arXiv subjects

Yifei Xu

Publications and source records attributed to Yifei Xu.

At least 19 recordsLinked to original sources

arXiv2026

Read Between the Stickers: Sentiment-Prior Reasoning with Learnable Verbalized Rules for Multimodal Chat Analysis

Multimodal chat analysis of social media stickers (MCAS) benefits from jointly modeling text and sticker semantics, yet it is inherently challenged by the interference between sentiment and intent recognition. Although existing multi-task approaches achieve competitive performance, they largely ignore this inter-task interference and offer little explicit reasoning about how these two predictions are made. To address this issue, we propose \textbf{ExCoVer}, an \textbf{Ex}plicit \textbf{C}hain-\textbf{o}f-Thought framework with \textbf{Ver}balized rules learning that integrates sentiment-prior reasoning with learnable discrimination rules to produce explicit reasoning chains for sentiment and intent predictions. Specifically, ExCoVer consists of two components: (1) Sentiment-Prior Chain-of-Thought (SP-CoT), which detects cross-modal sentiment conflicts and uses the dominant sentiment as a prior to mitigate inter-task interference and narrow the candidate intent space; and (2) Verbalized Rules Learning for Confusing Intent Discrimination (VRLCID), which treats discrimination rules as learnable parameters and optimizes them via learner, optimizer, and regularizer agents to suppress spurious correlations and distinguish confusing intents. Extensive experiments on CSMSA and MSAIRS datasets demonstrate that ExCoVer achieves state-of-the-art performance while providing explicit reasoning chains.

cs.HC↗

arXiv2026

Nonlocal correlations of fermionic entanglement in the spacetime of Einstein-Gauss-Bonnet black hole

The investigation of nonclassical correlations in curved spacetimes offers key insights into the intersection of quantum information theory and gravitational physics. This paper studies two nonlocal correlation measures, non local advantage of quantum coherence (NAQC) and Bell nonlocality (BN) in a $d$-dimensional spherically symmetric Einstein-Gauss-Bonnet (EGB) black hole spacetime. We consider two observers (Alice and Rob) initially sharing a maximally entangled Bell state: Alice freely falls into the black hole (inertial Kruskal frame), while Rob accelerates outside the horizon (non-inertial Schwarzschild-like frame). The Unruh-Hawking effect modifies Rob's field modes, requiring Bogoliubov transformations to relate the two frames. We derive the mixed bipartite density matrix for fermionic fields and analytical expressions for NAQC and BN, which depend on Hawking temperature (itself governed by $α$, $d$, and $r_h$). Our results show both correlations degrade monotonically with increasing Hawking temperature, confirm the NAQC-BN hierarchical relationship persists in EGB spacetime, and highlighting the impact of high curvature corrections on quantum resources.

gr-qc↗

arXiv2026

Orchestration for Domain-specific Edge-Cloud Language Models

The remarkable performance of Large Language Models (LLMs) has inspired many applications, which often necessitate edge-cloud collaboration due to connectivity, privacy, and cost considerations. Traditional methods primarily focus on selecting the best LLM model for optimizing performance, while neglecting the critical interplay between the components of the LLM serving pipeline (context retrieval, query preprocessing, etc.) or the changing latency and cost constraints. We introduce ECO-LLM (Edge-Cloud Orchestrator for LLMs), a novel system that reframes this problem as a joint optimization challenge and solves it by systematically exploring component configurations and dynamically selecting optimal strategies at the query level. ECO-LLM consists of two components: (1) the ECO-LLM Emulator, which efficiently explores the vast configuration space utilizing query clustering and pareto-optimal path selection, gathering domain-specific performance metrics without exhaustive evaluation; and (2) the ECO-LLM Runtime, which leverages these metrics to dynamically select optimal resolution strategies for user queries while meeting user-defined Service Level Objectives (SLOs). We evaluate ECO-LLM on a smart home and a smart car assistant scenarios. With an exhaustive exploration of all possible configurations for seen queries, ECO-LLM outperforms cloud-based models like GPT-4o in terms of accuracy (90% vs. 74% on average) while reducing costs by 90% and latency by 55%, demonstrating the value of its joint optimization at the query level. In practical deployment for previously unseen queries, ECO-LLM selects configurations that reduce costs by 62% or improve response times by 62% on average compared to state-of-the-art model routing approaches, while maintaining higher accuracy and consistently adhering to specified latency and cost constraints.

cs.DB↗

arXiv2026

Trustworthy AI/ML Regression and Unbiased Causal Inference for Real-World Data

Real-World Data (RWD), with its large sample sizes and rich clinical detail, offers a compelling alternative to randomized controlled trials (RCTs) for studying treatment effects in diverse and complex patient populations. However, its observational nature introduces confounding that prevents straightforward comparative effectiveness research. Target trial emulation leverages RWD to estimate average treatment effects (ATE) at the population scale and diversity that RCTs cannot achieve, yet its validity depends critically on unbiased ATE estimation under high-dimensional confounding. Many causal inference pipelines address high-dimensional confounding through machine learning and artificial intelligence (ML/AI) outcome regression. However, commonly used ML/AI regression models exhibit systematic prediction bias, with predicted outcomes shrinking toward the marginal outcome mean. This structural bias propagates into ATE estimation and cannot be corrected by cross-fitting, ensemble methods, or any standard ML practice. In this work, we first quantitatively characterize how systematic prediction bias in ML/AI outcome regression leads to biased ATE estimates in causal inference models. We further propose an unbiased ML/AI regression-based causal inference framework to ensure unbiased ATE estimation for observational studies. We demonstrate our approach by studying the effects of opioids on cardiovascular health in patients with chronic pain using UK Biobank data.

stat.AP↗

arXiv2026

Direct Reasoning Optimization: Token-Level Reasoning Reflectivity Meets Rubric Gates for Unverifiable Tasks

Reinforcement learning (RL) training of large language models (LLMs) on unverifiable tasks is challenging even when a reasonable-quality reference answer is available. We propose a constrained RL training framework that (i) optimizes a token-level dense Reasoning Reflection Reward (R3) aligned with reasoning quality, and (ii) enforces rubric-gating as feasibility constraints at the rollout group level. R3 measures the model's token-level certainty of a reference answer under its chain-of-thought (CoT) prefix, and selectively emphasizes tokens with high cross-rollout variance, which we call reasoning-reflective tokens, that would otherwise be diluted by the bulk of low-variance tokens. The same variance signal also drives a filter that discards queries with insufficient signal for comparative learning. Rubric-gating complements R3 by operationalizing principled task criteria as hard accept/reject checks on final answers. Empirically, across four datasets spanning scientific writing, medicine, legal contracts, and finance, our framework outperforms strong baselines, achieves faster, more sample-efficient learning, and respects feasibility constraints.

cs.CL↗

arXiv2026

Diagnosing Capability Gaps in Fine-Tuning Data

Fine-tuning large language models (LLMs) for domain-specific tasks requires training datasets that comprehensively cover the target capabilities a practitioner needs. Yet identifying which capabilities a dataset fails to support, and doing so before an expensive fine-tuning run, remains a largely unsolved problem. We introduce GoalCover, a framework that helps practitioners systematically detect capability gaps in fine-tuning datasets through interactive goal decomposition and automated coverage assessment. GoalCover guides a practitioner through structured decomposition of a high-level goal into atomic, independently evaluable subgoals; assigns each training sample an LLM-based alignment score against every subgoal; and surfaces missing capabilities through automated analysis of low-scoring sample explanations. We validate the framework along two complementary axes. First, through controlled corruption experiments across three domains (medical QA, legal summarization, code generation), we show that GoalCover reliably distinguishes targeted from non-targeted capability impacts: target subgoals degrade by 25.6% on average versus 2.1% for non-target subgoals (Cohen's d=1.24). Second, we demonstrate downstream utility on a financial-summarization Reinforcement Fine-Tuning (RFT) task with Qwen-3-14B: training on GoalCover-filtered data improves the LLM-judge reward from 3.77 to 4.12 (out of 5) over the unfiltered baseline, and combining filtered data with goal-conditioned synthetic samples yields the strongest result (4.20). The two results together show that GoalCover works as a practical pre-fine-tuning diagnostic: it detects capability gaps and produces concrete signal for closing them.

cs.LG↗

arXiv2026

GPU-Native Multi-Area State Estimation via SIMD Abstraction and Boundary Condensation

Power system state estimation (SE) is foundational for grid monitoring, yet conventional centralized solvers face increasing computational pressure as the system scale and real-time requirements grow. This paper presents a GPU-native framework for hierarchical multi-area state estimation (MASE) that addresses these bottlenecks through a single-instruction, multiple-data (SIMD) abstraction and sparse Schur local condensation. We partition the network into areas, evaluate measurement residuals and derivatives using fixed-sparsity templates, and directly assemble local normal-equation blocks through a fused GPU accumulation kernel without materializing explicit Jacobians. Each area is then factorized on the GPU in Schur mode to export a dense local boundary block and condensed right-hand side, after which a reduced global boundary system is assembled and solved on device. This design preserves device residency across measurement evaluation, local condensation, and boundary coordination while exposing parallelism across areas. Numerical experiments on partitioned PEGASE 2869-bus, PEGASE 9241-bus, and ACTIVSg10k benchmark systems demonstrate that the proposed approach effectively leverages GPU throughput by maintaining full device residency and high arithmetic intensity.

eess.SY↗

arXiv2026

AnyPro: Preference-Preserving Anycast Optimization based on Strategic AS-Path Prepending

Operating large-scale anycast networks is challenging because client-to-site mappings often misalign with operator's expectation due to opaque inter-domain routing. We present AnyPro, the first system to unlock the full potential of AS-path prepending (ASPP), efficiently deriving globally optimal configurations to steer clients toward performance-optimal sites at scale. AnyPro first employs an efficient polling mechanism to identify all clients sensitive to ASPP. By analyzing the routing changes during the process, the system derives a set of ASPP constraints that guide client traffic toward the desired sites. We then formulate the anycast optimization problem as a constraint-based program and compute optimal ASPP configurations. Extensive evaluation on a global testbed with 20 PoPs demonstrates the effectiveness of AnyPro: it reduces the 90th percentile latency by 37.7% compared to baseline configurations without ASPP. Furthermore, we show that AnyPro can be integrated with PoP-level anycast optimization techniques to achieve additional performance gains.

cs.NI↗

arXiv2026

SibylSense: Adaptive Rubric Learning via Memory Tuning and Adversarial Probing

Designing aligned and robust rewards for open-ended generation remains a key barrier to RL post-training. Rubrics provide structured, interpretable supervision, but scaling rubric construction is difficult: expert rubrics are costly, prompted rubrics are often superficial or inconsistent, and fixed-pool discriminative rubrics can saturate and drift, enabling reward hacking. We present SibylSense, an inference-time learning approach that adapts a frozen rubric generator through a tunable memory bank of validated rubric items. Memory is updated via verifier-based item rewards measured by reference-candidate answer discriminative gaps from a handful of examples. SibylSense alternates memory tuning with a rubric-adversarial policy update that produces rubric-satisfying candidate answers, shrinking discriminative gaps and driving the rubric generator to capture new quality dimensions. Experiments on two open-ended tasks show that SibylSense yields more discriminative rubrics and improves downstream RL performance over static and non-adaptive baselines.

cs.CL↗

arXiv2026

Geometric Algorithms for Neural Combinatorial Optimization with Constraints

Self-Supervised Learning (SSL) for Combinatorial Optimization (CO) is an emerging paradigm for solving combinatorial problems using neural networks. In this paper, we address a central challenge of SSL for CO: solving problems with discrete constraints. We design an end-to-end differentiable framework that enables us to solve discrete constrained optimization problems with neural networks. Concretely, we leverage algorithmic techniques from the literature on convex geometry and Carathéodory's theorem to decompose neural network outputs into convex combinations of polytope corners that correspond to feasible sets. This decomposition-based approach enables self-supervised training but also ensures efficient quality-preserving rounding of the neural net output into feasible solutions. Extensive experiments in cardinality-constrained optimization show that our approach can consistently outperform neural baselines. We further provide worked-out examples of how our method can be applied beyond cardinality-constrained problems to a diverse set of combinatorial optimization tasks, including finding independent sets in graphs, and solving matroid-constrained problems.

cs.LG↗

arXiv2025

VADTree: Explainable Training-Free Video Anomaly Detection via Hierarchical Granularity-Aware Tree

Video anomaly detection (VAD) focuses on identifying anomalies in videos. Supervised methods demand substantial in-domain training data and fail to deliver clear explanations for anomalies. In contrast, training-free methods leverage the knowledge reserves and language interactivity of large pre-trained models to detect anomalies. However, the current fixed-length temporal window sampling approaches struggle to accurately capture anomalies with varying temporal spans. Therefore, we propose VADTree that utilizes a Hierarchical Granularityaware Tree (HGTree) structure for flexible sampling in VAD. VADTree leverages the knowledge embedded in a pre-trained Generic Event Boundary Detection (GEBD) model to characterize potential anomaly event boundaries. Specifically, VADTree decomposes the video into generic event nodes based on boundary confidence, and performs adaptive coarse-fine hierarchical structuring and redundancy removal to construct the HGTree. Then, the multi-dimensional priors are injected into the visual language models (VLMs) to enhance the node-wise anomaly perception, and anomaly reasoning for generic event nodes is achieved via large language models (LLMs). Finally, an inter-cluster node correlation method is used to integrate the multi-granularity anomaly scores. Extensive experiments on three challenging datasets demonstrate that VADTree achieves state-of-the-art performance in training-free settings while drastically reducing the number of sampled video segments. The code will be available at https://github.com/wenlongli10/VADTree.

cs.CV↗

arXiv2025

LocalBench: Benchmarking LLMs on County-Level Local Knowledge and Reasoning

Large language models (LLMs) have been widely evaluated on macro-scale geographic tasks, such as global factual recall, event summarization, and regional reasoning. Yet, their ability to handle hyper-local knowledge remains poorly understood. This gap is increasingly consequential as real-world applications, from civic platforms to community journalism, demand AI systems that can reason about neighborhood-specific dynamics, cultural narratives, and local governance. Existing benchmarks fall short in capturing this complexity, often relying on coarse-grained data or isolated references. We present LocalBench, the first benchmark designed to systematically evaluate LLMs on county-level local knowledge across the United States. Grounded in the Localness Conceptual Framework, LocalBench includes 14,782 validated question-answer pairs across 526 U.S. counties in 49 states, integrating diverse sources such as Census statistics, local subreddit discourse, and regional news. It spans physical, cognitive, and relational dimensions of locality. Using LocalBench, we evaluate 13 state-of-the-art LLMs under both closed-book and web-augmented settings. Our findings reveal critical limitations: even the best-performing models reach only 56.8% accuracy on narrative-style questions and perform below 15.5% on numerical reasoning. Moreover, larger model size and web augmentation do not guarantee better performance, for example, search improves Gemini's accuracy by +13.6%, but reduces GPT-series performance by -11.4%. These results underscore the urgent need for language models that can support equitable, place-aware AI systems: capable of engaging with the diverse, fine-grained realities of local communities across geographic and cultural contexts.

cs.CL↗

arXiv2025

DeepSpecs: Expert-Level Questions Answering in 5G

5G technology enables mobile Internet access for billions of users. Answering expert-level questions about 5G specifications requires navigating thousands of pages of cross-referenced standards that evolve across releases. Existing retrieval-augmented generation (RAG) frameworks, including telecom-specific approaches, rely on semantic similarity and cannot reliably resolve cross-references or reason about specification evolution. We present DeepSpecs, a RAG system enhanced by structural and temporal reasoning via three metadata-rich databases: SpecDB (clause-aligned specification text), ChangeDB (line-level version diffs), and TDocDB (standardization meeting documents). DeepSpecs explicitly resolves cross-references by recursively retrieving referenced clauses through metadata lookup, and traces specification evolution by mining changes and linking them to Change Requests that document design rationale. We curate two 5G QA datasets: 573 expert-annotated real-world questions from practitioner forums and educational resources, and 350 evolution-focused questions derived from approved Change Requests. Across multiple LLM backends, DeepSpecs outperforms base models and state-of-the-art telecom RAG systems; ablations confirm that explicit cross-reference resolution and evolution-aware retrieval substantially improve answer quality, underscoring the value of modeling the structural and temporal properties of 5G standards.

cs.CL↗

arXiv2025

Forecasting the Buzz: Enriching Hashtag Popularity Prediction with LLM Reasoning

Hashtag trends ignite campaigns, shift public opinion, and steer millions of dollars in advertising spend, yet forecasting which tag goes viral is elusive. Classical regressors digest surface features but ignore context, while large language models (LLMs) excel at contextual reasoning but misestimate numbers. We present BuzzProphet, a reasoning-augmented hashtag popularity prediction framework that (1) instructs an LLM to articulate a hashtag's topical virality, audience reach, and timing advantage; (2) utilizes these popularity-oriented rationales to enrich the input features; and (3) regresses on these inputs. To facilitate evaluation, we release HashView, a 7,532-hashtag benchmark curated from social media. Across diverse regressor-LLM combinations, BuzzProphet reduces RMSE by up to 2.8% and boosts correlation by 30% over baselines, while producing human-readable rationales. Results demonstrate that using LLMs as context reasoners rather than numeric predictors injects domain insight into tabular models, yielding an interpretable and deployable solution for social media trend forecasting.

cs.SI↗

arXiv2025

RLTHF: Targeted Human Feedback for LLM Alignment

Fine-tuning large language models (LLMs) to align with user preferences is challenging due to the high cost of quality human annotations in Reinforcement Learning from Human Feedback (RLHF) and the generalizability limitations of AI Feedback. To address these challenges, we propose RLTHF, a human-AI hybrid framework that combines LLM-based initial alignment with selective human annotations to achieve full-human annotation alignment with minimal effort. RLTHF identifies hard-to-annotate samples mislabeled by LLMs using a reward model's reward distribution and iteratively enhances alignment by integrating strategic human corrections while leveraging LLM's correctly labeled samples. Evaluations on HH-RLHF and TL;DR datasets show that RLTHF reaches full-human annotation-level alignment with only 6-7% of the human annotation effort. Furthermore, models trained on RLTHF's curated datasets for downstream tasks outperform those trained on fully human-annotated datasets, underscoring the effectiveness of RLTHF.

cs.CL↗

arXiv2025

Beamforming for Movable and Rotatable Antenna Enabled Multi-User Communications

In the development of wireless communication technology, multiple-input multiple-output (MIMO) technology has emerged as a key enabler, significantly enhancing the capacity of communication systems. However, traditional MIMO systems, which rely on fixed-position antennas (FPAs) with spacing limitations, cannot fully exploit the channel variations in the continuous spatial domain, thus limiting the system's spatial multiplexing performance and diversity. To address these limitations, movable antennas (MAs) have been introduced, offering a breakthrough in signal processing and spatial multiplexing by overcoming the constraints of FPA-based systems. Furthermore, this paper extends the functionality of MAs by introducing movable rotatable antennas (MRAs), which enhance the system's ability to optimize performance in the spatial domain by adding rotational degrees of freedom. By incorporating a dynamic precoding framework based on both antenna position and rotation angle optimization, and employing the zero-forcing (ZF) precoding method, this paper proposes an efficient optimization approach aimed at improving signal quality, mitigating interference, and solving the non-linear, constrained optimization problem using the sequential quadratic programming (SQP) algorithm. This approach effectively enhances the communication system's performance.

eess.SP↗

arXiv2025

CoRNStack: High-Quality Contrastive Data for Better Code Retrieval and Reranking

Effective code retrieval plays a crucial role in advancing code generation, bug fixing, and software maintenance, particularly as software systems increase in complexity. While current code embedding models have demonstrated promise in retrieving code snippets for small-scale, well-defined tasks, they often underperform in more demanding real-world applications such as bug localization within GitHub repositories. We hypothesize that a key issue is their reliance on noisy and inconsistent datasets for training, which impedes their ability to generalize to more complex retrieval scenarios. To address these limitations, we introduce CoRNStack, a large-scale, high-quality contrastive training dataset for code that spans multiple programming languages. This dataset is curated using consistency filtering to eliminate noisy positives and is further enriched with mined hard negatives, thereby facilitating more effective learning. We demonstrate that contrastive training of embedding models using CoRNStack leads to state-of-the-art performance across a variety of code retrieval tasks. Furthermore, the dataset can be leveraged for training code reranking models, a largely underexplored area compared to text reranking. Our finetuned code reranking model significantly improves the ranking quality over the retrieved results. Finally, by employing our code retriever and reranker together, we demonstrate significant improvements in function localization for GitHub issues, an important component of real-world software development.

cs.CL↗

arXiv2024

System States Forecasting of Microservices with Dynamic Spatio-Temporal Data

In the AIOps (Artificial Intelligence for IT Operations) era, accurately forecasting system states is crucial. In microservices systems, this task encounters the challenge of dynamic and complex spatio-temporal relationships among microservice instances, primarily due to dynamic deployments, diverse call paths, and cascading effects among instances. Current time-series forecasting methods, which focus mainly on intrinsic patterns, are insufficient in environments where spatial relationships are critical. Similarly, spatio-temporal graph approaches often neglect the nature of temporal trend, concentrating mostly on message passing between nodes. Moreover, current research in microservices domain frequently underestimates the importance of network metrics and topological structures in capturing the evolving dynamics of systems. This paper introduces STMformer, a model tailored for forecasting system states in microservices environments, capable of handling multi-node and multivariate time series. Our method leverages dynamic network connection data and topological information to assist in modeling the intricate spatio-temporal relationships within the system. Additionally, we integrate the PatchCrossAttention module to compute the impact of cascading effects globally. We have developed a dataset based on a microservices system and conducted comprehensive experiments with STMformer against leading methods. In both short-term and long-term forecasting tasks, our model consistently achieved a 8.6% reduction in MAE(Mean Absolute Error) and a 2.2% reduction in MSE (Mean Squared Error). The source code is available at https://github.com/xuyifeiiie/STMformer.

cs.NI↗