Searcharxiv⌕ Search

arXiv subjects

Jianfeng Zhan

Publications and source records attributed to Jianfeng Zhan.

At least 19 recordsLinked to original sources

MSAR: Next-Scale Autoregressive Forecasting for Time Series via Modular Multi-Scale Decoupling

Time series forecasting underpins critical applications in finance, energy, healthcare, and transportation. Although deep models have achieved strong results, most adopt single-scale modeling or restrict multiscale processing to the input side, causing a misalignment between multiscale inputs and single-scale outputs and limiting predictive power. We introduce the Modular Scale-wise Autoregressive Framework (MSAR), a model-agnostic design that forecasts progressively across multiple temporal resolutions. MSAR offers three advantages: (1) scale-wise aligned modeling, which disentangles heterogeneous temporal patterns by aligning inputs and outputs at each scale; (2) scale-wise autoregression, where coarse-scale predictions guide finer-scale forecasting through hierarchical information flow; and (3) a modular architecture, enabling seamless integration with diverse backbones such as CNNs, MLPs, and Transformers. Extensive experiments across a broad set of datasets and forecasting models demonstrate that MSAR achieves consistent improvements in both accuracy and inference efficiency, validating the effectiveness of scale-aligned autoregression for multiscale time series forecasting. The code is available at: https://github.com/Oryxo/MSAR.

cs.LG↗

FlexCast: Flexible Context Learning for Time Series Foundation Models

Time series foundation models (TSFMs) based on autoregressive (AR) next-token prediction entangle two factors: variable-context conditioning and prediction-unit factorization with autoregressive rollout. Strong AR performance therefore does not reveal whether generalization stems from the flexible training distribution, the local prediction target, the left-to-right inference process, or their combination. We propose FlexCast, a flexible context-target learning framework that disentangles these factors. FlexCast reformulates forecasting as variable-length masked prediction over time series patches: during pretraining, it dynamically samples visible context and masked target patches, exposing the model to diverse context-target configurations without sequential decoding. At inference time, the same pretrained model supports one-step forecasting, confidence-based iterative refinement, and next-token-style generation through a unified mask-and-predict interface, enabling controlled comparison of inference schedules under identical architecture, data, and objective. Evaluated on BLAST and multiple zero-shot benchmarks, FlexCast achieves accuracy competitive with or better than strong TSFM baselines while substantially reducing inference latency. Ablations confirm that context diversity and contiguous target masking drive performance, while left-to-right decoding does not yield consistent gains once flexible context-target relationships have been learned, suggesting that the key benefit of AR-style TSFMs lies in flexible training rather than sequential inference. The code is available at https://github.com/BenchCouncil/FlexCast.

cs.AI↗

VecDBLens: An Evaluatology-Based Framework for Vector Database Diagnosis

Vector databases are increasingly deployed as retrieval pipelines in which representation, indexing, and query-time execution jointly determine accuracy, throughput, and cost. Existing vector database evaluations primarily focus on indexing while treating embeddings as fixed inputs and ignoring the role of query-time execution, making it difficult to decide whether a bottleneck should be addressed by changing the embedding model, the index, or the query-processing functions. To address the above limitation, we introduce VecDBLens, an Evaluatology-based framework that treats the function under diagnosis as an evaluated object and all remaining dataset, workload, parameter, and runtime factors as matched evaluation conditions. VecDBLens attributes performance differences to individual evaluated objects through comparisons under matched evaluation conditions, summarizing their performance and stability across multiple datasets and workloads. It further analyzes interactions among objects using analysis of variance (ANOVA), thereby turning vector database benchmarking from leaderboard ranking into a reproducible diagnostic procedure. VecDBLens reveals two key insights. First, retrieval performance is fundamentally upper-bounded by embedding quality, revealing that many apparent indexing failures are therefore representation-side limitations. Second, stronger index parameterization is not always beneficial: overly aggressive configurations can reshape neighborhood topology and cause recall regression. These results challenge the common practice of evaluating vector databases primarily through indexing and show that real-world limitations arise from constraints across retrieval functions rather than any single function alone. VecDBLens provides a principled basis for identifying the true performance drivers of vector databases and for guiding their design in practical applications.

cs.DB↗

Top-Down Microarchitecture Analysis Without Causal Modelling Is Dangerous!

Top-Down Microarchitecture Analysis (TMA) is the standard for identifying CPU bottlenecks. However, TMA implicitly models performance via a single causal path: Workload(dataset, algorithm) => CPU(CPI) => System(Performance) (microarchitectural efficiency path). It overlooks two hidden causal paths: Workload(dataset, algorithm) => Instructions(Retired) => System(Performance) (algorithmic efficiency path) and Workload(dataset, algorithm) => Instructions(Retired) => CPU(CPI) => System(Performance) (microarchitecture-algorithm efficiency path). This causal blindness often misleads architects into futile hardware optimizations. To address this, we propose Causal Top-Down Microarchitecture Analysis (CTMA). Utilizing the object causal model, CTMA explicitly models these three causal paths among the workload, instructions, CPU, and system. We validate CTMA across Intel, AMD, and ARM platforms. Our experiments reveal counterintuitive insights: (1) CPI Profiling Blindness: Algorithmic instruction bloat can double elapsed time while the CPI profile remains frozen. (2) Over-Speculation Congestion: Throttling the Frontend paradoxically improves performance in pointer-chasing algorithms. (3) ILP Window Collapse: Extreme data topologies shift bottlenecks from Backend ROB capacity to Frontend fill-rate starvation. (4) Op-Cache SMT Overlap: Ultra-small footprints yield SMT speedups on Intel but cause severe thrashing on AMD. (5) "Ghost State" Artifact: Stale pipeline mitigations on Intel Xeon Gold 5120T artificially boost performance by 22.75% on SPEC CPU2017 and 42.2% on SPEC CPU2026. Based on this, we propose a dynamic speculation throttling mechanism.

cs.PF↗

SmartANN: Object Causal Modeling Boosts Approximate Nearest Neighbor Diagnosis and Auto-Design

Approximate Nearest Neighbor (ANN) algorithms achieve high efficiency through interdependent phases across index construction and query execution. This coupling allows upstream performance loss to propagate downstream, affecting execution behavior and measurable outputs. Existing component-level analyses mainly compare isolated design choices, while end-to-end benchmarks report aggregate metrics; neither traces loss propagation across dependent phases, hindering root-cause attribution and automated redesign. We present SmartANN, a framework based on the object causal model (OCM) for ANN bottleneck attribution and automated redesign. SmartANN represents an ANN workflow as eight ordered, replaceable objects and diagnoses them with a sequential diagnose-and-replace loop. At each iteration, it identifies the first object deviating from expected behavior or output as a bottleneck. Because an upstream bottleneck can obscure downstream ones, SmartANN replaces it with a test oracle when available, or with an implementation producing a better outcome, then continues downstream diagnosis. From the diagnosed bottlenecks and failure causes, SmartANN composes compatible actions from a pluggable action library to generate an optimized end-to-end ANN design. We instantiate SmartANN for IVF-PQ and HNSW, covering partition-and-quantization and graph-based ANN families. Experiments on eight real-world datasets show that SmartANN improves Recall by 0.24--74.20%, and increases QPS by 28.8--256.5% at comparable Recall, with low diagnosis and auto-design overhead. The code is available at https://github.com/zhouyutong20/SmartANN.

cs.DB↗

When AI Designs AI: Innovation or Imitation?

Recent advances in LLM agents have made them increasingly capable of designing methods for complex AI tasks. This raises two central questions about agent-designed methods relative to human-designed methods: how well they perform, and how different their algorithmic designs are. To study these questions, this paper introduces an analysis that derives task-specific algorithmic design spaces from human-designed methods, maps both human- and agent-designed methods into these spaces, and quantifies their algorithmic differences at the module level. Widely used LLM agents are evaluated on a suite of representative, open-ended AI tasks spanning multiple modalities, and the methods they design are analyzed in terms of both task performance and algorithmic differences from human-designed methods. Experimental results show that current agents can occasionally match or surpass human state-of-the-art (SOTA) performance (10/72 configurations), but such success does not generalize reliably across tasks or agents. Moreover, 96.8% of agent-designed methods fall within human-derived algorithmic design spaces, largely recombining algorithmic choices found in human-designed methods, while nearly half exactly match an existing human algorithmic design. Taken together, these findings suggest that although current agents can occasionally match or surpass human SOTA performance, their algorithmic designs remain within human-derived algorithmic design spaces, reflecting the reuse and recombination of algorithmic choices.

cs.AI↗

Inference of Component Effect on System Performance

In a computer system, multiple components--such as the CPU, memory, and others--work together as a system whose performance can be directly measured. However, the effect of a component under investigation (CUI), e.g., CPU, on system performance cannot be directly measured and can only be inferred. Accurately inferring CUI effect on system performance is a critical issue. Our experiments reveal that the general-purpose rigorous methodologies, like Design of Experiments (DoE), Randomized Controlled Trials (RCTs), and a single-purpose empirical methodology, like SPEC CPU2017, can not address this issue effectively and efficiently. We propose a rigorous methodology to address this issue: First, we identify a self-contained system (SCS) under the context of which we can completely understand how CUI and other essential components affect the system performance, and then we use a structural causal model methodology to represent and infer the causal effect of CUI on the system performance. We utilize this methodology and verify its correctness in the context of CPU design and evaluation. Through theoretical analysis and pioneering controlled experiments, we systematically compare our methodology against three established methodologies: SPEC CPU2017, DoE, and RCTs. The results show that our methodology can achieve its goal effectively and efficiently, whereas others exhibit inherent limitations.

cs.PF↗

BigPower: Hierarchical Source-Level Module Power Estimation for CPUs with Large Language Models

Accurate power estimation is important for understanding and optimizing CPU power behavior, yet practical workflows often rely on simulation-derived information or post-silicon analysis. In this work, we present BigPower, a hierarchical source-level surrogate model for fine-grained module-level power estimation during CPU design. BigPower leverages large language model-based representations together with architectural hierarchy, module connectivity, configuration parameters, and workload context to estimate module-level power consumption directly from source-level design information, without requiring additional simulation during inference. Experimental results in the open-source XiangShan processor family demonstrate practical fine-grained power estimation across diverse configurations and workloads, offering an efficient alternative to conventional simulation-based workflows.

cs.AR↗

CombinationTS: A Modular Framework for Understanding Time-Series Forecasting Models

Recent progress in time-series forecasting has led to rapidly increasing architectural complexity, yet many reported State-of-the-Art gains are statistically fragile or misattributed. We argue that progress requires a shift from model selection to modular attribution, identifying which components truly drive performance. We propose CombinationTS, a self-contained probabilistic evaluation framework that decomposes forecasting models into orthogonal modules--Input Transformation, Embedding, Encoder, Decoder, and Output Transformation--and evaluates them under a shared evaluation condition space. By quantifying each component via marginalized performance ($μ$) and stability ($σ$), CombinationTS enables robust attribution beyond fragile point estimates. Through large-scale paired evaluation, we uncover the Identity Paradox: once the data view (Embedding) is well-designed, a parameter-free Identity Encoder often matches or outperforms complex backbones. We further show that explicit structural priors introduced via Input Transformations yield a more favorable performance-stability trade-off than increasing Encoder complexity, establishing a principled baseline for architectural necessity.

cs.LG↗

GraDE: A Graph Diffusion Estimator for Frequent Subgraph Discovery in Neural Architectures

Finding frequently occurring subgraph patterns or network motifs in neural architectures is crucial for optimizing efficiency, accelerating design, and uncovering structural insights. However, as the subgraph size increases, enumeration-based methods are perfectly accurate but computationally prohibitive, while sampling-based methods are computationally tractable but suffer from a severe decline in discovery capability. To address these challenges, this paper proposes GraDE, a diffusion-guided search framework that ensures both computational feasibility and discovery capability. The key innovation is the Graph Diffusion Estimator (GraDE), which is the first to introduce graph diffusion models to identify frequent subgraphs by scoring their typicality within the learned distribution. Comprehensive experiments demonstrate that the estimator achieves superior ranking accuracy, with up to 114\% improvement compared to sampling-based baselines. Benefiting from this, the proposed framework successfully discovers large-scale frequent patterns, achieving up to 30$\times$ higher median frequency than sampling-based methods.

cs.LG↗

Probing Memes in LLMs: A Paradigm for the Entangled Evaluation World

Current evaluation paradigms for large language models (LLMs) characterize models and datasets separately, yielding coarse descriptions: items in datasets are treated as pre-labeled entries, and models are summarized by overall scores such as accuracy, together ignoring the diversity of population-level model behaviors across items with varying properties. To address this gap, this paper conceptualizes LLMs as composed of memes, a notion introduced by Dawkins as cultural genes that replicate knowledge and behavior. Building on this perspective, the Probing Memes paradigm reconceptualizes evaluation as an entangled world of models and data. It centers on a Perception Matrix that captures model-item interactions, enabling Probe Properties for characterizing items and Meme Scores for depicting model behavioral traits. Applied to 9 datasets and 4,507 LLMs, Probing Memes reveals hidden capability structures and quantifies phenomena invisible under traditional paradigms (e.g., elite models failing on problems that most models answer easily). It not only supports more informative and extensible benchmarks but also enables population-based evaluation of LLMs.

cs.CL↗

TimeMosaic: Temporal Heterogeneity Guided Time Series Forecasting via Adaptive Granularity Patch and Segment-wise Decoding

Multivariate time series forecasting is essential in domains such as finance, transportation, climate, and energy. However, existing patch-based methods typically adopt fixed-length segmentation, overlooking the heterogeneity of local temporal dynamics and the decoding heterogeneity of forecasting. Such designs lose details in information-dense regions, introduce redundancy in stable segments, and fail to capture the distinct complexities of short-term and long-term horizons. We propose TimeMosaic, a forecasting framework that aims to address temporal heterogeneity. TimeMosaic employs adaptive patch embedding to dynamically adjust granularity according to local information density, balancing motif reuse with structural clarity while preserving temporal continuity. In addition, it introduces segment-wise decoding that treats each prediction horizon as a related subtask and adapts to horizon-specific difficulty and information requirements, rather than applying a single uniform decoder. Extensive evaluations on benchmark datasets demonstrate that TimeMosaic delivers consistent improvements over existing methods, and our model trained on the large-scale corpus with 321 billion observations achieves performance competitive with state-of-the-art TSFMs.

cs.LG↗

On Meta-Evaluation

Evaluation is the foundation of empirical science, yet the evaluation of evaluation itself -- so-called meta-evaluation -- remains strikingly underdeveloped. While methods such as observational studies, design of experiments (DoE), and randomized controlled trials (RCTs) have shaped modern scientific practice, there has been little systematic inquiry into their comparative validity and utility across domains. Here we introduce a formal framework for meta-evaluation by defining the evaluation space, its structured representation, and a benchmark we call AxiaBench. AxiaBench enables the first large-scale, quantitative comparison of ten widely used evaluation methods across eight representative application domains. Our analysis reveals a fundamental limitation: no existing method simultaneously achieves accuracy and efficiency across diverse scenarios, with DoE and observational designs in particular showing significant deviations from real-world ground truth. We further evaluate a unified method of entire-space stratified sampling from previous evaluatology research, and the results report that it consistently outperforms prior approaches across all tested domains. These results establish meta-evaluation as a scientific object in its own right and provide both a conceptual foundation and a pragmatic tool set for advancing trustworthy evaluation in computational and experimental research.

stat.ME↗

KAIROS: Unified Training for Universal Non-Autoregressive Time Series Forecasting

In the World Wide Web, reliable time series forecasts provide the forward-looking signals that drive resource planning, cache placement, and anomaly response, enabling platforms to operate efficiently as user behavior and content distributions evolve. Compared with other domains, time series forecasting for Web applications requires much faster responsiveness to support real-time decision making. We present KAIROS, a non-autoregressive time series forecasting framework that directly models segment-level multi-peak distributions. Unlike autoregressive approaches, KAIROS avoids error accumulation and achieves just-in-time inference, while improving over existing non-autoregressive models that collapse to over-smoothed predictions. Trained on the large-scale corpus, KAIROS demonstrates strong zero-shot generalization on six widely used benchmarks, delivering forecasting performance comparable to state-of-the-art foundation models with similar scale, at a fraction of their inference cost. Beyond empirical results, KAIROS highlights the importance of non-autoregressive design as a scalable paradigm for foundation models in time series.

cs.LG↗

DualSG: A Dual-Stream Explicit Semantic-Guided Multivariate Time Series Forecasting Framework

Multivariate Time Series Forecasting plays a key role in many applications. Recent works have explored using Large Language Models for MTSF to take advantage of their reasoning abilities. However, many methods treat LLMs as end-to-end forecasters, which often leads to a loss of numerical precision and forces LLMs to handle patterns beyond their intended design. Alternatively, methods that attempt to align textual and time series modalities within latent space frequently encounter alignment difficulty. In this paper, we propose to treat LLMs not as standalone forecasters, but as semantic guidance modules within a dual-stream framework. We propose DualSG, a dual-stream framework that provides explicit semantic guidance, where LLMs act as Semantic Guides to refine rather than replace traditional predictions. As part of DualSG, we introduce Time Series Caption, an explicit prompt format that summarizes trend patterns in natural language and provides interpretable context for LLMs, rather than relying on implicit alignment between text and time series in the latent space. We also design a caption-guided fusion module that explicitly models inter-variable relationships while reducing noise and computation. Experiments on real-world datasets from diverse domains show that DualSG consistently outperforms 15 state-of-the-art baselines, demonstrating the value of explicitly combining numerical forecasting with semantic guidance.

cs.AI↗

Achieving Consistent and Comparable CPU Evaluation

The challenge of CPU evaluation lies in the fact that user-perceived performance metrics can only be measured on an independently running system consisting of the CPU and other indispensable components, and hence it is difficult to accurately attribute the deviations in the evaluation outcomes to the differences between the CPUs. Our experiments reveal that the industry-standard CPU benchmark, SPEC CPU2017, suffers from a significant flaw: for the identical CPU, undefined configurations of other indispensable components introduce uncontrolled variability in evaluation outcomes. We propose a rigorous CPU evaluation methodology. Through theoretical analysis and pioneering controlled experiments, we systematically compare our methodology against four established methodologies: the SPEC CPU 2017, two DOE variants, and one RCTs approach. The results show our methodology can achieve consistent and comparable evaluation outcomes, while others exhibit inherent limations.

cs.PF↗

Quantifying the Dynamics of Harm Caused by Retracted Research

Despite enormous efforts devoted to understand the characteristics and impacts of retracted papers, little is known about the mechanisms underlying the dynamics of their harm and the dynamics of its propagation. Here, we propose a citation-based framework to quantify the harm caused by retracted papers, aiming to uncover why their harm persists and spreads so widely. We uncover an ''attention escape'' mechanism, wherein retracted papers postpone significant harm, more prominently affect indirectly citing papers, and inflict greater harm on citations in journals with an impact factor less than 10. This mechanism allows retracted papers to inflict harm outside the attention of authors and publishers, thereby evading their intervention. This study deepens understanding of the harm caused by retracted papers, emphasizes the need to activate and enhance the attention of authors and publishers, and offers new insights and a foundation for strategies to mitigate their harm and prevent its spread.

cs.DL↗

Could Bibliometrics Reveal Top Science and Technology Achievements and Researchers? The Case for Evaluatology-based Science and Technology Evaluation

By utilizing statistical methods to analyze bibliographic data, bibliometrics faces inherent limitations in identifying the most significant science and technology achievements and researchers. To overcome this challenge, we present an evaluatology-based science and technology evaluation methodology. At the heart of this approach lies the concept of an extended evaluation condition, encompassing eight crucial components derived from a field. We define four relationships that illustrate the connections among various achievements based on their mapped extended EC components, as well as their temporal and citation links. Within a relationship under an extended evaluation condition, evaluators can effectively compare these achievements by carefully addressing the influence of confounding variables. We establish a real-world evaluation system encompassing an entire collection of achievements, each of which is mapped to several components of an extended EC. Within a specific field like chip technology or open source, we construct a perfect evaluation model that can accurately trace the evolution and development of all achievements in terms of four relationships based on the real-world evaluation system. Building upon the foundation of the perfect evaluation model, we put forth four-round rules to eliminate non-significant achievements by utilizing four relationships. This process allows us to establish a pragmatic evaluation model that effectively captures the essential achievements, serving as a curated collection of the top N achievements within a specific field during a specific timeframe. We present a case study on the top 100 Chip achievements which highlights its practical application and efficacy in identifying significant achievements and researchers that otherwise can not be identified by using bibliometrics.

cs.CE↗