arXiv subjects

Yadong Zhang

Publications and source records attributed to Yadong Zhang.

At least 19 recordsLinked to original sources

arXiv2026

Towards Robust Argumentative Essay Understanding via TIDE: An Interactive Framework with Trial and Debate

Argumentative essays serve as a vital medium for assessing critical thinking and reasoning skills, yet there is limited works on accurately understanding and evaluating such texts via prompt. In this work, we propose TIDE, a novel framework designed to improve criteria-based prompt optimization for argument-related tasks by integrating TrIal and DEbate mechanism. Our method addresses key limitations of criteria-based prompt optimizing by mitigating the influence of noisy training data and enhancing optimization stability. We evaluate TIDE on three core tasks: Automated Essay Scoring, Argument Component Detection, and Argument Relation Identification. Results demonstrate that our framework improves performance across tasks. These findings underscore the potential of combining prompt-based methods for advanced argument understanding.

cs.AI↗

arXiv2026

Retrieve, Integrate, and Synthesize: Spatial-Semantic Grounded Latent Visual Reasoning

Multimodal Large Language Models (MLLMs) have made remarkable progress on vision-language reasoning, yet most methods still compress visual evidence into discrete textual thoughts, creating an information bottleneck for fine-grained perception. Recent latent visual reasoning methods attempt to reason in continuous hidden states, but we find that they suffer from insufficient manifold compatibility: latent trajectories drift away from pretrained reasoning circuits, collapse into instance-agnostic patterns, and are often bypassed during answer generation. To address these issues, we propose RIS (Retrieve, Integrate, and Synthesize), a spatial-semantic grounded framework that develops latent reasoning as a compatible extension of pretrained MLLM computation. We first construct a step-wise grounded reasoning dataset with bounding boxes and region-specific semantic descriptions. Built on this supervision, RIS anchors latent tokens to both spatial and semantic evidence, enforces their causal role through a progressive attention bottleneck, and introduces short language transition tokens to bridge synthesized latent states back to vocabulary-aligned decoding. Experiments on V*, HRBench4K, HRBench8K, MMVP, and BLINK show consistent improvements over closed/open-source and latent reasoning baselines. Further analyses demonstrate that RIS learns diverse, interpretable, and progressively integrated latent trajectories, offering a practical path toward faithful internal visual reasoning in MLLMs.

cs.CL↗

arXiv2025

H2EAL: Hybrid-Bonding Architecture with Hybrid Sparse Attention for Efficient Long-Context LLM Inference

Large language models (LLMs) have demonstrated remarkable proficiency in a wide range of natural language processing applications. However, the high energy and latency overhead induced by the KV cache limits the edge deployment, especially for long contexts. Emerging hybrid bonding (HB) technology has been proposed as a promising alternative to conventional near-memory processing (NMP) architectures, offering improved bandwidth efficiency and lower power consumption while exhibiting characteristics of distributed memory. In this paper, we propose H2EAL, a hybrid bonding-based accelerator with sparse attention algorithm-hardware co-design for efficient LLM inference at the edge. At the algorithm level, we propose a hybrid sparse attention scheme with static and dynamic sparsity for different heads to fully leverage the sparsity with high accuracy. At the hardware level, we co-design the hardware to support hybrid sparse attention and propose memory-compute co-placement to address the distributed memory bottleneck. Since different attention heads exhibit different sparse patterns and the attention structure often mismatches the HB architecture, we further develop a load-balancing scheduler with parallel tiled attention to address workload imbalance and optimize the mapping strategy. Extensive experiments demonstrate H2EAL achieves 5.20~48.21x speedup and 6.22~73.48x energy efficiency improvement over baseline HB implementation, with a negligible average accuracy drop of 0.87% on multiple benchmarks.

cs.PF↗

arXiv2025

KernelDNA: Dynamic Kernel Sharing via Decoupled Naive Adapters

Dynamic convolution enhances model capacity by adaptively combining multiple kernels, yet faces critical trade-offs: prior works either (1) incur significant parameter overhead by scaling kernel numbers linearly, (2) compromise inference speed through complex kernel interactions, or (3) struggle to jointly optimize dynamic attention and static kernels. We observe that pre-trained Convolutional Neural Networks (CNNs) exhibit inter-layer redundancy akin to that in Large Language Models (LLMs). Specifically, dense convolutional layers can be efficiently replaced by derived "child" layers generated from a shared "parent" convolutional kernel through an adapter. To address these limitations and implement the weight-sharing mechanism, we propose a lightweight convolution kernel plug-in, named KernelDNA. It decouples kernel adaptation into input-dependent dynamic routing and pre-trained static modulation, ensuring both parameter efficiency and hardware-friendly inference. Unlike existing dynamic convolutions that expand parameters via multi-kernel ensembles, our method leverages cross-layer weight sharing and adapter-based modulation, enabling dynamic kernel specialization without altering the standard convolution structure. This design preserves the native computational efficiency of standard convolutions while enhancing representation power through input-adaptive kernel adjustments. Experiments on image classification and dense prediction tasks demonstrate that KernelDNA achieves a state-of-the-art accuracy-efficiency balance among dynamic convolution variants.

cs.CV↗

arXiv2025

SelecTKD: Selective Token-Weighted Knowledge Distillation for LLMs

Knowledge distillation (KD) is a standard route to compress Large Language Models (LLMs) into compact students, yet most pipelines uniformly apply token-wise loss regardless of teacher confidence. This indiscriminate supervision amplifies noisy, high-entropy signals and is especially harmful under large teacher-student capacity gaps. We introduce SelecTKD, a plug-and-play Selective Token-Weighted distillation framework that shifts the focus from "how to measure divergence" to "where to apply learning". At each step, the student proposes tokens that are verified by the teacher through a robust propose-and-verify procedure with two variants: greedy Top-k and non-greedy Spec-k. Accepted tokens receive full loss, while rejected tokens are masked or down-weighted. This objective-agnostic design works with on- and off-policy data, induces an implicit curriculum quantified by Token Acceptance Rate (TAR), and stabilizes optimization. Across instruction following, mathematical reasoning, code generation, and a VLM setting, SelecTKD consistently improves strong baselines and achieves state-of-the-art results for small models without architectural changes or extra reference models.

cs.CL↗

arXiv2025

Deposition-Dependent Coverage and Performance of Phosphonic Acid Interface Modifiers in Halide Perovskite Optoelectronics

In this work, we study the effect of various deposition methods for phosphonic acid interface modifiers commonly pursued as self-assembled monolayers in high-performance metal halide perovskite photovoltaics and light-emitting diodes. We compare the deposition of (2-(3,6-diiodo-9H-carbazol-9-yl)ethyl)phosphonic acid onto indium tin oxide (ITO) bottom contacts by varying three parameters: the method of deposition, specifically spin coating or prolonged dip coating, ITO surface treatment via HCl/FeCl3 etching, and use in combination with a second modifier, 1,6-hexylenediphosphonic acid. We demonstrate that varying these modification protocols can impact time-resolved photoluminescence carrier lifetimes and quasi-Fermi level splitting of perovskite films deposited onto the phosphonic-acid-modified ITO. Ultraviolet photoelectron spectroscopy shows an increase in effective work function after phosphonic acid modification and clear evidence for photoemission from carbazole functional groups at the ITO surface. We use X-ray photoelectron spectroscopy to probe differences in phosphonic acid coverage on the metal oxide contact and show that perovskite samples grown on ITO with the highest phosphonic acid coverage exhibit the longest carrier lifetimes. Finally, we establish that device performance follows these same trends. These results indicate that the reactivity, heterogeneity, and composition of the bottom contact help to control recombination rates and therefore power conversion efficiencies. ITO etching, prolonged deposition times for phosphonic acids via dip coating, and the use of a secondary, more hydrophilic bis-phosphonic acid, all contribute to improvements in surface coverage, carrier lifetime, and device efficiency. These improvements each have a positive impact, and we achieve the best results when all three strategies are implemented.

physics.app-ph↗

arXiv2025

DeepKD: A Deeply Decoupled and Denoised Knowledge Distillation Trainer

Recent advances in knowledge distillation have emphasized the importance of decoupling different knowledge components. While existing methods utilize momentum mechanisms to separate task-oriented and distillation gradients, they overlook the inherent conflict between target-class and non-target-class knowledge flows. Furthermore, low-confidence dark knowledge in non-target classes introduces noisy signals that hinder effective knowledge transfer. To address these limitations, we propose DeepKD, a novel training framework that integrates dual-level decoupling with adaptive denoising. First, through theoretical analysis of gradient signal-to-noise ratio (GSNR) characteristics in task-oriented and non-task-oriented knowledge distillation, we design independent momentum updaters for each component to prevent mutual interference. We observe that the optimal momentum coefficients for task-oriented gradient (TOG), target-class gradient (TCG), and non-target-class gradient (NCG) should be positively related to their GSNR. Second, we introduce a dynamic top-k mask (DTM) mechanism that gradually increases K from a small initial value to incorporate more non-target classes as training progresses, following curriculum learning principles. The DTM jointly filters low-confidence logits from both teacher and student models, effectively purifying dark knowledge during early training. Extensive experiments on CIFAR-100, ImageNet, and MS-COCO demonstrate DeepKD's effectiveness. Our code is available at https://github.com/haiduo/DeepKD.

cs.CV↗

arXiv2025

Balancing Efficiency and Effectiveness: An LLM-Infused Approach for Optimized CTR Prediction

Click-Through Rate (CTR) prediction is essential in online advertising, where semantic information plays a pivotal role in shaping user decisions and enhancing CTR effectiveness. Capturing and modeling deep semantic information, such as a user's preference for "Häagen-Dazs' HEAVEN strawberry light ice cream" due to its health-conscious and premium attributes, is challenging. Traditional semantic modeling often overlooks these intricate details at the user and item levels. To bridge this gap, we introduce a novel approach that models deep semantic information end-to-end, leveraging the comprehensive world knowledge capabilities of Large Language Models (LLMs). Our proposed LLM-infused CTR prediction framework(Multi-level Deep Semantic Information Infused CTR model via Distillation, MSD) is designed to uncover deep semantic insights by utilizing LLMs to extract and distill critical information into a smaller, more efficient model, enabling seamless end-to-end training and inference. Importantly, our framework is carefully designed to balance efficiency and effectiveness, ensuring that the model not only achieves high performance but also operates with optimal resource utilization. Online A/B tests conducted on the Meituan sponsored-search system demonstrate that our method significantly outperforms baseline models in terms of Cost Per Mile (CPM) and CTR, validating its effectiveness, scalability, and balanced approach in real-world applications.

cs.LG↗

arXiv2024

EGEAN: An Exposure-Guided Embedding Alignment Network for Post-Click Conversion Estimation

Accurate post-click conversion rate (CVR) estimation is crucial for online advertising systems. Despite significant advances in causal approaches designed to address the Sample Selection Bias problem, CVR estimation still faces challenges due to Covariate Shift. Given the intrinsic connection between the distribution of covariates in the click and non-click spaces, this study proposes an Exposure-Guided Embedding Alignment Network (EGEAN) to address estimation bias caused by covariate shift. Additionally, we propose a Parameter Varying Doubly Robust Estimator with steady-state control to handle small propensities better. Online A/B tests conducted on the Meituan advertising system demonstrate that our method significantly outperforms baseline models with respect to CVR and GMV, validating its effectiveness. Code is available: https://github.com/hydrogen-maker/EGEAN.

cs.LG↗

arXiv2024

Mind's Eye of LLMs: Visualization-of-Thought Elicits Spatial Reasoning in Large Language Models

Large language models (LLMs) have exhibited impressive performance in language comprehension and various reasoning tasks. However, their abilities in spatial reasoning, a crucial aspect of human cognition, remain relatively unexplored. Human possess a remarkable ability to create mental images of unseen objects and actions through a process known as the Mind's Eye, enabling the imagination of the unseen world. Inspired by this cognitive capacity, we propose Visualization-of-Thought (VoT) prompting. VoT aims to elicit spatial reasoning of LLMs by visualizing their reasoning traces, thereby guiding subsequent reasoning steps. We employed VoT for multi-hop spatial reasoning tasks, including natural language navigation, visual navigation, and visual tiling in 2D grid worlds. Experimental results demonstrated that VoT significantly enhances the spatial reasoning abilities of LLMs. Notably, VoT outperformed existing multimodal large language models (MLLMs) in these tasks. While VoT works surprisingly well on LLMs, the ability to generate mental images to facilitate spatial reasoning resembles the mind's eye process, suggesting its potential viability in MLLMs. Please find the dataset and codes at https://microsoft.github.io/visualization-of-thought

cs.CL↗

arXiv2024

K-Level Reasoning: Establishing Higher Order Beliefs in Large Language Models for Strategic Reasoning

Strategic reasoning is a complex yet essential capability for intelligent agents. It requires Large Language Model (LLM) agents to adapt their strategies dynamically in multi-agent environments. Unlike static reasoning tasks, success in these contexts depends on anticipating other agents' beliefs and actions while continuously adjusting strategies to achieve individual goals. LLMs and LLM agents often struggle with strategic reasoning due to the absence of a reasoning framework that enables them to dynamically infer others' perspectives and adapt to changing environments. Inspired by the Level-K framework from game theory and behavioral economics, which extends reasoning from simple reactions to structured strategic depth, we propose a novel framework: "K-Level Reasoning with Large Language Models (K-R)." This framework employs recursive mechanisms to enable LLMs to achieve varying levels of strategic depth, allowing agents to form higher order beliefs - beliefs about others' beliefs. We validate this framework through rigorous testing on four testbeds: two classical game theory problems and two social intelligence tasks. The results demonstrate the advantages of K-R in strategic reasoning. Our work presents the first recursive implementation of strategic depth in large language models (LLMs). It establishes a foundation for future research into theory of mind and strategic reasoning in LLMs.

cs.CL↗

arXiv2024

Increased Brightness and Reduced Efficiency Droop in Perovskite Quantum Dot Light-Emitting Diodes using Carbazole-Based Phosphonic Acid Interface Modifiers

We demonstrate the use of [2-($\textit{9H}$-carbazol-9-yl)ethyl]phosphonic acid (2PACz) and [2-(3,6-di-$\textit{tert}$-butyl-$\textit{9H}$-carbazol-9-yl)ethyl]phosphonic acid (t-Bu-2PACz) as anode modification layers in metal-halide perovskite quantum dot light-emitting diodes (QLEDs). Compared to conventional QLED structures with PEDOT:PSS (poly(3,4-ethylenedioxythiophene) polystyrene sulfonate)/PVK (poly(9-vinylcarbazole)) hole-transport layers, QLEDs made with phosphonic acid (PA)-modified indium tin oxide (ITO) anodes show an over 7-fold increase in brightness, achieving a brightness of 373,000 cd m$^{-2}$, one of the highest brightnesses reported to date for colloidal perovskite QLEDs. Importantly, the onset of efficiency roll-off, or efficiency droop, occurs at ~1000-fold higher current density for QLEDs made with PA-modified anodes compared to control QLEDs made with conventional PEDOT:PSS/PVK hole transport layers, allowing the devices to sustain significantly higher levels of external quantum efficiency at a brightness of >10$^{5}$ cd m$^{-2}$. Steady-state and time-resolved photoluminescence measurements indicate these improvements are due to a combination of multiple factors, including reducing quenching of photoluminescence at the PEDOT:PSS interface and reducing photoluminescence efficiency loss at high levels of current density.

cond-mat.mtrl-sci↗

arXiv2024

Enhancing Language Model Rationality with Bi-Directional Deliberation Reasoning

This paper introduces BI-Directional DEliberation Reasoning (BIDDER), a novel reasoning approach to enhance the decision rationality of language models. Traditional reasoning methods typically rely on historical information and employ uni-directional (left-to-right) reasoning strategy. This lack of bi-directional deliberation reasoning results in limited awareness of potential future outcomes and insufficient integration of historical context, leading to suboptimal decisions. BIDDER addresses this gap by incorporating principles of rational decision-making, specifically managing uncertainty and predicting expected utility. Our approach involves three key processes: Inferring hidden states to represent uncertain information in the decision-making process from historical data; Using these hidden states to predict future potential states and potential outcomes; Integrating historical information (past contexts) and long-term outcomes (future contexts) to inform reasoning. By leveraging bi-directional reasoning, BIDDER ensures thorough exploration of both past and future contexts, leading to more informed and rational decisions. We tested BIDDER's effectiveness in two well-defined scenarios: Poker (Limit Texas Hold'em) and Negotiation. Our experiments demonstrate that BIDDER significantly improves the decision-making capabilities of LLMs and LLM agents.

cs.CL↗

arXiv2024

Graph neural networks for power grid operational risk assessment under evolving grid topology

This article investigates the ability of graph neural networks (GNNs) to identify risky conditions in a power grid over the subsequent few hours, without explicit, high-resolution information regarding future generator on/off status (grid topology) or power dispatch decisions. The GNNs are trained using supervised learning, to predict the power grid's aggregated bus-level (either zonal or system-level) or individual branch-level state under different power supply and demand conditions. The variability of the stochastic grid variables (wind/solar generation and load demand), and their statistical correlations, are rigorously considered while generating the inputs for the training data. The outputs in the training data, obtained by solving numerous mixed-integer linear programming (MILP) optimal power flow problems, correspond to system-level, zonal and transmission line-level quantities of interest (QoIs). The QoIs predicted by the GNNs are used to conduct hours-ahead, sampling-based reliability and risk assessment w.r.t. zonal and system-level (load shedding) as well as branch-level (overloading) failure events. The proposed methodology is demonstrated for three synthetic grids with sizes ranging from 118 to 2848 buses. Our results demonstrate that GNNs are capable of providing fast and accurate prediction of QoIs and can be good proxies for computationally expensive MILP algorithms. The excellent accuracy of GNN-based reliability and risk assessment suggests that GNN models can substantially improve situational awareness by quickly providing rigorous reliability and risk estimates.

eess.SY↗

arXiv2024

Operational risk quantification of power grids using graph neural network surrogates of the DC OPF

A DC OPF surrogate modeling framework is developed for Monte Carlo (MC) sampling-based risk quantification in power grid operation. MC simulation necessitates solving a large number of DC OPF problems corresponding to the samples of stochastic grid variables (power demand and renewable generation), which is computationally prohibitive. Computationally inexpensive surrogates of OPF provide an attractive alternative for expedited MC simulation. Graph neural network (GNN) surrogates of DC OPF, which are especially suitable to graph-structured data, are employed in this work. Previously developed DC OPF surrogate models have focused on accurate operational decision-making and not on risk quantification. Here, risk quantification-specific aspects of DC OPF surrogate evaluation is the main focus. To this end, the proposed GNN surrogates are evaluated using realistic joint probability distributions, quantification of their risk estimation accuracy, and investigation of their generalizability. Four synthetic grids (Case118, Case300, Case1354pegase, and Case2848rte) are used for surrogate model performance evaluation. It is shown that the GNN surrogates are sufficiently accurate for predicting the (bus-level, branch-level and system-level) grid state and enable fast as well as accurate operational risk quantification for power grids. The article thus develops tools for fast reliability and risk quantification in real-world power grids using GNN-based surrogates.

eess.SY↗

arXiv2024

LLM as a Mastermind: A Survey of Strategic Reasoning with Large Language Models

This paper presents a comprehensive survey of the current status and opportunities for Large Language Models (LLMs) in strategic reasoning, a sophisticated form of reasoning that necessitates understanding and predicting adversary actions in multi-agent settings while adjusting strategies accordingly. Strategic reasoning is distinguished by its focus on the dynamic and uncertain nature of interactions among multi-agents, where comprehending the environment and anticipating the behavior of others is crucial. We explore the scopes, applications, methodologies, and evaluation metrics related to strategic reasoning with LLMs, highlighting the burgeoning development in this area and the interdisciplinary approaches enhancing their decision-making performance. It aims to systematize and clarify the scattered literature on this subject, providing a systematic review that underscores the importance of strategic reasoning as a critical cognitive capability and offers insights into future research directions and potential improvements.

cs.CL↗

arXiv2024

Surface doping of rubrene single crystals by molecular electron donors and acceptors

The surface molecular doping of organic semiconductors can play an important role in the development of organic electronic or optoelectronic devices. Single-crystal rubrene remains a leading molecular candidate for applications in electronics due to its high hole mobility. In parallel, intensive research into the fabrication of flexible organic electronics requires the careful design of functional interfaces to enable optimal device characteristics. To this end, the present work seeks to understand the effect of surface molecular doping on the electronic band structure of rubrene single crystals. Our angle-resolved photoemission measurements reveal that the Fermi level moves in the band gap of rubrene depending on the direction of surface electron-transfer reactions with the molecular dopants, yet the valence band dispersion remains essentially unperturbed. This indicates that surface electron-transfer doping of a molecular single crystal can effectively modify the near-surface charge density, while retaining good charge-carrier mobility.

cond-mat.mtrl-sci↗

arXiv2023

Power grid operational risk assessment using graph neural network surrogates

We investigate the utility of graph neural networks (GNNs) as proxies of power grid operational decision-making algorithms (optimal power flow (OPF) and security-constrained unit commitment (SCUC)) to enable rigorous quantification of the operational risk. To conduct principled risk analysis, numerous Monte Carlo (MC) samples are drawn from the (foretasted) probability distributions of spatio-temporally correlated stochastic grid variables. The corresponding OPF and SCUC solutions, which are needed to quantify the risk, are generated using traditional OPF and SCUC solvers to generate data for training GNN model(s). The GNN model performance is evaluated in terms of the accuracy of predicting quantities of interests (QoIs) derived from the decision variables in OPF and SCUC. Specifically, we focus on thermal power generation and load shedding at system and individual zone level. We also perform reliability and risk quantification based on GNN predictions and compare with that obtained from OPF/SCUC solutions. Our results demonstrate that GNNs are capable of providing fast and accurate prediction of QoIs and thus can be good surrogate models for OPF and SCUC. The excellent accuracy of GNN-based reliability and risk assessment further suggests that GNN surrogate has the potential to be applied in real-time and hours-ahead risk quantification.

cs.LG↗