arXiv subjects

Yiran Liu

Publications and source records attributed to Yiran Liu.

At least 19 recordsLinked to original sources

arXiv2026

Semantic-Topological Graph Reasoning for Language-Guided Pulmonary Screening

Medical image segmentation driven by free-text clinical instructions is a critical frontier in computer-aided diagnosis. However, existing multimodal and foundation models struggle with the semantic ambiguity of clinical reports and fail to disambiguate complex anatomical overlaps in low-contrast scans. Furthermore, fully fine-tuning these massive architectures on limited medical datasets invariably leads to severe overfitting. To address these challenges, we propose a novel Semantic-Topological Graph Reasoning (STGR) framework for language-guided pulmonary screening. Our approach elegantly synergizes the reasoning capabilities of large language models (LLaMA-3-V) with the zero-shot delineation of vision foundation models (MedSAM). Specifically, we introduce a Text-to-Vision Intent Distillation (TVID) module to extract precise diagnostic guidance. To resolve anatomical ambiguity, we formulate mask selection as a dynamic graph reasoning problem, where candidate lesions are modeled as nodes and edges capture spatial and semantic affinities. To ensure deployment feasibility, we introduce a Selective Asymmetric Fine-Tuning (SAFT) strategy that updates less than 1% of the parameters. Rigorous 5-fold cross-validation on the LIDC-IDRI and LNDb datasets demonstrates that our framework establishes a new state-of-the-art. Notably, it achieves an 81.5% Dice Similarity Coefficient (DSC) on LIDC-IDRI, outperforming leading LLM-based tools like LISA by over 5%. Crucially, our SAFT strategy acts as a powerful regularizer, yielding exceptional cross-fold stability (0.6% DSC variance) and paving the way for robust, context-aware clinical deployment.

cs.CV↗

arXiv2026

MALLES: A Multi-agent LLMs-based Economic Sandbox with Consumer Preference Alignment

In the real economy, modern decision-making is fundamentally challenged by high-dimensional, multimodal environments, which are further complicated by agent heterogeneity and combinatorial data sparsity. This paper introduces a Multi-Agent Large Language Model-based Economic Sandbox (MALLES), leveraging the inherent generalization capabilities of large-sacle models to establish a unified simulation framework applicable to cross-domain and cross-category scenarios. Central to our approach is a preference learning paradigm in which LLMs are economically aligned via post-training on extensive, heterogeneous transaction records across diverse product categories. This methodology enables the models to internalize and transfer latent consumer preference patterns, thereby mitigating the data sparsity issues prevalent in individual categories. To enhance simulation stability, we implement a mean-field mechanism designed to model the dynamic interactions between the product environment and customer populations, effectively stabilizing sampling processes within high-dimensional decision spaces. Furthermore, we propose a multi-agent discussion framework wherein specialized agents collaboratively process extensive product information. This architecture distributes cognitive load to alleviate single-agent attention bottlenecks and captures critical decision factors through structured dialogue. Experiments demonstrate that our framework achieves significant improvements in product selection accuracy, purchase quantity prediction, and simulation stability compared to existing economic and financial LLM simulation baselines. Our results substantiate the potential of large language models as a foundational pillar for high-fidelity, scalable decision simulation and latter analysis in the real economy based on foundational database.

cs.AI↗

arXiv2026

Suppression of Spectral Gap and Flat Bands on a Cuprate Superconductor Side-Surface

Side surfaces of cuprate superconductors are expected to display a suppressed $d$-wave order parameter and zero-energy topological flat bands with a large density of states, making them susceptible to symmetry broken orders. Yet such surfaces have never been investigated with momentum-resolved, surface-sensitive probes, because high-temperature superconductors rarely cleave along them. Using focused-ion-beam milling to define a controlled breaking point, we expose pristine (110) side surfaces of overdoped La$_{2-x}$Sr$_x$CuO$_4$ ($x=0.22$) suitable for angle-resolved photoemission. We observe the suppression of the superconducting spectral gap within our energy resolution ($\sim 4~\mathrm{meV}$), and surprisingly, the expected zero-energy flat band peak is also suppressed, despite the high topographic quality of the surface. Self-consistent Bogoliubov--de~Gennes calculations show that the measured geometric roughness of the cleaved surface is too weak to eliminate these modes. The calculations further demonstrate that bulk inhomogeneities characteristic of high-temperature superconductors, modelled as moderate Anderson-type disorder, can broaden the flat-band states beyond detectability. Our results provide the first momentum-resolved view of the electronic structure on a cuprate side surface and reveal disorder as the key factor currently preventing appearance of flat bands and their associated correlated orders.

cond-mat.supr-con↗

arXiv2026

Ultrasensitive strain modulation of terahertz magnons at a magnetic phase transition

Antiferromagnets typically host spin-wave (magnon) excitations in the terahertz (THz) regime, offering a promising platform for high-speed magnonic information technologies. Harnessing these excitations requires sensitive control of their spectral properties. Here we use resonant x-ray diffraction and Raman scattering to demonstrate uniaxial-strain control of the antiferromagnetic (AFM) ground state and THz magnon excitations in the layered Mott insulator Ca$_2$RuO$_4$. Although the states separated by the strain-induced phase transition differ only by the sign of the weak and partially frustrated interlayer interaction, their magnon energies differ by more than 10% (~ 0.3 THz). Our theoretical analysis explains this surprising observation by tracing the origin of both the sign reversal of the interlayer coupling and the magnon energy to the spin-orbital composition of the Ru valence electrons. The extreme strain sensitivity of the THz magnon energy near a magnetic phase transition opens up pathways towards a new generation of transition-edge magnonic devices.

cond-mat.mtrl-sci↗

arXiv2026

FOREVER: Forgetting Curve-Inspired Memory Replay for Language Model Continual Learning

Continual learning (CL) for large language models (LLMs) aims to enable sequential knowledge acquisition without catastrophic forgetting. Memory replay methods are widely used for their practicality and effectiveness, but most rely on fixed, step-based heuristics that often misalign with the model's actual learning progress, since identical training steps can result in varying degrees of parameter change. Motivated by recent findings that LLM forgetting mirrors the Ebbinghaus human forgetting curve, we propose FOREVER (FORgEtting curVe-inspired mEmory Replay), a novel CL framework that aligns replay schedules with a model-centric notion of time. FOREVER defines model time using the magnitude of optimizer updates, allowing forgetting curve-inspired replay intervals to align with the model's internal evolution rather than raw training steps. Building on this approach, FOREVER incorporates a forgetting curve-based replay scheduler to determine when to replay and an intensity-aware regularization mechanism to adaptively control how to replay. Extensive experiments on three CL benchmarks and models ranging from 0.6B to 13B parameters demonstrate that FOREVER consistently mitigates catastrophic forgetting.

cs.LG↗

arXiv2025

PRISM: A Personality-Driven Multi-Agent Framework for Social Media Simulation

Traditional agent-based models (ABMs) of opinion dynamics often fail to capture the psychological heterogeneity driving online polarization due to simplistic homogeneity assumptions. This limitation obscures the critical interplay between individual cognitive biases and information propagation, thereby hindering a mechanistic understanding of how ideological divides are amplified. To address this challenge, we introduce the Personality-Refracted Intelligent Simulation Model (PRISM), a hybrid framework coupling stochastic differential equations (SDE) for continuous emotional evolution with a personality-conditional partially observable Markov decision process (PC-POMDP) for discrete decision-making. In contrast to continuous trait approaches, PRISM assigns distinct Myers-Briggs Type Indicator (MBTI) based cognitive policies to multimodal large language model (MLLM) agents, initialized via data-driven priors from large-scale social media datasets. PRISM achieves superior personality consistency aligned with human ground truth, significantly outperforming standard homogeneous and Big Five benchmarks. This framework effectively replicates emergent phenomena such as rational suppression and affective resonance, offering a robust tool for analyzing complex social media ecosystems.

cs.CL↗

arXiv2025

A2Seek: Towards Reasoning-Centric Benchmark for Aerial Anomaly Understanding

While unmanned aerial vehicles (UAVs) offer wide-area, high-altitude coverage for anomaly detection, they face challenges such as dynamic viewpoints, scale variations, and complex scenes. Existing datasets and methods, mainly designed for fixed ground-level views, struggle to adapt to these conditions, leading to significant performance drops in drone-view scenarios. To bridge this gap, we introduce A2Seek (Aerial Anomaly Seek), a large-scale, reasoning-centric benchmark dataset for aerial anomaly understanding. This dataset covers various scenarios and environmental conditions, providing high-resolution real-world aerial videos with detailed annotations, including anomaly categories, frame-level timestamps, region-level bounding boxes, and natural language explanations for causal reasoning. Building on this dataset, we propose A2Seek-R1, a novel reasoning framework that generalizes R1-style strategies to aerial anomaly understanding, enabling a deeper understanding of "Where" anomalies occur and "Why" they happen in aerial frames. To this end, A2Seek-R1 first employs a graph-of-thought (GoT)-guided supervised fine-tuning approach to activate the model's latent reasoning capabilities on A2Seek. Then, we introduce Aerial Group Relative Policy Optimization (A-GRPO) to design rule-based reward functions tailored to aerial scenarios. Furthermore, we propose a novel "seeking" mechanism that simulates UAV flight behavior by directing the model's attention to informative regions. Extensive experiments demonstrate that A2Seek-R1 achieves up to a 22.04% improvement in AP for prediction accuracy and a 13.9% gain in mIoU for anomaly localization, exhibiting strong generalization across complex environments and out-of-distribution scenarios. Our dataset and code are released at https://2-mo.github.io/A2Seek/.

cs.CV↗

arXiv2025

Relief of EGFR/FOS-downregulated miR-103a by loganin alleviates NF-kappaB-triggered inflammation and gut barrier disruption in colitis

Due to the ever-rising global incidence rate of inflammatory bowel disease (IBD) and the lack of effective clinical treatment drugs, elucidating the detailed pathogenesis, seeking novel targets, and developing promising drugs are the top priority for IBD treatment. Here, we demonstrate that the levels of microRNA (miR)-103a were significantly downregulated in the inflamed mucosa of ulcerative colitis (UC) patients, along with elevated inflammatory cytokines (IL-1beta/TNF-alpha) and reduced tight junction protein (Occludin/ZO-1) levels, as compared with healthy control objects. Consistently, miR-103a deficient intestinal epithelial cells Caco-2 showed serious inflammatory responses and increased permeability, and DSS induced more severe colitis in miR-103a-/- mice than wild-type ones. Mechanistic studies unraveled that c-FOS suppressed miR-103a transcription via binding to its promoter, then miR-103a-targeted NF-kappaB activation contributes to inflammatory responses and barrier disruption by targeting TAB2 and TAK1. Notably, the traditional Chinese medicine Cornus officinalis (CO) and its core active ingredient loganin potently mitigated inflammation and barrier disruption in UC by specifically blocking the EGFR/RAS/ERK/c-FOS signaling axis, these effects mainly attributed to modulated miR-103a levels as the therapeutic activities of them were almost completely shielded in miR-103a KO mice. Taken together, this work reveals that loganin relieves EGFR/c-FOS axis-suppressed epithelial miR-103a expression, thereby inhibiting NF-kappaB pathway activation, suppressing inflammatory responses, and preserving tight junction integrity in UC. Thus, our data enrich mechanistic insights and promising targets for UC treatment.

q-bio.BM↗

arXiv2025

Robust Charge-Density Wave Correlations in Optimally-Doped YBa2Cu3Oy

Charge-density wave (CDW) order is a key property of high-Tc cuprates, but its boundaries in the phase diagram and potential connections to other phases remain controversial. We report nuclear magnetic resonance (NMR) measurements in the prototypical cuprate YBa2Cu3Oy demonstrating that short-range static CDW order remains robust at optimal doping (p=0.165), exhibiting a strength and temperature dependence in the normal state similar to those observed at p=0.11 in the underdoped regime. For an overdoped sample with p=0.184, we detect no static CDW down to T=Tc, though weak CDW order plausibly emerges below Tc. More broadly, we argue that both quenched disorder and competition with superconductivity influence the apparent boundary of the CDW phase, likely causing an underestimation of its intrinsic extent in doping. These findings challenge the view that the CDW phase boundary lies below p*=0.19, widely regarded as the critical doping where the pseudogap phase ends in YBa2Cu3Oy.

cond-mat.supr-con↗

arXiv2025

Bias and Volatility: A Statistical Framework for Evaluating Large Language Model's Stereotypes and the Associated Generation Inconsistency

We present a novel statistical framework for analyzing stereotypes in large language models (LLMs) by systematically estimating the bias and variation in their generation. Current alignment evaluation metrics often overlook stereotypes' randomness caused by LLMs' inconsistent generative behavior. For instance, LLMs may display contradictory stereotypes, such as those related to gender or race, for identical professions in different contexts. Ignoring this inconsistency risks misleading conclusions in alignment assessments and undermines efforts to evaluate the potential of LLMs to perpetuate or amplify social biases and unfairness. To address this, we propose the Bias-Volatility Framework (BVF), which estimates the probability distribution of stereotypes in LLM outputs. By capturing the variation in generative behavior, BVF assesses both the likelihood and degree to which LLM outputs negatively impact vulnerable groups, enabling a quantification of aggregated discrimination risk. Additionally, we introduce a mathematical framework to decompose this risk into bias risk (from the mean of the stereotype distribution) and volatility risk (from its variation). Applying BVF to 12 widely used LLMs, we find: i) Bias risk is the dominant contributor to discrimination; ii) Most LLMs exhibit substantial pro-male stereotypes across nearly all professions; iii) Reinforcement learning from human feedback reduces bias but increases volatility; iv) Discrimination risk correlates with socio-economic factors, such as professional salaries. Finally, we highlight BVF's broader applicability for assessing how generation inconsistencies in LLMs impact behavior beyond stereotypes.

cs.CL↗

arXiv2025

HandDiffuse: Generative Controllers for Two-Hand Interactions via Diffusion Models

Existing hands datasets are largely short-range and the interaction is weak due to the self-occlusion and self-similarity of hands, which can not yet fit the need for interacting hands motion generation. To rescue the data scarcity, we propose HandDiffuse12.5M, a novel dataset that consists of temporal sequences with strong two-hand interactions. HandDiffuse12.5M has the largest scale and richest interactions among the existing two-hand datasets. We further present a strong baseline method HandDiffuse for the controllable motion generation of interacting hands using various controllers. Specifically, we apply the diffusion model as the backbone and design two motion representations for different controllers. To reduce artifacts, we also propose Interaction Loss which explicitly quantifies the dynamic interaction process. Our HandDiffuse enables various applications with vivid two-hand interactions, i.e., motion in-betweening and trajectory control. Experiments show that our method outperforms the state-of-the-art techniques in motion generation and can also contribute to data augmentation for other datasets. Our dataset, corresponding codes, and pre-trained models will be disseminated to the community for future research towards two-hand interaction modeling.

cs.CV↗

arXiv2024

Beyond Euclidean: Dual-Space Representation Learning for Weakly Supervised Video Violence Detection

While numerous Video Violence Detection (VVD) methods have focused on representation learning in Euclidean space, they struggle to learn sufficiently discriminative features, leading to weaknesses in recognizing normal events that are visually similar to violent events (\emph{i.e.}, ambiguous violence). In contrast, hyperbolic representation learning, renowned for its ability to model hierarchical and complex relationships between events, has the potential to amplify the discrimination between visually similar events. Inspired by these, we develop a novel Dual-Space Representation Learning (DSRL) method for weakly supervised VVD to utilize the strength of both Euclidean and hyperbolic geometries, capturing the visual features of events while also exploring the intrinsic relations between events, thereby enhancing the discriminative capacity of the features. DSRL employs a novel information aggregation strategy to progressively learn event context in hyperbolic spaces, which selects aggregation nodes through layer-sensitive hyperbolic association degrees constrained by hyperbolic Dirichlet energy. Furthermore, DSRL attempts to break the cyber-balkanization of different spaces, utilizing cross-space attention to facilitate information interactions between Euclidean and hyperbolic space to capture better discriminative features for final violence detection. Comprehensive experiments demonstrate the effectiveness of our proposed DSRL.

cs.CV↗

arXiv2024

An End-to-End Reinforcement Learning Based Approach for Micro-View Order-Dispatching in Ride-Hailing

Assigning orders to drivers under localized spatiotemporal context (micro-view order-dispatching) is a major task in Didi, as it influences ride-hailing service experience. Existing industrial solutions mainly follow a two-stage pattern that incorporate heuristic or learning-based algorithms with naive combinatorial methods, tackling the uncertainty of both sides' behaviors, including emerging timings, spatial relationships, and travel duration, etc. In this paper, we propose a one-stage end-to-end reinforcement learning based order-dispatching approach that solves behavior prediction and combinatorial optimization uniformly in a sequential decision-making manner. Specifically, we employ a two-layer Markov Decision Process framework to model this problem, and present \underline{D}eep \underline{D}ouble \underline{S}calable \underline{N}etwork (D2SN), an encoder-decoder structure network to generate order-driver assignments directly and stop assignments accordingly. Besides, by leveraging contextual dynamics, our approach can adapt to the behavioral patterns for better performance. Extensive experiments on Didi's real-world benchmarks justify that the proposed approach significantly outperforms competitive baselines in optimizing matching efficiency and user experience tasks. In addition, we evaluate the deployment outline and discuss the gains and experiences obtained during the deployment tests from the view of large-scale engineering implementation.

cs.LG↗

arXiv2024

Magnetic field expulsion in optically driven YBa$_2$Cu$_3$O$_{6.48}$

Coherent optical driving in quantum solids is emerging as a new research frontier, with many demonstrations of exotic non-equilibrium quantum phases. These are based on engineered band structures, and on stimulated nonlinear interactions between driven modes. Enhanced functionalities like ferroelectricity, magnetism and superconductivity have been reported in these non-equilibrium settings. In high-Tc cuprates, coherent driving of certain phonon modes induces a transient state with superconducting-like optical properties, observed far above T$_c$ and throughout the pseudogap phase. Questions remain not only on the microscopic nature of this phenomenon, but also on the macroscopic properties of these transient states, beyond the documented optical conductivities. Crucially, it is not clear if driven cuprates exhibit Meissner-like diamagnetism. Here, the time-dependent magnetic-field amplitude surrounding a driven YBa$_2$Cu$_3$O$_{6.48}$ sample is probed by measuring Faraday rotation in a GaP layer adjacent to the superconductor. For the same driving conditions that result in superconducting-like optical properties, an enhancement of magnetic field at the edge of the sample is detected, indicative of induced diamagnetism. The dynamical field expulsion measured after pumping is comparable in size to the one expected in an equilibrium type II superconductor of similar shape and size with a volume susceptibility $χ_v$ of order -0.3. Crucially, this value is incompatible with a photo-induced increase in mobility without superconductivity. Rather, it underscores the notion of a pseudogap phase in which incipient superconducting correlations are enhanced or synchronized by the optical drive.

cond-mat.supr-con↗

arXiv2024

Using strain to uncover the interplay between two- and three-dimensional charge density waves in high-temperature superconducting YBa$_{2}$Cu$_{3}$O$_{\rm{y}}$

Uniaxial pressure provides an efficient approach to control the competition between charge density waves (CDWs) and superconductivity in underdoped YBa$_{2}$Cu$_{3}$O$_{\rm{y}}$. It can enhance the correlation volume of ubiquitous short-range 2D CDW correlations, and induces a long-range 3D CDW otherwise only accessible at large magnetic fields. Here, we use x-ray diffraction to study the strain and doping evolution of these CDWs. No signatures of discommensurations nor pair density waves are observed in the investigated strain-temperature parameter space, but direct evidence for a form of competition between 2D and 3D CDWs is uncovered. We show that the interplay between the 3D CDW, the 2D CDWs and superconductivity is qualitatively well described by including strain effects in simulations of a nonlinear sigma model of competing superconducting and CDW orders. From a broader perspective, our results underscore the potential of strain tuning as a powerful tool for probing and manipulating competing orders in quantum materials.

cond-mat.supr-con↗

arXiv2023

Valuing Post-Revenue Biopharmaceutical Assets with Pfizer's Current Portfolio as a Case Study

This research paper addresses the critical challenge of accurately valuing post-revenue drug assets in the biotechnology and pharmaceutical sectors, a key factor influencing a wide range of strategic operations and investment decisions. Recognizing the importance of reliable valuations for stakeholders such as pharmaceutical companies, venture capitalists, and private equity firms, this study introduces a novel model for forecasting future sales of post-revenue biopharmaceutical assets. The proposed model leverages historical sales data, a resource known for its high quality and availability in company financial records, to produce distributional estimates of cumulative sales for individual assets. These estimates are instrumental in calculating the Net Present Value of each asset, thereby facilitating more informed and strategic investment decisions. A practical application of this model is demonstrated through its implementation in analyzing Pfizer's portfolio of post-revenue assets. This precision highlights the model's potential as a valuable tool in the financial assessment and decision-making processes within the biotech and pharmaceutical industries, offering a methodical approach to identifying investment opportunities and optimizing capital allocation.

q-fin.PR↗

arXiv2023

Tuna: Instruction Tuning using Feedback from Large Language Models

Instruction tuning of open-source large language models (LLMs) like LLaMA, using direct outputs from more powerful LLMs such as Instruct-GPT and GPT-4, has proven to be a cost-effective way to align model behaviors with human preferences. However, the instruction-tuned model has only seen one response per instruction, lacking the knowledge of potentially better responses. In this paper, we propose finetuning an instruction-tuned LLM using our novel \textit{probabilistic ranking} and \textit{contextual ranking} approaches to increase the likelihood of generating better responses. Probabilistic ranking enables the instruction-tuned model to inherit the relative rankings of high-quality and low-quality responses from the teacher LLM. On the other hand, learning with contextual ranking allows the model to refine its own response distribution using the contextual understanding ability of stronger LLMs. Furthermore, we apply probabilistic ranking and contextual ranking sequentially to the instruction-tuned LLM. The resulting model, which we call \textbf{Tuna}, consistently improves the performance on Super Natural Instructions (119 test tasks), LMentry (25 test tasks), Vicuna QA, and can even obtain better results than several strong reinforcement learning baselines. Our code and data are available at \url{ https://github.com/microsoft/LMOps}.

cs.CL↗

arXiv2023

Does Debiasing Inevitably Degrade the Model Performance

Gender bias in language models has attracted sufficient attention because it threatens social justice. However, most of the current debiasing methods degraded the model's performance on other tasks while the degradation mechanism is still mysterious. We propose a theoretical framework explaining the three candidate mechanisms of the language model's gender bias. We use our theoretical framework to explain why the current debiasing methods cause performance degradation. We also discover a pathway through which debiasing will not degrade the model performance. We further develop a causality-detection fine-tuning approach to correct gender bias. The numerical experiment demonstrates that our method is able to lead to double dividends: partially mitigating gender bias while avoiding performance degradation.

cs.CL↗