving state-of-the-art zero-shot performance without any task-specific tuning.

cs.CL↗

arXiv2024

Site-ordering/disordering-induced magnetic textures in a vdW ferromagnet by competing global and broken inversion-symmetry

Fe5GeTe2 single crystals can be divided into nonquenched (NQ) and quench-cooled (QC) phases with different magnetic properties. A comprehensive understanding of the magnetic property variations in the NQ and QC phases is imperative for guiding Fe5GeTe2 towards spintronics applications; however, it remains elusive. Here, we report a real-space study on the structural and magnetic properties of these two magnetic phases using cryogenic magnetic force microscopy and scanning tunneling microscopy. The thermal history introduces disorder and order to the Fe(1) sites, resulting in the NQ and QC phases exhibiting global and broken inversion symmetry, respectively. The observed magnetic domain transitions (branching to labyrinthine) in the spin reorientation process and the distinct 3D spin textures stabilized by magnetic dipolar interaction observed in field-dependent studies allow the NQ phase to exhibit a more resilient global magnetic state. In contrast, the QC phase exhibits enhanced magnetic anisotropy, resulting in a higher TC. Meanwhile, the Dzyaloshinskii-Moriya interaction (DMI) introduced by the broken inversion symmetry causes the QC phase to exhibit a localized magnetic state: no domain transformation occurs during spin reorientation, and irregular domain states are observed in field-related studies. Our work provides an important reference for understanding the complex magnetic properties in Fe5GeTe2.

cond-mat.mtrl-sci↗

arXiv2024

Towards Robust Multi-Modal Reasoning via Model Selection

The reasoning capabilities of LLM (Large Language Model) are widely acknowledged in recent research, inspiring studies on tool learning and autonomous agents. LLM serves as the "brain" of the agent, orchestrating multiple tools for collaborative multi-step task solving. Unlike methods invoking tools like calculators or weather APIs for straightforward tasks, multi-modal agents excel by integrating diverse AI models for complex challenges. However, current multi-modal agents neglect the significance of model selection: they primarily focus on the planning and execution phases, and will only invoke predefined task-specific models for each subtask, making the execution fragile. Meanwhile, other traditional model selection methods are either incompatible with or suboptimal for the multi-modal agent scenarios, due to ignorance of dependencies among subtasks arising by multi-step reasoning. To this end, we identify the key challenges therein and propose the $\textit{M}^3$ framework as a plug-in with negligible runtime overhead at test-time. This framework improves model selection and bolsters the robustness of multi-modal agents in multi-step reasoning. In the absence of suitable benchmarks, we create MS-GQA, a new dataset specifically designed to investigate the model selection challenge in multi-modal agents. Our experiments reveal that our framework enables dynamic model selection, considering both user inputs and subtask dependencies, thereby robustifying the overall reasoning process. Our code and benchmark: https://github.com/LINs-lab/M3.

cs.LG↗

arXiv2024

Dysen-VDM: Empowering Dynamics-aware Text-to-Video Diffusion with LLMs

Text-to-video (T2V) synthesis has gained increasing attention in the community, in which the recently emerged diffusion models (DMs) have promisingly shown stronger performance than the past approaches. While existing state-of-the-art DMs are competent to achieve high-resolution video generation, they may largely suffer from key limitations (e.g., action occurrence disorders, crude video motions) with respect to the intricate temporal dynamics modeling, one of the crux of video synthesis. In this work, we investigate strengthening the awareness of video dynamics for DMs, for high-quality T2V generation. Inspired by human intuition, we design an innovative dynamic scene manager (dubbed as Dysen) module, which includes (step-1) extracting from input text the key actions with proper time-order arrangement, (step-2) transforming the action schedules into the dynamic scene graph (DSG) representations, and (step-3) enriching the scenes in the DSG with sufficient and reasonable details. Taking advantage of the existing powerful LLMs (e.g., ChatGPT) via in-context learning, Dysen realizes (nearly) human-level temporal dynamics understanding. Finally, the resulting video DSG with rich action scene details is encoded as fine-grained spatio-temporal features, integrated into the backbone T2V DM for video generating. Experiments on popular T2V datasets suggest that our Dysen-VDM consistently outperforms prior arts with significant margins, especially in scenarios with complex actions. Codes at https://haofei.vip/Dysen-VDM

cs.AI↗

arXiv2024

Observation of unconventional van der Waals multiferroics near room temperature

The search for two-dimensional (2D) van der Waals (vdW) multiferroics is an exciting yet challenging endeavor. Room-temperature 2D vdW few-layer multiferroic is a much bigger insurmountable obstacle. Here we report the discovery of an unconventional 2D vdW multiferroic with out-of-plane ferroelectric polarization and long-range magnetic orders in trilayer NiI2 device from 10 K to 295 K. The evolutions of magnetic domains with magnetic field, and the evolutions between ferroelectric and antiferroelectric phase have been unambiguously observed. More significantly, we realize a robust mutual control of magnetism and ferroelectricity at room temperature. The magnetic domains are manipulated by a small voltage ranging from 1 V to 6 V at 0 T and 295 K. This work opens opportunities for exploring multiferroic physics at the limit of few atomic layers.

cond-mat.mtrl-sci↗

arXiv2024

Composed Image Retrieval with Text Feedback via Multi-grained Uncertainty Regularization

We investigate composed image retrieval with text feedback. Users gradually look for the target of interest by moving from coarse to fine-grained feedback. However, existing methods merely focus on the latter, i.e., fine-grained search, by harnessing positive and negative pairs during training. This pair-based paradigm only considers the one-to-one distance between a pair of specific points, which is not aligned with the one-to-many coarse-grained retrieval process and compromises the recall rate. In an attempt to fill this gap, we introduce a unified learning approach to simultaneously modeling the coarse- and fine-grained retrieval by considering the multi-grained uncertainty. The key idea underpinning the proposed method is to integrate fine- and coarse-grained retrieval as matching data points with small and large fluctuations, respectively. Specifically, our method contains two modules: uncertainty modeling and uncertainty regularization. (1) The uncertainty modeling simulates the multi-grained queries by introducing identically distributed fluctuations in the feature space. (2) Based on the uncertainty modeling, we further introduce uncertainty regularization to adapt the matching objective according to the fluctuation range. Compared with existing methods, the proposed strategy explicitly prevents the model from pushing away potential candidates in the early stage, and thus improves the recall rate. On the three public datasets, i.e., FashionIQ, Fashion200k, and Shoes, the proposed method has achieved +4.03%, +3.38%, and +2.40% Recall@50 accuracy over a strong baseline, respectively.

cs.CV↗

arXiv2024

Panoptic Scene Graph Generation with Semantics-Prototype Learning

Panoptic Scene Graph Generation (PSG) parses objects and predicts their relationships (predicate) to connect human language and visual scenes. However, different language preferences of annotators and semantic overlaps between predicates lead to biased predicate annotations in the dataset, i.e. different predicates for same object pairs. Biased predicate annotations make PSG models struggle in constructing a clear decision plane among predicates, which greatly hinders the real application of PSG models. To address the intrinsic bias above, we propose a novel framework named ADTrans to adaptively transfer biased predicate annotations to informative and unified ones. To promise consistency and accuracy during the transfer process, we propose to measure the invariance of representations in each predicate class, and learn unbiased prototypes of predicates with different intensities. Meanwhile, we continuously measure the distribution changes between each presentation and its prototype, and constantly screen potential biased data. Finally, with the unbiased predicate-prototype representation embedding space, biased annotations are easily identified. Experiments show that ADTrans significantly improves the performance of benchmark models, achieving a new state-of-the-art performance, and shows great generalization and effectiveness on multiple datasets.

cs.CV↗