arXiv subjects

Yi Yang

Publications and source records attributed to Yi Yang.

At least 307 records · Page 17Linked to original sources

arXiv2025

Evaluating and Aligning Human Economic Risk Preferences in LLMs

Large Language Models (LLMs) are increasingly used in decision-making scenarios that involve risk assessment, yet their alignment with human economic rationality remains unclear. In this study, we investigate whether LLMs exhibit risk preferences consistent with human expectations across different personas. Specifically, we assess whether LLM-generated responses reflect appropriate levels of risk aversion or risk-seeking behavior based on individual's persona. Our results reveal that while LLMs make reasonable decisions in simplified, personalized risk contexts, their performance declines in more complex economic decision-making tasks. To address this, we propose an alignment method designed to enhance LLM adherence to persona-specific risk preferences. Our approach improves the economic rationality of LLMs in risk-related applications, offering a step toward more human-aligned AI decision-making.

econ.GN↗

arXiv2025

GAPrune: Gradient-Alignment Pruning for Domain-Aware Embeddings

Domain-specific embedding models have shown promise for applications that require specialized semantic understanding, such as coding agents and financial retrieval systems, often achieving higher performance gains than general models. However, state-of-the-art embedding models are typically based on LLMs, which contain billions of parameters, making deployment challenging in resource-constrained environments. Model compression through pruning offers a promising solution, but existing pruning methods treat all parameters uniformly, failing to distinguish between general semantic representations and domain-specific patterns, leading to suboptimal pruning decisions. Thus, we propose GAPrune, a pruning framework that addresses this challenge by considering both domain importance and preserving general linguistic foundation. Our method uses Fisher Information to measure importance and general-domain gradient alignment to assess parameter behavior, then combines these signals using our Domain Alignment Importance (DAI) scoring. Lower DAI scores indicate that the parameter is either less important for the domain task or creates conflicts between domain and general objectives. Experiments on two domain benchmarks, FinMTEB and ChemTEB, show that GAPrune maintains performance within 2.5% of dense models in one-shot pruning at 50% sparsity, while outperforming all baselines. With retraining in 100 steps, GAPrune achieves +4.51% improvement on FinMTEB and +1.73% on ChemTEB, demonstrating that our pruning strategy not only preserves but enhances domain-specific capabilities. Our findings demonstrate that principled pruning strategies can achieve model compression and enhanced domain specialization, providing the research community with a new approach for development.

cs.CL↗

arXiv2025

FinMTEB: Finance Massive Text Embedding Benchmark

Embedding models play a crucial role in representing and retrieving information across various NLP applications. Recent advances in large language models (LLMs) have further enhanced the performance of embedding models. While these models are often benchmarked on general-purpose datasets, real-world applications demand domain-specific evaluation. In this work, we introduce the Finance Massive Text Embedding Benchmark (FinMTEB), a specialized counterpart to MTEB designed for the financial domain. FinMTEB comprises 64 financial domain-specific embedding datasets across 7 tasks that cover diverse textual types in both Chinese and English, such as financial news articles, corporate annual reports, ESG reports, regulatory filings, and earnings call transcripts. We also develop a finance-adapted model, Fin-E5, using a persona-based data synthetic method to cover diverse financial embedding tasks for training. Through extensive evaluation of 15 embedding models, including Fin-E5, we show three key findings: (1) performance on general-purpose benchmarks shows limited correlation with financial domain tasks; (2) domain-adapted models consistently outperform their general-purpose counterparts; and (3) surprisingly, a simple Bag-of-Words (BoW) approach outperforms sophisticated dense embeddings in financial Semantic Textual Similarity (STS) tasks, underscoring current limitations in dense embedding techniques. Our work establishes a robust evaluation framework for financial NLP applications and provides crucial insights for developing domain-specific embedding models.

cs.CL↗

arXiv2025

Time-dependent correlations of the Edwards-Anderson order parameter above the spin-glass transition

In 1975 Edwards and Anderson introduced a new paradigm that interacting quenched systems, such as a spin-glass, have a phase transition in which long time memory of spatial patterns is realized without spatial correlations. We show here that the information about the time-dependent correlations above the spin-glass transition are embedded in the four spin correlations of the intensity of speckle pattern. This encodes the spin-orientation memory and can be measured by the technique of resonant magnetic x-ray photon correlation spectroscopy (RM- XPCS). We have implemented this method to observe and accurately characterize the critical slowing down of the spin orientation fluctuations in the classic metallic spin glass alloy $Cu_{1-x}{Mn}_x$ over time scales of ${2}$ sec. to $2 \times 10^{\mathbf{4}}$ secs. Remarkably the divergence of the correlation time as a function of temperature is consistent with the Vogel-Vulcher law, universally used to characterize the viscous relaxation time in structural glasses. Our method also opens the way for studying phase transitions in systems such as spin ices, quantum spin liquids, the structural glass transition, as well as possibly provide new perspectives on the multifarious problems in which spin-glass concepts have found applications.

cond-mat.dis-nn↗

arXiv2025

Dropping Experts, Recombining Neurons: Retraining-Free Pruning for Sparse Mixture-of-Experts LLMs

Sparse Mixture-of-Experts (SMoE) architectures are widely used in large language models (LLMs) due to their computational efficiency. However, though only a few experts are activated for each token, SMoE still requires loading all expert parameters, leading to high memory usage and challenges in deployment. Previous work has tried to reduce the overhead by pruning and merging experts, but primarily focused on expert-level operations, leaving neuron-level structure underexplored. We propose DERN (Dropping Experts, Recombining Neurons), a task-agnostic and retraining-free framework for expert pruning and reconstruction. We observe that experts are often misaligned and contain semantic conflicts at the neuron level, which poses challenges for direct merging. To solve this, DERN works in three steps: it first prunes redundant experts using router statistics; then it decomposes them into neuron-level expert segments, assigning each segment to its most compatible retained expert; and finally, it merges segments within each retained expert to build a compact representation. Experiments on Mixtral, Qwen, and DeepSeek SMoE models show that DERN improves performance by more than 5% on commonsense reasoning and MMLU benchmarks under 50% expert sparsity, without extra training. It also greatly reduces the number of experts and memory usage, making SMoE LLMs easier to deploy in practice.

cs.CL↗

arXiv2025

PersonaFuse: A Personality Activation-Driven Framework for Enhancing Human-LLM Interactions

Recent advancements in Large Language Models (LLMs) demonstrate remarkable capabilities across various fields. These developments have led to more direct communication between humans and LLMs in various situations, such as social companionship and psychological support. However, LLMs often exhibit limitations in emotional perception and social competence during real-world conversations. These limitations partly originate from their inability to adapt their communication style and emotional expression to different social and task contexts. In this work, we introduce PersonaFuse, a novel LLM post-training framework that enables LLMs to adapt and express different personalities for varying situations. Inspired by Trait Activation Theory and the Big Five personality model, PersonaFuse employs a Mixture-of-Expert architecture that combines persona adapters with a dynamic routing network, enabling contextual trait expression. Experimental results show that PersonaFuse substantially outperforms baseline models across multiple dimensions of social-emotional intelligence. Importantly, these gains are achieved without sacrificing general reasoning ability or model safety, which remain common limitations of direct prompting and supervised fine-tuning approaches. PersonaFuse also delivers consistent improvements in downstream human-centered applications, such as mental health counseling and review-based customer service. Finally, human preference evaluations against leading LLMs, including GPT-4o and DeepSeek, demonstrate that PersonaFuse achieves competitive response quality despite its comparatively smaller model size. These findings demonstrate that PersonaFuse offers a theoretically grounded and practical approach for developing social-emotional enhanced LLMs, marking a significant advancement toward more human-centric AI systems.

cs.CL↗

arXiv2025

Non-Abelian Electric Field and Zitterbewegung on a Photonic Frequency Chain

The synthetic frequency dimension, which realizes fictitious spatial dimensions from the spectral degree of freedom, has emerged as a promising platform for engineering artificial gauge fields in studying quantum simulations and topological physics with photons. A current central task for frequency-domain photons is the creation and manipulation of nontrivial non-Abelian field strength tensors and observing their governing dynamics. Here, we experimentally demonstrate a miniaturized scheme for creating non-Abelian electric fields in a photonic frequency chain using a polarization-multiplexed, time-modulated ring resonator. By engineering spin-orbit coupling via modulation dephasing, polarization rotation, and polarization retardation, we achieve programmable control over synthetic Floquet bands and their quasimomentum spin-resolved textures. Leveraging self-heterodyne coherent detection, we demonstrate Zitterbewegung -- a trembling motion of photons -- induced by non-Abelian electric fields on the frequency chain. We further observe the interference between Zitterbewegung and Bloch oscillations arising from the coexistence of non-Abelian and Abelian electric fields. Our work bridges synthetic dimensions with non-Abelian gauge theory for versatile photonic emulation of relativistic quantum mechanics and spinor dynamics, and can be instrumental in applications like frequency-domain optical computation and multimodal frequency comb control.

physics.optics↗

arXiv2025

A Survey of World Models for Autonomous Driving

Recent breakthroughs in autonomous driving have been propelled by advances in robust world modeling, fundamentally transforming how vehicles interpret dynamic scenes and execute safe decision-making. World models have emerged as a linchpin technology, offering high-fidelity representations of the driving environment that integrate multi-sensor data, semantic cues, and temporal dynamics. This paper systematically reviews recent advances in world models for autonomous driving, proposing a three-tiered taxonomy: (i) Generation of Future Physical World, covering Image-, BEV-, OG-, and PC-based generation methods that enhance scene evolution modeling through diffusion models and 4D occupancy forecasting; (ii) Behavior Planning for Intelligent Agents, combining rule-driven and learning-based paradigms with cost map optimization and reinforcement learning for trajectory generation in complex traffic conditions; (ii) Interaction between Prediction and Planning, achieving multi-agent collaborative decision-making through latent space diffusion and memory-augmented architectures. The study further analyzes training paradigms, including self-supervised learning, multimodal pretraining, and generative data augmentation, while evaluating world models' performance in scene understanding and motion prediction tasks. Future research must address key challenges in self-supervised representation learning, multimodal fusion, and advanced simulation to advance the practical deployment of world models in complex urban environments. Overall, the comprehensive analysis provides a technical roadmap for harnessing the transformative potential of world models in advancing safe and reliable autonomous driving solutions.

cs.RO↗

arXiv2025

OmniMap: A General Mapping Framework Integrating Optics, Geometry, and Semantics

Robotic systems demand accurate and comprehensive 3D environment perception, requiring simultaneous capture of photo-realistic appearance (optical), precise layout shape (geometric), and open-vocabulary scene understanding (semantic). Existing methods typically achieve only partial fulfillment of these requirements while exhibiting optical blurring, geometric irregularities, and semantic ambiguities. To address these challenges, we propose OmniMap. Overall, OmniMap represents the first online mapping framework that simultaneously captures optical, geometric, and semantic scene attributes while maintaining real-time performance and model compactness. At the architectural level, OmniMap employs a tightly coupled 3DGS-Voxel hybrid representation that combines fine-grained modeling with structural stability. At the implementation level, OmniMap identifies key challenges across different modalities and introduces several innovations: adaptive camera modeling for motion blur and exposure compensation, hybrid incremental representation with normal constraints, and probabilistic fusion for robust instance-level understanding. Extensive experiments show OmniMap's superior performance in rendering fidelity, geometric accuracy, and zero-shot semantic segmentation compared to state-of-the-art methods across diverse scenes. The framework's versatility is further evidenced through a variety of downstream applications, including multi-domain scene Q&A, interactive editing, perception-guided manipulation, and map-assisted navigation.

cs.RO↗

arXiv2025

Revealing the Numeracy Gap: An Empirical Investigation of Text Embedding Models

Text embedding models are widely used in natural language processing applications. However, their capability is often benchmarked on tasks that do not require understanding nuanced numerical information in text. As a result, it remains unclear whether current embedding models can precisely encode numerical content, such as numbers, into embeddings. This question is critical because embedding models are increasingly applied in domains where numbers matter, such as finance and healthcare. For example, Company X's market share grew by 2\% should be interpreted very differently from Company X's market share grew by 20\%, even though both indicate growth in market share. This study aims to examine whether text embedding models can capture such nuances. Using synthetic data in a financial context, we evaluate 13 widely used text embedding models and find that they generally struggle to capture numerical details accurately. Our further analyses provide deeper insights into embedding numeracy, informing future research to strengthen embedding model-based NLP systems with improved capacity for handling numerical content.

cs.CL↗

arXiv2025

2nd Place Solution for CVPR2024 E2E Challenge: End-to-End Autonomous Driving Using Vision Language Model

End-to-end autonomous driving has drawn tremendous attention recently. Many works focus on using modular deep neural networks to construct the end-to-end archi-tecture. However, whether using powerful large language models (LLM), especially multi-modality Vision Language Models (VLM) could benefit the end-to-end driving tasks remain a question. In our work, we demonstrate that combining end-to-end architectural design and knowledgeable VLMs yield impressive performance on the driving tasks. It is worth noting that our method only uses a single camera and is the best camera-only solution across the leaderboard, demonstrating the effectiveness of vision-based driving approach and the potential for end-to-end driving tasks.

cs.CV↗

arXiv2025

OpenMulti: Open-Vocabulary Instance-Level Multi-Agent Distributed Implicit Mapping

Multi-agent distributed collaborative mapping provides comprehensive and efficient representations for robots. However, existing approaches lack instance-level awareness and semantic understanding of environments, limiting their effectiveness for downstream applications. To address this issue, we propose OpenMulti, an open-vocabulary instance-level multi-agent distributed implicit mapping framework. Specifically, we introduce a Cross-Agent Instance Alignment module, which constructs an Instance Collaborative Graph to ensure consistent instance understanding across agents. To alleviate the degradation of mapping accuracy due to the blind-zone optimization trap, we leverage Cross Rendering Supervision to enhance distributed learning of the scene. Experimental results show that OpenMulti outperforms related algorithms in both fine-grained geometric accuracy and zero-shot semantic accuracy. In addition, OpenMulti supports instance-level retrieval tasks, delivering semantic annotations for downstream applications. The project website of OpenMulti is publicly available at https://openmulti666.github.io/.

cs.RO↗

arXiv2025

The Most Luminous Known Fast Blue Optical Transient AT 2024wpp: Unprecedented Evolution and Properties in the Ultraviolet to the Near-Infrared

We present an extensive photometric and spectroscopic ultraviolet-optical-infrared campaign on the luminous fast blue optical transient (LFBOT) AT 2024wpp over the first ~100 d. AT 2024wpp is the most luminous LFBOT discovered to date, with $L_{\rm{pk}}\approx(2-4)\times10^{45}$ erg s$^{-1}$ (5-10 times that of the prototypical AT 2018cow). This extreme luminosity enabled the acquisition of the most detailed LFBOT UV light curve thus far. In the first ~45 d, AT 2024wpp radiated $>10^{51}$ erg, surpassing AT 2018cow by an order of magnitude and requiring a power source beyond the radioactive $^{56}$Ni decay of traditional supernovae. Like AT 2018cow, the UV-optical spectrum of AT 2024wpp is dominated by a persistently blue thermal continuum throughout our monitoring, with blackbody parameters at peak of T>30,000 K and $R_{\rm{BB}}/t\approx0.2-0.3c$. A temperature of $\gtrsim$20,000 K is maintained thereafter without evidence for cooling. We interpret the featureless spectra as a consequence of continuous energy injection from a central source of high-energy emission which maintains high ejecta ionization. After 35 d, faint (equivalent width <10 Å) H and He spectral features with kinematically separate velocity components centered at 0 km s$^{-1}$ and -6400 km s$^{-1}$ emerge, implying spherical symmetry deviations. A near-infrared excess of emission above the optical blackbody emerges between 20-30 d with a power-law spectrum $F_{\rmν,NIR}\proptoν^{-0.3}$ at 30 d. We interpret this distinct emission component as either reprocessing of early UV emission in a dust echo or free-free emission in an extended medium above the optical photosphere. LFBOT asphericity and multiple outflow components (including mildly relativistic ejecta) together with the large radiated energy are naturally realized by super-Eddington accretion disks around neutron stars or black holes and their outflows.

astro-ph.HE↗

arXiv2025

Local Observability of a Class of Feedforward Neural Networks

Beyond the traditional neural network training methods based on gradient descent and its variants, state estimation techniques have been proposed to determine a set of ideal weights from a control-theoretic perspective. Hence, the concept of observability becomes relevant in neural network training. In this paper, we investigate local observability of a class of two-layer feedforward neural networks~(FNNs) with rectified linear unit~(ReLU) activation functions. We analyze local observability of FNNs by evaluating an observability rank condition with respect to the weight matrix and the input sequence. First, we show that, in general, the weights of FNNs are not locally observable. Then, we provide sufficient conditions on the network structures and the weights that lead to local observability. Moreover, we propose an input design approach to render the weights distinguishable and show that this input also excites other weights inside a neighborhood. Finally, we validate our results through a numerical example.

eess.SY↗

arXiv2025

Image-based modelling of elastic properties using unresolved rock images

The trade-off between image resolution and model field-of-view has long been a limitation for numerical simulations in digital rock models. A significant amount of sub-resolution pore space cannot be captured in the unresolved digital rock images, which hinders the accuracy of numerical simulations, especially those predicting the rock effective elastic properties. This work uses paired digital rock images at multiple resolutions to investigate the sub-resolution solid and pore fraction distributions. It demonstrates that the cumulative Beta distribution function can effectively represent the solid and pore fractions in unresolved rock images. Based on this finding, we propose a novel methodology to predict the sub-resolution pore fractions. Compared to the pore fractions extracted from paired Bentheimer sandstone images at resolutions of 2, 6, and 18um, the proposed method yields double-weighted mean absolute percentage errors (WWMAPE) of 3.67% (6um) and 13.78% (18um), significantly lower than the SLOT technique's errors of 24.55% (6um) and 59.33% (18um). By incorporating the modelled solid and pore fractions with effective medium theory, this work achieves improved predictions of effective elastic moduli and wave velocities across various rock types. This method has significant implications for conducting enhanced simulations of geophysical properties in unresolved rock models.

physics.geo-ph↗

arXiv2025

AutoScale: Linear Scalarization Guided by Multi-Task Optimization Metrics

Recent multi-task learning studies suggest that linear scalarization, when using well-chosen fixed task weights, can achieve comparable to or even better performance than complex multi-task optimization (MTO) methods. It remains unclear why certain weights yield optimal performance and how to determine these weights without relying on exhaustive hyperparameter search. This paper establishes a direct connection between linear scalarization and MTO methods, revealing through extensive experiments that well-performing scalarization weights exhibit specific trends in key MTO metrics, such as high gradient magnitude similarity. Building on this insight, we introduce AutoScale, a simple yet effective two-phase framework that uses these MTO metrics to guide weight selection for linear scalarization, without expensive weight search. AutoScale consistently shows superior performance with high efficiency across diverse datasets including a new large-scale benchmark.

cs.LG↗

arXiv2025

Topological quantum electrodynamics in synthetic non-Abelian gauge fields

Quantum electrodynamics (QED), a cornerstone framework that describes light-matter interactions rooted in Abelian symmetries, renders the harnessing of synthetic non-Abelian gauge fields as a fundamental yet uncharted frontier. Here, we develop a general theory of light-matter interaction of quantum emitters embedded in non-Abelian photonic lattices. Based on analytical solutions to the non-Abelian Landau dressed states beyond the continuum limit, we reveal chiral photon emission and vortices with emergent nonreciprocity enabled by selective coupling between emitters and spin-momentum-locked bands. When coexisting with Abelian and non-Abelian magnetic fields, emitters hybridize with Landau dressed orbits to form spin-polarized, squeezed Landau polaritons that carry quantized angular momenta, with Rabi frequencies tunable via Landau levels and pseudospin interactions. Multi-emitter dynamics further exhibit collective phenomena governed by real-space staggered phases induced by nonsymmorphic crystalline symmetry. These results bridge non-Abelian physics with quantum optics, and establish non-Abelian gauge fields as a versatile tool for synthesizing topological quantum optical states, angular momentum transfer, and controlling photon-mediated correlations in QED systems, relevant for applications in quantum simulations and chiral quantum optical networks.

physics.optics↗

arXiv2025

TDDBench: A Benchmark for Training data detection

Training Data Detection (TDD) is a task aimed at determining whether a specific data instance is used to train a machine learning model. In the computer security literature, TDD is also referred to as Membership Inference Attack (MIA). Given its potential to assess the risks of training data breaches, ensure copyright authentication, and verify model unlearning, TDD has garnered significant attention in recent years, leading to the development of numerous methods. Despite these advancements, there is no comprehensive benchmark to thoroughly evaluate the effectiveness of TDD methods. In this work, we introduce TDDBench, which consists of 13 datasets spanning three data modalities: image, tabular, and text. We benchmark 21 different TDD methods across four detection paradigms and evaluate their performance from five perspectives: average detection performance, best detection performance, memory consumption, and computational efficiency in both time and memory. With TDDBench, researchers can identify bottlenecks and areas for improvement in TDD algorithms, while practitioners can make informed trade-offs between effectiveness and efficiency when selecting TDD algorithms for specific use cases. Our extensive experiments also reveal the generally unsatisfactory performance of TDD algorithms across different datasets. To enhance accessibility and reproducibility, we open-source TDDBench for the research community at https://github.com/zzh9568/TDDBench.

cs.CR↗