arXiv subjects

Shiwei Chen

Publications and source records attributed to Shiwei Chen.

18 recordsLinked to original sources

arXiv2026

SemiConLens: Visual Analytics for 2D Semiconductor Discovery

The past few years have witnessed vibrant efforts in discovering new two-dimensional (2D) semiconductor materials from both academia and the industry, due to their promising potential in resolving the severe performance deterioration of traditional semiconductors resulting from condensed silicon thickness. However, existing methods (e.g., Density Functional Theory (DFT) or machine-learning-based approaches) suffer from various challenges such as small datasets, and reliability and trustworthiness issues. To bridge this gap, we propose SemiConLens, a visual analytics approach to combine human expertise with the power of ML to enable effective and reliable 2D semiconductor discovery. Specifically, we first develop a new Correlation Aware Multivariate Imputation (CAMI) method and use ML models like autoencoder, which can better learn from limited data and reveal uncertainty, to address the challenge of sparse data in semiconductivity prediction. Built upon this, our visualization module, consisting of three visualization views with linked interactions, allows material researchers to interactively filter, discover and compare 2D semiconductor candidates. A novel circular glyph design and a new cluster-aware layout optimization approach are proposed to effectively display all the user-configurable key attributes and possible prediction uncertainties of each semiconductor candidate, ensuring a reliable and trustable 2D semiconductor discovery. We assess SemiConLens through quantitative evaluations, expert interviews, and use cases. The results demonstrate SemiConLens's capability to help material researchers conduct effective discovery of desirable 2D semiconductors.

cs.HC↗

arXiv2026

Towards Effective In-context Cross-domain Knowledge Transfer via Domain-invariant-neurons-based Retrieval

Large language models (LLMs) have made notable progress in logical reasoning, yet still fall short of human-level performance. Current boosting strategies rely on expert-crafted in-domain demonstrations, limiting their applicability in expertise-scarce domains, such as specialized mathematical reasoning, formal logic, or legal analysis. In this work, we demonstrate the feasibility of leveraging cross-domain demonstrating examples to boost the LLMs' reasoning performance. Despite substantial domain differences, many reusable implicit logical structures are shared across domains. In order to effectively retrieve cross-domain examples for unseen domains under investigation, in this work, we further propose an effective retrieval method, called domain-invariant neurons-based retrieval (\textbf{DIN-Retrieval}). Concisely, DIN-Retrieval first summarizes a hidden representation that is universal across different domains. Then, during the inference stage, we use the DIN vector to retrieve structurally compatible cross-domain demonstrations for the in-context learning. Experimental results in multiple settings for the transfer of mathematical and logical reasoning demonstrate that our method achieves an average improvement of 1.8 over the state-of-the-art methods \footnote{Our implementation is available at https://github.com/Leon221220/DIN-Retrieval}.

cs.AI↗

arXiv2026

Reason Analogically via Cross-domain Prior Knowledge: An Empirical Study of Cross-domain Knowledge Transfer for In-Context Learning

Despite its success, existing in-context learning (ICL) relies on in-domain expert demonstrations, limiting its applicability when expert annotations are scarce. We posit that different domains may share underlying reasoning structures, enabling source-domain demonstrations to improve target-domain inference despite semantic mismatch. To test this hypothesis, we conduct a comprehensive empirical study of different retrieval methods to validate the feasibility of achieving cross-domain knowledge transfer under the in-context learning setting. Our results demonstrate conditional positive transfer in cross-domain ICL. We identify a clear example absorption threshold: beyond it, positive transfer becomes more likely, and additional demonstrations yield larger gains. Further analysis suggests that these gains stem from reasoning structure repair by retrieved cross-domain examples, rather than semantic cues. Overall, our study validates the feasibility of leveraging cross-domain knowledge transfer to improve cross-domain ICL performance, motivating the community to explore designing more effective retrieval approaches for this novel direction.\footnote{Our implementation is available at https://github.com/littlelaska/ICL-TF4LR}

cs.AI↗

arXiv2026

When the Chain Breaks: Interactive Diagnosis of LLM Chain-of-Thought Reasoning Errors

Current Large Language Models (LLMs), especially Large Reasoning Models, can generate Chain-of-Thought (CoT) reasoning traces to illustrate how they produce final outputs, thereby facilitating trust calibration for users. However, these CoT reasoning traces are usually lengthy and tedious, and can contain various issues, such as logical and factual errors, which make it difficult for users to interpret the reasoning traces efficiently and accurately. To address these challenges, we develop an error detection pipeline that combines external fact-checking with symbolic formal logical validation to identify errors at the step level. Building on this pipeline, we propose ReasonDiag, an interactive visualization system for diagnosing CoT reasoning traces. ReasonDiag provides 1) an integrated arc diagram to show reasoning-step distributions and error-propagation patterns, and 2) a hierarchical node-link diagram to visualize high-level reasoning flows and premise dependencies. We evaluate ReasonDiag through a technical evaluation for the error detection pipeline, two case studies, and user interviews with 16 participants. The results indicate that ReasonDiag helps users effectively understand CoT reasoning traces, identify erroneous steps, and determine their root causes.

cs.HC↗

arXiv2025

Interlayer Coupling Driven Correlated and Charge-Ordered Electronic States in a Transition Metal Dichalcogenide Superlattice

4Hb-TaS_2, a van der Waals superlattice comprising alternate stacked Ising superconducting 1H-TaS_2 and cluster Mott insulating 1T-TaS_2, exhibits emergent properties beyond those of its constituent layers. Notable phenomena include time-reversal-symmetry-breaking superconductivity and spontaneous vortex phases, which are driven by nontrivial interlayer interactions that remain debated. Using area-selective angle-resolved photoemission spectroscopy, we provide direct spectroscopic evidence of such interaction by systematically probing the electronic structures of 1T- and 1H-terminted surfaces of 4Hb-TaS_2. The metallic states of subsurface 1H-layers are folded to the Brillouin zone center by the sqrt(13) by sqrt(13) modulation of the surface 1T-layer, forming chiral "windmill" Fermi surfaces via Umklapp scattering. These conducting states further hybridize with the incipient flat band of the surface 1T-layer, producing a Kondo-like peak at the Fermi level. Interlayer charge transfer induces distinct 3 by 3 and 2 by 2 charge orders on the surface and subsurface 1H-layers, respectively, which result in characteristic segmented Fermi surfaces and dichotomously shift the van Hove singularities. These findings reconcile the competing Kondo and Mott-Hubbard models in this material and emphasize the interplay of flat bands, van hove singularities, charge orders, and unconventional superconductivity in correlated superlattices.

cond-mat.str-el↗

arXiv2025

Spin Fluctuations-induced Unconventional Transverse Spin Current in Spin Degenerate Antiferromagnet

Modern magnetic memory technology requires unconventional transverse spin current to achieve deterministic switching of perpendicular magnetization. Spin current in antiferromagnets (AFMs) has been long thought to be trivial as nonmagnets. Recently, a class of spin-splitting AFMs has been shown to be able to generate unconventional spin current for spin-orbit torque (SOT) applications. However, such AFMs requires specific symmetry breaking, and have been largely restricted by the limited material candidates. Here, we reveal universal spin fluctuation-induced unconventional transverse spin current in conventional spin degenerate AFMs at finite temperature by taking collinear (AFM) L10-MnPt as an example. The field-free switching of perpendicular magnetization was achieved via out-of-plane anti-damping-like SOT, leveraging the spin current with both y- and z-spin polarizations in heterostructures with MnPt functioning as spin current source. Based on symmetry analyses and experimental characterizations of current-induced spin torques, we find that the spin current generated by L10-MnPt exhibits anisotropy that varies with the current direction, enforced by the low symmetry magnetic point group when its N\'eel vector is along [110]. From a fundamental perspective, it would be intriguing to uncover a mechanism underlying the emergence of unconventional spin currents in spin degenerate antiferromagnets, which is highly desirable for spintronic applications.

physics.app-ph↗

arXiv2025

Towards Efficient CoT Distillation: Self-Guided Rationale Selector for Better Performance with Fewer Rationales

Chain-of-thought (CoT) distillation aims to enhance small language models' (SLMs) reasoning by transferring multi-step reasoning capability from the larger teacher models. However, existing work underestimates rationale quality, focusing primarily on data quantity, which may transfer noisy or incorrect information to the student model. To address the above issues, we proposed \textbf{M}odel-\textbf{O}riented \textbf{R}ationale \textbf{S}election \textbf{D}istillation (MoRSD), which can discern and select high quality rationales for distillation to improve performance further. We further propose a Rationale Difficulty (RD) metric to measure the ability of the student model to generate the correct answer under a given rationale. Compared to the baseline, we achieved 4.6$\%$ average improvement on seven datasets over three tasks, using fewer rationales by controlling their accuracy, diversity, and difficulty. Our results reveal that a small portion of the high quality rationales can enhance the reasoning ability of student models than the entire dataset. Our method promises to be a possible solution for efficient CoT distillation. Our code will be released in https://github.com/Leon221220/MoRSD.

cs.CL↗

arXiv2025

From Long to Lean: Performance-aware and Adaptive Chain-of-Thought Compression via Multi-round Refinement

Chain-of-Thought (CoT) reasoning improves performance on complex tasks but introduces significant inference latency due to verbosity. We propose Multiround Adaptive Chain-of-Thought Compression (MACC), a framework that leverages the token elasticity phenomenon--where overly small token budgets can paradoxically increase output length--to progressively compress CoTs via multiround refinement. This adaptive strategy allows MACC to determine the optimal compression depth for each input. Our method achieves an average accuracy improvement of 5.6 percent over state-of-the-art baselines, while also reducing CoT length by an average of 47 tokens and significantly lowering latency. Furthermore, we show that test-time performance--accuracy and token length--can be reliably predicted using interpretable features like perplexity and compression rate on the training set. Evaluated across different models, our method enables efficient model selection and forecasting without repeated fine-tuning, demonstrating that CoT compression is both effective and predictable. Our code will be released in https://github.com/Leon221220/MACC.

cs.CL↗

arXiv2025

GlyphWeaver: Unlocking Glyph Design Creativity with Uniform Glyph DSL and AI

Expressive glyph visualizations provide a powerful and versatile means to represent complex multivariate data through compact visual encodings, but creating custom glyphs remains challenging due to the gap between design creativity and technical implementation. We present GlyphWeaver, a novel interactive system to enable an easy creation of expressive glyph visualizations. Our system comprises three key components: a glyph domain-specific language (GDSL), a GDSL operation management mechanism, and a multimodal interaction interface. The GDSL is a hierarchical container model, where each container is independent and composable, providing a rigorous yet practical foundation for complex glyph visualizations. The operation management mechanism restricts modifications of the GDSL to atomic operations, making it accessible without requiring direct coding. The multimodal interaction interface enables direct manipulation, natural language commands, and parameter adjustments. A multimodal large language model acts as a translator, converting these inputs into GDSL operations. GlyphWeaver significantly lowers the barrier for designers, who often do not have extensive programming skills, to create sophisticated glyph visualizations.

cs.HC↗

arXiv2025

GEPO: Group Expectation Policy Optimization for Stable Heterogeneous Reinforcement Learning

As single-center computing approaches power constraints, decentralized training becomes essential. However, traditional Reinforcement Learning (RL) methods, crucial for enhancing large model post-training, cannot adapt to decentralized distributed training due to the tight coupling between parameter learning and rollout sampling. For this, we propose HeteroRL, a heterogeneous RL architecture that decouples these processes, enabling stable training across geographically distributed nodes connected via the Internet. The core component is Group Expectation Policy Optimization (GEPO), an asynchronous RL algorithm robust to latency caused by network delays or heterogeneity in computational resources. Our study reveals that high latency significantly increases KL divergence, leading to higher variance of importance weights and training instability. GEPO mitigates this issue by using group expectation weighting to exponentially reduce the variance of importance weights, with theoretical guarantees. Experiments show GEPO achieves superior stability - only a 3% performance drop from online to 1800s latency-and reduces the best-to-last gap by 85% versus GSPO (1.8 vs. 12.0) while attaining the highest scores, highlighting its effectiveness in decentralized, resource-heterogeneous environments.

cs.LG↗

arXiv2025

Learning Event Completeness for Weakly Supervised Video Anomaly Detection

Weakly supervised video anomaly detection (WS-VAD) is tasked with pinpointing temporal intervals containing anomalous events within untrimmed videos, utilizing only video-level annotations. However, a significant challenge arises due to the absence of dense frame-level annotations, often leading to incomplete localization in existing WS-VAD methods. To address this issue, we present a novel LEC-VAD, Learning Event Completeness for Weakly Supervised Video Anomaly Detection, which features a dual structure designed to encode both category-aware and category-agnostic semantics between vision and language. Within LEC-VAD, we devise semantic regularities that leverage an anomaly-aware Gaussian mixture to learn precise event boundaries, thereby yielding more complete event instances. Besides, we develop a novel memory bank-based prototype learning mechanism to enrich concise text descriptions associated with anomaly-event categories. This innovation bolsters the text's expressiveness, which is crucial for advancing WS-VAD. Our LEC-VAD demonstrates remarkable advancements over the current state-of-the-art methods on two benchmark datasets XD-Violence and UCF-Crime.

cs.CV↗

arXiv2024

Distilling Fine-grained Sentiment Understanding from Large Language Models

Fine-grained sentiment analysis (FSA) aims to extract and summarize user opinions from vast opinionated text. Recent studies demonstrate that large language models (LLMs) possess exceptional sentiment understanding capabilities. However, directly deploying LLMs for FSA applications incurs high inference costs. Therefore, this paper investigates the distillation of fine-grained sentiment understanding from LLMs into small language models (SLMs). We prompt LLMs to examine and interpret the sentiments of given reviews and then utilize the generated content to pretrain SLMs. Additionally, we develop a comprehensive FSA benchmark to evaluate both SLMs and LLMs. Extensive experiments on this benchmark reveal that: (1) distillation significantly enhances the performance of SLMs in FSA tasks, achieving a 6.00\% improvement in $F_1$-score, and the distilled model can outperform Llama-2-7b with only 220M parameters; (2) distillation equips SLMs with excellent zero-shot sentiment classification capabilities, enabling them to match or even exceed their teacher models. These results suggest that distillation from LLMs is a highly promising direction for FSA. We will release our code, data, and pretrained model weights at https://github.com/HITSZ-HLT/FSA-Distillation.

cs.CL↗

arXiv2024

Mimicking synaptic plasticity with wedged Pt/Co/Pt spin-orbit torque device

We fabricated a wedge-shaped Pt/Co/Pt device with perpendicular magnetic anisotropy and manifested that the Co magnetization can be solely switched by spin-orbit torque without any magnetic field. Similar to the synaptic weight, we observed that the state of Co magnetization (presented by the anomalous Hall resistance RH) of the wedged Pt/Co/Pt device can be tuned continuously with a large number of nonvolatile levels by applied pulse currents. Furthermore, we studied the synaptic plasticity of the wedged Pt/Co/Pt device, including the excitatory postsynaptic potentials or inhibitory postsynaptic potentials and spiking-time-dependent plasticity. The work elucidates the promise of the wedged Pt/Co/Pt device as a candidate for a new type of artificial synaptic device that is induced by a spin current and paves a substantial pathway toward the combination of spintronics and synaptic devices.

physics.app-ph↗

arXiv2024

Self-Training with Pseudo-Label Scorer for Aspect Sentiment Quad Prediction

Aspect Sentiment Quad Prediction (ASQP) aims to predict all quads (aspect term, aspect category, opinion term, sentiment polarity) for a given review, which is the most representative and challenging task in aspect-based sentiment analysis. A key challenge in the ASQP task is the scarcity of labeled data, which limits the performance of existing methods. To tackle this issue, we propose a self-training framework with a pseudo-label scorer, wherein a scorer assesses the match between reviews and their pseudo-labels, aiming to filter out mismatches and thereby enhance the effectiveness of self-training. We highlight two critical aspects to ensure the scorer's effectiveness and reliability: the quality of the training dataset and its model architecture. To this end, we create a human-annotated comparison dataset and train a generative model on it using ranking-based objectives. Extensive experiments on public ASQP datasets reveal that using our scorer can greatly and consistently improve the effectiveness of self-training. Moreover, we explore the possibility of replacing humans with large language models for comparison dataset annotation, and experiments demonstrate its feasibility. We release our code and data at https://github.com/HITSZ-HLT/ST-w-Scorer-ABSA .

cs.CL↗

arXiv2024

SafePaint: Anti-forensic Image Inpainting with Domain Adaptation

Existing image inpainting methods have achieved remarkable accomplishments in generating visually appealing results, often accompanied by a trend toward creating more intricate structural textures. However, while these models excel at creating more realistic image content, they often leave noticeable traces of tampering, posing a significant threat to security. In this work, we take the anti-forensic capabilities into consideration, firstly proposing an end-to-end training framework for anti-forensic image inpainting named SafePaint. Specifically, we innovatively formulated image inpainting as two major tasks: semantically plausible content completion and region-wise optimization. The former is similar to current inpainting methods that aim to restore the missing regions of corrupted images. The latter, through domain adaptation, endeavors to reconcile the discrepancies between the inpainted region and the unaltered area to achieve anti-forensic goals. Through comprehensive theoretical analysis, we validate the effectiveness of domain adaptation for anti-forensic performance. Furthermore, we meticulously crafted a region-wise separated attention (RWSA) module, which not only aligns with our objective of anti-forensics but also enhances the performance of the model. Extensive qualitative and quantitative evaluations show our approach achieves comparable results to existing image inpainting methods while offering anti-forensic capabilities not available in other methods.

cs.CV↗

arXiv2023

Stochastic spin-orbit-torque synapse and its application in uncertainty quantification

Stochasticity plays a significant role in the low-power operation of a biological neural network. In an artificial neural network (ANN), stochasticity also contributes to critical functions such as the uncertainty quantification (UQ) for estimating the probability for the correctness of prediction. This UQ is vital for cutting-edge applications, including medical diagnostics, autopilots, and large language models. Thanks to high computing velocity and low dissipation, a spin-orbit-torque (SOT) device exhibits significant potential for implementing the UQ. However, up until now, the application of UQ for stochastic SOT devices remains unexplored. In this study, based on SOT-induced stochastic magnetic domain wall (DW) motion with varying velocity, we fabricated an SOT synapse that could emulate stochastic weight update following the Spike-Timing-Dependent-Plasticity (STDP) rule. Furthermore, we set up a stochastic Spiking-Neural-Network (SNN), which, when compared to its deterministic counterpart, demonstrates a clear advantage in quantifying uncertainty for diagnosing the type of breast tumor (benign or malignant).

physics.app-ph↗

arXiv2022

Anomalous anisotropy of spin current in a cubic spin source with noncollinear antiferromagnetism

Cubic materials host high crystal symmetry and hence are not expected to support anisotropy in transport phenomena. In contrast to this common expectation, here we report an anomalous anisotropy of spin current can emerge in the (001) film of Mn${_3}$Pt, a noncollinear antiferromagnetic spin source with face-centered cubic structure. Such spin current anisotropy originates from the intertwined time reversal-odd ($T$-odd) and time reversal-even ($T$-even) spin Hall effects. Based on symmetry analyses and experimental characterizations of the current-induced spin torques in Mn${_3}$Pt-based heterostructures, we find that the spin current generated by Mn${_3}$Pt (001) exhibits exotic dependences on the current direction for all the spin components, deviating from that in conventional cubic systems. We also demonstrate that such an anisotropic spin current can be used to realize low-power spintronic applications such as the efficient field-free switching of the perpendicular magnetizations.

cond-mat.mtrl-sci↗

arXiv2017

Current induced magnetization switching in PtCoCr structures with enhanced perpendicular magnetic anisotropy and spin-orbit torques

Magnetic trilayers having large perpendicular magnetic anisotropy (PMA) and high spin-orbit torques (SOTs) efficiency are the key to fabricate nonvolatile magnetic memory and logic devices. In this work, PMA and SOTs are systematically studied in Pt/Co/Cr stacks as a function of Cr thickness. An enhanced perpendicular anisotropy field around 10189 Oe is obtained and is related to the interface between Co and Cr layers. In addition, an effective spin Hall angle up to 0.19 is observed due to the improved antidamping-like torque by employing dissimilar metals Pt and Cr with opposite signs of spin Hall angles on opposite sides of Co layer. Finally, we observed a nearly linear dependence between spin Hall angle and longitudinal resistivity from their temperature dependent properties, suggesting that the spin Hall effect may arise from extrinsic skew scattering mechanism. Our results indicate that 3d transition metal Cr with a large negative spin Hall angle could be used to engineer the interfaces of trilayers to enhance PMA and SOTs.

physics.app-ph↗