arXiv subjects

Yilin Zhao

Publications and source records attributed to Yilin Zhao.

At least 37 records · Page 2Linked to original sources

arXiv2025

Wavefunction-Free Approach for Predicting Nonlinear Responses in Weyl Semimetals

By sidestepping the intractable calculations of many-body wavefunctions, density functional theory (DFT) has revolutionized the prediction of ground states of materials. However, predicting nonlinear responses--critical for next-generation quantum devices--still relies heavily on explicit wavefunctions, limiting computational efficiency. In this letter, using the circular photogalvanic effect (CPGE) in Weyl semimetals as a representative example, we realize a 1000-fold computational speedup by eliminating the explicit dependence on wavefunctions. Our approach leverages the one-to-one correspondence between free parameters of Weyl fermions and the associated responses to obtain precise wavefunction-free formulations. Applying our methodology, we systematically investigated known Weyl semimetals and revealed that Ta$_3$S$_2$ exhibits photocurrents an order of magnitude greater than those observed in TaAs, with potential for an additional order-of-magnitude enhancement under strain. To further demonstrate the generality of our approach, we obtained a wavefunction-free formula for the Berry-curvature dipole in Weyl semimetals. Our work paves the way for substantially more efficient screening and optimization of nonlinear electromagnetic properties in topological quantum materials.

cond-mat.mtrl-sci↗

arXiv2025

VoxelFormer: Parameter-Efficient Multi-Subject Visual Decoding from fMRI

Recent advances in fMRI-based visual decoding have enabled compelling reconstructions of perceived images. However, most approaches rely on subject-specific training, limiting scalability and practical deployment. We introduce \textbf{VoxelFormer}, a lightweight transformer architecture that enables multi-subject training for visual decoding from fMRI. VoxelFormer integrates a Token Merging Transformer (ToMer) for efficient voxel compression and a query-driven Q-Former that produces fixed-size neural representations aligned with the CLIP image embedding space. Evaluated on the 7T Natural Scenes Dataset, VoxelFormer achieves competitive retrieval performance on subjects included during training with significantly fewer parameters than existing methods. These results highlight token merging and query-based transformers as promising strategies for parameter-efficient neural decoding.

cs.CV↗

arXiv2025

The 9th AI City Challenge

The ninth AI City Challenge continues to advance real-world applications of computer vision and AI in transportation, industrial automation, and public safety. The 2025 edition featured four tracks and saw a 17% increase in participation, with 245 teams from 15 countries registered on the evaluation server. Public release of challenge datasets led to over 30,000 downloads to date. Track 1 focused on multi-class 3D multi-camera tracking, involving people, humanoids, autonomous mobile robots, and forklifts, using detailed calibration and 3D bounding box annotations. Track 2 tackled video question answering in traffic safety, with multi-camera incident understanding enriched by 3D gaze labels. Track 3 addressed fine-grained spatial reasoning in dynamic warehouse environments, requiring AI systems to interpret RGB-D inputs and answer spatial questions that combine perception, geometry, and language. Both Track 1 and Track 3 datasets were generated in NVIDIA Omniverse. Track 4 emphasized efficient road object detection from fisheye cameras, supporting lightweight, real-time deployment on edge devices. The evaluation framework enforced submission limits and used a partially held-out test set to ensure fair benchmarking. Final rankings were revealed after the competition concluded, fostering reproducibility and mitigating overfitting. Several teams achieved top-tier results, setting new benchmarks in multiple tasks.

cs.CV↗

arXiv2025

Innovative Thinking, Infinite Humor: Humor Research of Large Language Models through Structured Thought Leaps

Humor is previously regarded as a gift exclusive to humans for the following reasons. Humor is a culturally nuanced aspect of human language, presenting challenges for its understanding and generation. Humor generation necessitates a multi-hop reasoning process, with each hop founded on proper rationales. Although many studies, such as those related to GPT-o1, focus on logical reasoning with reflection and correction, they still fall short in humor generation. Due to the sparsity of the knowledge graph in creative thinking, it is arduous to achieve multi-hop reasoning. Consequently, in this paper, we propose a more robust framework for addressing the humor reasoning task, named LoL. LoL aims to inject external information to mitigate the sparsity of the knowledge graph, thereby enabling multi-hop reasoning. In the first stage of LoL, we put forward an automatic instruction-evolution method to incorporate the deeper and broader thinking processes underlying humor. Judgment-oriented instructions are devised to enhance the model's judgment capability, dynamically supplementing and updating the sparse knowledge graph. Subsequently, through reinforcement learning, the reasoning logic for each online-generated response is extracted using GPT-4o. In this process, external knowledge is re-introduced to aid the model in logical reasoning and the learning of human preferences. Finally, experimental results indicate that the combination of these two processes can enhance both the model's judgment ability and its generative capacity. These findings deepen our comprehension of the creative capabilities of large language models (LLMs) and offer approaches to boost LLMs' creative abilities for cross-domain innovative applications.

cs.AI↗

arXiv2025

Eagle: Exploring The Design Space for Multimodal LLMs with Mixture of Encoders

The ability to accurately interpret complex visual information is a crucial topic of multimodal large language models (MLLMs). Recent work indicates that enhanced visual perception significantly reduces hallucinations and improves performance on resolution-sensitive tasks, such as optical character recognition and document analysis. A number of recent MLLMs achieve this goal using a mixture of vision encoders. Despite their success, there is a lack of systematic comparisons and detailed ablation studies addressing critical aspects, such as expert selection and the integration of multiple vision experts. This study provides an extensive exploration of the design space for MLLMs using a mixture of vision encoders and resolutions. Our findings reveal several underlying principles common to various existing strategies, leading to a streamlined yet effective design approach. We discover that simply concatenating visual tokens from a set of complementary vision encoders is as effective as more complex mixing architectures or strategies. We additionally introduce Pre-Alignment to bridge the gap between vision-focused encoders and language tokens, enhancing model coherence. The resulting family of MLLMs, Eagle, surpasses other leading open-source models on major MLLM benchmarks.

cs.CV↗

arXiv2025

Broken symmetries associated with a Kagome chiral charge order

Chirality or handedness manifests in all fields of science, ranging from cell biology, molecular interaction, and catalysis to different branches of physics. In condensed matter physics, chirality is intrinsic to enigmatic quantum phases, such as chiral charge density waves and chiral superconductivity. Here, the underlying chiral response is subtle and leads to broken symmetries in the ground state. Detection of subtle broken symmetries is the key to understand these quantum states but they are extremely challenging to expose leading to debate and controversy. Here, using second-order optical response, we uncover the broken symmetries of a chiral charge density wave in the Kagome lattice KV3Sb5, revealing the relevant broken symmetries of its charge order. KV3Sb5 undergoes a phase transition to a charge-ordered state at low temperatures. Our polarization-dependent mid-infrared photocurrent microscopy reveals an intrinsic, longitudinal helicity-dependent photocurrent associated with the charge order. Our measurements, supported by our theoretical analysis, provide direct evidence for broken inversion and mirror symmetries at the charge order transition, indicating a chiral charge ordered state. On the other hand, we do not observe a circular photogalvanic effect along the direction perpendicular to that of the incident light, imposing stringent constraints on the rotational and point group symmetries of the charge order. Our study not only visualizes the chiral nature of the Kagome charge order revealing its broken symmetries, but also highlights the nonlinear photogalvanic effect as a sensitive probe for detecting subtle symmetry breakings.

cond-mat.str-el↗

arXiv2025

Eagle 2: Building Post-Training Data Strategies from Scratch for Frontier Vision-Language Models

Recently, promising progress has been made by open-source vision-language models (VLMs) in bringing their capabilities closer to those of proprietary frontier models. However, most open-source models only publish their final model weights, leaving the critical details of data strategies and implementation largely opaque. In this work, we address VLM post-training from a data-centric perspective, showing the key role of data strategy in developing frontier VLMs. By studying and building our post-training data strategy from scratch, we share detailed insights into the development processes, aiming to benefit the development of competitive models for the open-source community. Our introduced data strategy, together with training recipes and model design, leads to a family of performant VLMs named Eagle2. Specifically, Eagle2-9B achieves state-of-the-art results across various multimodal benchmarks, matching certain competitive models with up to 70B parameters.

cs.CV↗

arXiv2024

This took us a Weyl: synthesis of a semimetallic Weyl ferromagnet with point Fermi surface

Quantum materials governed by emergent topological fermions have become a cornerstone of physics. Dirac fermions in graphene form the basis for moiré quantum matter, and Dirac fermions in magnetic topological insulators enabled the discovery of the quantum anomalous Hall effect. In contrast, there are few materials whose electromagnetic response is dominated by emergent Weyl fermions. Nearly all known Weyl materials are overwhelmingly metallic, and are largely governed by irrelevant, conventional electrons. Here we theoretically predict and experimentally observe a semimetallic Weyl ferromagnet in van der Waals (Cr,Bi)$_2$Te$_3$. In transport, we find a record bulk anomalous Hall angle $> 0.5$ along with non-metallic conductivity, a regime sharply distinct from conventional ferromagnets. Together with symmetry analysis, our data suggest a semimetallic Fermi surface composed of two Weyl points, with a giant separation $> 75\%$ of the linear dimension of the bulk Brillouin zone, and no other electronic states. Using state-of-the-art crystal synthesis techniques, we widely tune the electronic structure, allowing us to annihilate the Weyl state and visualize a unique topological phase diagram exhibiting broad Chern insulating, Weyl semimetallic and magnetic semiconducting regions. Our observation of a semimetallic Weyl ferromagnet offers an avenue toward novel correlated states and non-linear phenomena, as well as zero-magnetic-field Weyl spintronic and optical devices.

cond-mat.mes-hall↗

arXiv2024

MetaTool: Facilitating Large Language Models to Master Tools with Meta-task Augmentation

Utilizing tools with Large Language Models (LLMs) is essential for grounding AI agents in real-world applications. The prevailing approach involves few-shot prompting with demonstrations or fine-tuning with expert annotations. However, mere in-context demonstrations may fail to cover sufficient knowledge for complex tools and tasks. Training on solution paths is also hindered by the high cost of expert annotations and generalizing to new tools. A core challenge of generalizable tool use lies in understanding the "meta", or fundamental natures of tools that are transferable across tasks, such as causality and constraints. In this paper, we present MetaTool, a novel tool learning methodology designed to generalize across any reusable toolset. Our approach incorporates a self-supervised augmentation technique derived from a series of meta-tasks. This involves predicting masked elements in the tool execution process. The self-supervised procedure enables scalable generation of high-quality QA data, which is handy for supervising tool understanding. By incorporating meta-task data into task-oriented training, our method significantly enhances the performance of open-source LLMs, achieving results comparable to ChatGPT in both tool-based planning and chatting scenarios. Through large-scale instruction tuning, the MetaTool model demonstrates impressive zero-shot generalizability on new tasks.

cs.CL↗

arXiv2024

PLLaVA : Parameter-free LLaVA Extension from Images to Videos for Video Dense Captioning

Vision-language pre-training has significantly elevated performance across a wide range of image-language applications. Yet, the pre-training process for video-related tasks demands exceptionally large computational and data resources, which hinders the progress of video-language models. This paper investigates a straight-forward, highly efficient, and resource-light approach to adapting an existing image-language pre-trained model for dense video understanding. Our preliminary experiments reveal that directly fine-tuning pre-trained image-language models with multiple frames as inputs on video datasets leads to performance saturation or even a drop. Our further investigation reveals that it is largely attributed to the bias of learned high-norm visual features. Motivated by this finding, we propose a simple but effective pooling strategy to smooth the feature distribution along the temporal dimension and thus reduce the dominant impacts from the extreme features. The new model is termed Pooling LLaVA, or PLLaVA in short. PLLaVA achieves new state-of-the-art performance on modern benchmark datasets for both video question-answer and captioning tasks. Notably, on the recent popular VideoChatGPT benchmark, PLLaVA achieves a score of 3.48 out of 5 on average of five evaluated dimensions, exceeding the previous SOTA results from GPT4V (IG-VLM) by 9%. On the latest multi-choice benchmark MVBench, PLLaVA achieves 58.1% accuracy on average across 20 sub-tasks, 14.5% higher than GPT4V (IG-VLM). Code is available at https://pllava.github.io/

cs.CV↗

arXiv2023

ChatAnything: Facetime Chat with LLM-Enhanced Personas

In this technical report, we target generating anthropomorphized personas for LLM-based characters in an online manner, including visual appearance, personality and tones, with only text descriptions. To achieve this, we first leverage the in-context learning capability of LLMs for personality generation by carefully designing a set of system prompts. We then propose two novel concepts: the mixture of voices (MoV) and the mixture of diffusers (MoD) for diverse voice and appearance generation. For MoV, we utilize the text-to-speech (TTS) algorithms with a variety of pre-defined tones and select the most matching one based on the user-provided text description automatically. For MoD, we combine the recent popular text-to-image generation techniques and talking head algorithms to streamline the process of generating talking objects. We termed the whole framework as ChatAnything. With it, users could be able to animate anything with any personas that are anthropomorphic using just a few text inputs. However, we have observed that the anthropomorphic objects produced by current generative models are often undetectable by pre-trained face landmark detectors, leading to failure of the face motion generation, even if these faces possess human-like appearances because those images are nearly seen during the training (e.g., OOD samples). To address this issue, we incorporate pixel-level guidance to infuse human face landmarks during the image generation phase. To benchmark these metrics, we have built an evaluation dataset. Based on it, we verify that the detection rate of the face landmark is significantly increased from 57.0% to 92.5% thus allowing automatic face animation based on generated speech content. The code and more results can be found at https://chatanything.github.io/.

cs.CV↗

arXiv2023

Multi-grained Evidence Inference for Multi-choice Reading Comprehension

Multi-choice Machine Reading Comprehension (MRC) is a major and challenging task for machines to answer questions according to provided options. Answers in multi-choice MRC cannot be directly extracted in the given passages, and essentially require machines capable of reasoning from accurate extracted evidence. However, the critical evidence may be as simple as just one word or phrase, while it is hidden in the given redundant, noisy passage with multiple linguistic hierarchies from phrase, fragment, sentence until the entire passage. We thus propose a novel general-purpose model enhancement which integrates multi-grained evidence comprehensively, named Multi-grained evidence inferencer (Mugen), to make up for the inability. Mugen extracts three different granularities of evidence: coarse-, middle- and fine-grained evidence, and integrates evidence with the original passages, achieving significant and consistent performance improvement on four multi-choice MRC benchmarks.

cs.CL↗

arXiv2023

Spinor-dominated magnetoresistance driven by the topological phase transition in $β$-Ag$_2$Se

A topological insulator is a quantum material which possesses conducting surfaces and an insulating bulk. Despite extensive researches on the properties of Dirac surface states, the characteristics of bulk states have remained largely unexplored. Here we report the observation of spinor-dominated magnetoresistance anomalies in the topological insulator $β$-Ag$_2$Se, induced by a magnetic-field-driven band topological phase transition. These anomalies are caused by intrinsic orthogonality in the wave-function spinors of the last Landau bands of the bulk states, in which backscattering is strictly forbidden during a band topological phase transition. This new type of longitudinal magnetoresistance, purely controlled by the wave-function spinors of the last Landau bands, highlights a unique signature of electrical transport around the band topological phase transition. With further reducing the quantum limit and gap size in $β$-Ag$_2$Se, our results may also suggest possible device applications based on this spinor-dominated mechanism and signify a rare case where topology enters the realm of magnetoresistance control.

cond-mat.mes-hall↗

arXiv2023

Transverse circular photogalvanic effect associated with Lorentz-violating Weyl fermions

Nonlinear optical responses of quantum materials have recently undergone dramatic developments to unveil nontrivial geometry and topology. A remarkable example is the quantized longitudinal circular photogalvanic effect (CPGE) associated with the Chern number of Weyl fermions, while the physics of transverse CPGE in Weyl semimetals remains exclusive. Here, we show that the transverse CPGE of Lorentz invariant Weyl fermions is forced to be zero. We find that the transverse photocurrents of Weyl fermions are associated not only with the Chern numbers but also with the degree of Lorentz-symmetry breaking in condensed matter systems. Based on the generic two-band model analysis, we provide a new powerful equation to calculate the transverse CPGE based on the tilting and warping terms of Weyl fermions. Our results are more capable in designing large transverse CPGE of Weyl semimetals in experiments and are applied to more than tens of Weyl materials to estimate their photocurrents. Our method paves the way to study the CPGE of massless or massive quasiparticles to design next-generation quantum optoelectronics.

cond-mat.mtrl-sci↗

arXiv2022

Reference Knowledgeable Network for Machine Reading Comprehension

Multi-choice Machine Reading Comprehension (MRC) as a challenge requires models to select the most appropriate answer from a set of candidates with a given passage and question. Most of the existing researches focus on the modeling of specific tasks or complex networks, without explicitly referring to relevant and credible external knowledge sources, which are supposed to greatly make up for the deficiency of the given passage. Thus we propose a novel reference-based knowledge enhancement model called Reference Knowledgeable Network (RekNet), which simulates human reading strategies to refine critical information from the passage and quote explicit knowledge in necessity. In detail, RekNet refines finegrained critical information and defines it as Reference Span, then quotes explicit knowledge quadruples by the co-occurrence information of Reference Span and candidates. The proposed RekNet is evaluated on three multi-choice MRC benchmarks: RACE, DREAM and Cosmos QA, obtaining consistent and remarkable performance improvement with observable statistical significance level over strong baselines. Our code is available at https://github.com/Yilin1111/RekNet.

cs.CL↗

arXiv2022

Lite Unified Modeling for Discriminative Reading Comprehension

As a broad and major category in machine reading comprehension (MRC), the generalized goal of discriminative MRC is answer prediction from the given materials. However, the focuses of various discriminative MRC tasks may be diverse enough: multi-choice MRC requires model to highlight and integrate all potential critical evidence globally; while extractive MRC focuses on higher local boundary preciseness for answer extraction. Among previous works, there lacks a unified design with pertinence for the overall discriminative MRC tasks. To fill in above gap, we propose a lightweight POS-Enhanced Iterative Co-Attention Network (POI-Net) as the first attempt of unified modeling with pertinence, to handle diverse discriminative MRC tasks synchronously. Nearly without introducing more parameters, our lite unified design brings model significant improvement with both encoder and decoder components. The evaluation results on four discriminative MRC benchmarks consistently indicate the general effectiveness and applicability of our model, and the code is available at https://github.com/Yilin1111/poi-net.

cs.CL↗

arXiv2021

DocOIE: A Document-level Context-Aware Dataset for OpenIE

Open Information Extraction (OpenIE) aims to extract structured relational tuples (subject, relation, object) from sentences and plays critical roles for many downstream NLP applications. Existing solutions perform extraction at sentence level, without referring to any additional contextual information. In reality, however, a sentence typically exists as part of a document rather than standalone; we often need to access relevant contextual information around the sentence before we can accurately interpret it. As there is no document-level context-aware OpenIE dataset available, we manually annotate 800 sentences from 80 documents in two domains (Healthcare and Transportation) to form a DocOIE dataset for evaluation. In addition, we propose DocIE, a novel document-level context-aware OpenIE model. Our experimental results based on DocIE demonstrate that incorporating document-level context is helpful in improving OpenIE performance. Both DocOIE dataset and DocIE model are released for public.

cs.CL↗

arXiv2021

A new type-II lepidocrocite-type TiO2/GaSe heterostructure: Electronic and optical properties, bandgap engineering, interaction with ultrafast laser pulses

Recently, van der Waals heterostructure has attracted interest both theoretically and experimentally for their potential applications in photoelectronic devices, photovoltaic devices, plasmonic devices and photocatalysis. Inspired by this, we design a lepidocrocite-type TiO2/GaSe heterostructure. Via first-principles simulations, we show that such a heterostructure is a direct bandgap semiconductor with a strong and broad optical absorption, ranging from visible light to UV region, exhibiting its potential application in photoelectronic and photovoltaic devices. With the planar-averaged electron density difference and Bader charge analysis, the heterostructure shows a strong capacity of enhancing the charge redistribution especially at the interface, prolonging the lifetime of excitons, and hence improving photocatalytic performance. By applying biaxial strain and interlayer coupling, the heterostructure exhibits a direct-indirect bandgap transition and shows a potential for mechanical sensors due to the smooth and linear variation of bandgaps. Furthermore, our result indicates that a lower interlayer distance leads to a stronger charge redistribution. The calculation of irradiating ultrafast on the heterostructure further reveals a semiconductor-metal transition for the heterostructure. Moreover, we find an enhanced induced plasmonic current in the heterostructure under both x-polarized and z-polarized laser, which is beneficial to plasmonic devices designs. Our research provides valuable insight in applying the lepidocrocite-type TiO2/GaSe heterostructure in photoelectronic, photovoltaic, photocatalytic, mechanical sensing and plasmonic realms.

cond-mat.mtrl-sci↗