arXiv subjects

Sheng Di

Publications and source records attributed to Sheng Di.

At least 37 records · Page 2Linked to original sources

arXiv2026

Retrieval-of-Thought: Efficient Reasoning via Reusing Thoughts

Large reasoning models improve accuracy by producing long reasoning traces, but this inflates latency and cost, motivating inference-time efficiency. We propose Retrieval-of-Thought (RoT), which reuses prior reasoning as composable ``thought" steps to guide new problems. RoT organizes steps into a thought graph with sequential and semantic edges to enable fast retrieval and flexible recombination. At inference, RoT retrieves query-relevant nodes and applies reward-guided traversal to assemble a problem-specific template that guides generation. This dynamic template reuse reduces redundant exploration and, therefore, reduces output tokens while preserving accuracy. We evaluate RoT on reasoning benchmarks with multiple models, measuring accuracy, token usage, latency, and memory overhead. Findings show small prompt growth but substantial efficiency gains, with RoT reducing output tokens by up to 40%, inference latency by 82%, and cost by 59% while maintaining accuracy. RoT establishes a scalable paradigm for efficient LRM reasoning via dynamic template construction through retrieval.

cs.AI↗

arXiv2026

Enabling Homomorphic Analytical Operations on Compressed Scientific Data with Multi-stage Decompression

Error-controlled lossy compressors have been widely used in scientific applications to reduce the unprecedented size of scientific data while keeping data distortion within a user-specified threshold. While they significantly mitigate the pressure for data storage and transmission, they prolong the time to access the data because decompression is required to transform the binary compressed data into meaningful floating-point numbers. This incurs noticeable overhead for common analytical operations on scientific data that extract or derive useful information, because the time cost of the operations could be much lower than that of decompression. In this work, we design an error-controlled lossy compression and analytical framework that features multi-stage decompression and homomorphic analytical operation algorithms on intermediate decompressed data for reduced data access latency. Our contributions are threefold. (1) We abstract a generic compression pipeline with partial decompression to multiple intermediate data representations and implement four instances based on state-of-the-art high-throughput scientific data compressors. (2) We carefully design homomorphic algorithms to enable direct operations on intermediate decompressed data for three types of analytical operations on scientific data. (3) We evaluate our approach using five real-world scientific datasets. Experimental evaluations demonstrate that our method achieves significant speedups when performing analytical operations on compressed scientific data across all three targeted analytical operation types.

cs.DB↗

arXiv2026

Rate-Distortion Bounds for Heterogeneous Random Fields on Finite Lattices

Since Shannon's foundational work, rate-distortion theory has defined the fundamental limits of lossy compression. Classical results, derived for memoryless and stationary ergodic sources in the asymptotic regime, have shaped both transform and predictive coding architectures, as well as practical standards such as JPEG. Finite-blocklength refinements, initiated by the non-asymptotic achievability and converse bounds of Kostina and Verdu, provide precise characterizations under excess-distortion probability constraints, but primarily for memoryless or statistically homogeneous models. In contrast, error-bounded practical lossy compressors for scientific computing, such as SZ, ZFP, MGARD, and SPERR, are designed for finite, high-dimensional, spatially correlated, and statistically heterogeneous random fields. These compressors partition data into fixed-size tiles that are processed independently, making tile size a central architectural constraint. Structural heterogeneity, finite lattice effects, and tiling constraints are not addressed by existing finite-blocklength analyses. This paper introduces a finite-blocklength rate-distortion framework for heterogeneous random fields on finite lattices, explicitly accounting for the tile-based architectures used in high-performance scientific compressors. The field is modeled as piecewise homogeneous with regionwise stationary second-order statistics, and tiling constraints are incorporated directly into the source model. Under an excess-distortion probability criterion, we establish non-asymptotic achievability, converse bounds and derive a second-order expansion that quantifies the impact of spatial correlation, region geometry, heterogeneity, and tile size on the rate and dispersion.

cs.IT↗

arXiv2026

LUMOS: Democratizing SciML Workflows with L0-Regularized Learning for Unified Feature and Parameter Adaptation

The rapid growth of scientific machine learning (SciML) has accelerated discovery across diverse domains, yet designing effective SciML models remains a challenging task. In practice, building such models often requires substantial prior knowledge and manual expertise, particularly in determining which input features to use and how large the model should be. We introduce LUMOS, an end-to-end framework based on L0-regularized learning that unifies feature selection and model pruning to democratize SciML model design. By employing semi-stochastic gating and reparameterization techniques, LUMOS dynamically selects informative features and prunes redundant parameters during training, reducing the reliance on manual tuning while maintaining predictive accuracy. We evaluate LUMOS across 13 diverse SciML workloads, including cosmology and molecular sciences, and demonstrate its effectiveness and generalizability. Experiments on 13 SciML models show that LUMOS achieves 71.45% parameter reduction and a 6.4x inference speedup on average. Furthermore, Distributed Data Parallel (DDP) training on up to eight GPUs confirms the scalability of

cs.LG↗

arXiv2026

Mitigating Artifacts in Pre-quantization Based Scientific Data Compressors with Quantization-aware Interpolation

Error-bounded lossy compression has been regarded as a promising way to address the ever-increasing amount of scientific data in today's high-performance computing systems. Pre-quantization, a critical technique to remove sequential dependency and enable high parallelism, is widely used to design and develop high-throughput error-controlled data compressors. Despite the extremely high throughput of pre-quantization based compressors, they generally suffer from low data quality with medium or large user-specified error bounds. In this paper, we investigate the artifacts generated by pre-quantization based compressors and propose a novel algorithm to mitigate them. Our contributions are fourfold: (1) We carefully characterize the artifacts in pre-quantization based compressors to understand the correlation between the quantization index and compression error; (2) We propose a novel quantization-aware interpolation algorithm to improve the decompressed data; (3) We parallelize our algorithm in both shared-memory and distributed-memory environments to obtain high performance; (4) We evaluate our algorithm and validate it with two leading pre-quantization based compressors using five real-world datasets. Experiments demonstrate that our artifact mitigation algorithm can effectively improve the quality of decompressed data produced by pre-quantization based compressors while maintaining their high compression throughput.

cs.DC↗

arXiv2026

TopoSZp: Lightweight Topology-Aware Error-controlled Compression for Scientific Data

Error-bounded lossy compression is essential for managing the massive data volumes produced by large-scale HPC simulations. While state-of-the-art compressors such as SZ and ZFP provide strong numerical error guarantees, they often fail to preserve topological structures (example, minima, maxima, and saddle points) that are critical for scientific analysis. Existing topology-aware compressors address this limitation but incur substantial computational overhead. We present TopoSZp, a lightweight, topology-aware, error-controlled lossy compressor that preserves critical points and their relationships while maintaining high compression and decompression performance. Built on the high-throughput SZp compressor, TopoSZp integrates efficient critical point detection, local ordering preservation, and targeted saddle point refinement, all within a relaxed but strictly enforced error bound. Experimental results on real-world scientific datasets show that TopoSZp achieves 3 to 100 times fewer non-preserved critical points, introduces no false positives or incorrect critical point types, and delivers 100 to 10000 times faster compression and 10 to 500 times faster decompression compared to existing topology-aware compressors, while maintaining competitive compression ratios.

cs.DC↗

arXiv2026

HII-DPO: Eliminate Hallucination via Accurate Hallucination-Inducing Counterfactual Images

Large Vision-Language Models (VLMs) have achieved remarkable success across diverse multimodal tasks but remain vulnerable to hallucinations rooted in inherent language bias. Despite recent progress, existing hallucination mitigation methods often overlook the underlying hallucination patterns driven by language bias. In this work, we design a novel pipeline to accurately synthesize Hallucination-Inducing Images (HIIs). Using synthesized HIIs, we reveal a consistent scene-conditioned hallucination pattern: models tend to mention objects that are highly typical of the scene even when visual evidence is removed. To quantify the susceptibility of VLMs to this hallucination pattern, we establish the Masked-Object-Hallucination (MOH) benchmark to rigorously evaluate existing state-of-the-art alignment frameworks. Finally, we leverage HIIs to construct high-quality preference datasets for fine-grained alignment. Experimental results demonstrate that our approach effectively mitigates hallucinations while preserving general model capabilities. Specifically, our method achieves up to a 38% improvement over the current state-of-the-art on standard hallucination benchmarks.

cs.CV↗

arXiv2026

PackKV: Reducing KV Cache Memory Footprint through LLM-Aware Lossy Compression

Transformer-based large language models (LLMs) have demonstrated remarkable potential across a wide range of practical applications. However, long-context inference remains a significant challenge due to the substantial memory requirements of the key-value (KV) cache, which can scale to several gigabytes as sequence length and batch size increase. In this paper, we present \textbf{PackKV}, a generic and efficient KV cache management framework optimized for long-context generation. %, which synergistically supports both latency-critical and throughput-critical inference scenarios. PackKV introduces novel lossy compression techniques specifically tailored to the characteristics of KV cache data, featuring a careful co-design of compression algorithms and system architecture. Our approach is compatible with the dynamically growing nature of the KV cache while preserving high computational efficiency. Experimental results show that, under the same and minimum accuracy drop as state-of-the-art quantization methods, PackKV achieves, on average, \textbf{153.2}\% higher memory reduction rate for the K cache and \textbf{179.6}\% for the V cache. Furthermore, PackKV delivers extremely high execution throughput, effectively eliminating decompression overhead and accelerating the matrix-vector multiplication operation. Specifically, PackKV achieves an average throughput improvement of \textbf{75.7}\% for K and \textbf{171.7}\% for V across A100 and RTX Pro 6000 GPUs, compared to cuBLAS matrix-vector multiplication kernels, while demanding less GPU memory bandwidth. Code available on https://github.com/BoJiang03/PackKV

cs.DC↗

arXiv2026

pMSz: A Distributed Parallel Algorithm for Correcting Extrema and Morse Smale Segmentations in Lossy Compression

Lossy compression, widely used by scientists to reduce data from simulations, experiments, and observations, can distort features of interest even under bounded error. Such distortions may compromise downstream analyses and lead to incorrect scientific conclusions in applications such as combustion and cosmology. This paper presents a distributed and parallel algorithm for correcting topological features, specifically, piecewise linear Morse Smale segmentations (PLMSS), which decompose the domain into monotone regions labeled by their corresponding local minima and maxima. While a single GPU algorithm (MSz) exists for PLMSS correction after compression, no methodology has been developed that scales beyond a single GPU for extreme scale data. We identify the key bottleneck in scaling PLMSS correction as the parallel computation of integral paths, a communication-intensive computation that is notoriously difficult to scale. Instead of explicitly computing and correcting integral paths, our algorithm simplifies MSz by preserving steepest ascending and descending directions across all locations, thereby minimizing interprocess communication while introducing negligible additional storage overhead. With this simplified algorithm and relaxed synchronization, our method achieves over 90% parallel efficiency on 128 GPUs on the Perlmutter supercomputer for real world datasets.

cs.DC↗

arXiv2026

FFCz: Fast Fourier Correction for Spectrum-Preserving Lossy Compression of Scientific Data

This paper introduces a novel technique to preserve spectral features in lossy compression based on a novel fast Fourier correction algorithm\added{ for regular-grid data}. Preserving both spatial and frequency representations of data is crucial for applications such as cosmology, turbulent combustion, and X-ray diffraction, where spatial and frequency views provide complementary scientific insights. In particular, many analysis tasks rely on frequency-domain representations to capture key features, including the power spectrum of cosmology simulations, the turbulent energy spectrum in combustion, and diffraction patterns in reciprocal space for ptychography. However, existing compression methods guarantee accuracy only in the spatial domain while disregarding the frequency domain. To address this limitation, we propose an algorithm that corrects the errors produced by off-the-shelf ``base'' compressors such as SZ3, ZFP, and SPERR, thereby preserving both spatial and frequency representations by bounding errors in both domains. By expressing frequency-domain errors as linear combinations of spatial-domain errors, we derive a region that jointly bounds errors in both domains. Given as input the spatial errors from a base compressor and user-defined error bounds in the spatial and frequency domains, we iteratively project the spatial error vector onto the regions defined by the spatial and frequency constraints until it lies within their intersection. We further accelerate the algorithm using GPU parallelism to achieve practical performance. We validate our approach with datasets from cosmology simulations, X-ray diffraction, combustion simulation, and electroencephalography demonstrating its effectiveness in preserving critical scientific information in both spatial and frequency domains.

cs.DC↗

arXiv2025

DeepCQ: General-Purpose Deep-Surrogate Framework for Lossy Compression Quality Prediction

Error-bounded lossy compression techniques have become vital for scientific data management and analytics, given the ever-increasing volume of data generated by modern scientific simulations and instruments. Nevertheless, assessing data quality post-compression remains computationally expensive due to the intensive nature of metric calculations. In this work, we present a general-purpose deep-surrogate framework for lossy compression quality prediction (DeepCQ), with the following key contributions: 1) We develop a surrogate model for compression quality prediction that is generalizable to different error-bounded lossy compressors, quality metrics, and input datasets; 2) We adopt a novel two-stage design that decouples the computationally expensive feature-extraction stage from the light-weight metrics prediction, enabling efficient training and modular inference; 3) We optimize the model performance on time-evolving data using a mixture-of-experts design. Such a design enhances the robustness when predicting across simulation timesteps, especially when the training and test data exhibit significant variation. We validate the effectiveness of DeepCQ on four real-world scientific applications. Our results highlight the framework's exceptional predictive accuracy, with prediction errors generally under 10\% across most settings, significantly outperforming existing methods. Our framework empowers scientific users to make informed decisions about data compression based on their preferred data quality, thereby significantly reducing I/O and computational overhead in scientific data analysis.

cs.LG↗

arXiv2025

An Efficient Gradient-Aware Error-Bounded Lossy Compressor for Federated Learning

Federated learning (FL) enables collaborative model training without exposing clients' private data, but its deployment is often constrained by the communication cost of transmitting gradients between clients and the central server, especially under system heterogeneity where low-bandwidth clients bottleneck overall performance. Lossy compression of gradient data can mitigate this overhead, and error-bounded lossy compression (EBLC) is particularly appealing for its fine-grained utility-compression tradeoff. However, existing EBLC methods (e.g., SZ), originally designed for smooth scientific data with strong spatial locality, rely on generic predictors such as Lorenzo and interpolation for entropy reduction to improve compression ratio. Gradient tensors, in contrast, exhibit low smoothness and weak spatial correlation, rendering these predictors ineffective and leading to poor compression ratios. To address this limitation, we propose an EBLC framework tailored for FL gradient data to achieve high compression ratios while preserving model accuracy. The core of it is an innovative prediction mechanism that exploits temporal correlations across FL training rounds and structural regularities within convolutional kernels to reduce residual entropy. The predictor is compatible with standard quantizers and entropy coders and comprises (1) a cross-round magnitude predictor based on a normalized exponential moving average, and (2) a sign predictor that leverages gradient oscillation and kernel-level sign consistency. Experiments show that this new EBLC yields up to 1.53x higher compression ratios than SZ3 with lower accuracy loss. Integrated into a real-world FL framework, APPFL, it reduces end-to-end communication time by 76.1%-96.2% under various constrained-bandwidth scenarios, demonstrating strong scalability for real-world FL deployments.

cs.LG↗

arXiv2025

MoE-Compression: How the Compression Error of Experts Affects the Inference Accuracy of MoE Model?

With the widespread application of Mixture of Experts (MoE) reasoning models in the field of LLM learning, efficiently serving MoE models under limited GPU memory constraints has emerged as a significant challenge. Offloading the non-activated experts to main memory has been identified as an efficient approach to address such a problem, while it brings the challenges of transferring the expert between the GPU memory and main memory. We need to explore an efficient approach to compress the expert and analyze how the compression error affects the inference performance. To bridge this gap, we propose employing error-bounded lossy compression algorithms (such as SZ3 and CuSZp) to compress non-activated experts, thereby reducing data transfer overhead during MoE inference. We conduct extensive experiments across various benchmarks and present a comprehensive analysis of how compression-induced errors in different experts affect overall inference accuracy. The results indicate that experts in the shallow layers, which are primarily responsible for the attention mechanism and the transformation of input tokens into vector representations, exhibit minimal degradation in inference accuracy when subjected to bounded errors. In contrast, errors in the middle-layer experts, which are central to model reasoning, significantly impair inference accuracy. Interestingly, introducing bounded errors in the deep-layer experts, which are mainly responsible for instruction following and output integration, can sometimes lead to improvements in inference accuracy.

cs.LG↗

arXiv2025

Boosting Scientific Error-Bounded Lossy Compression through Optimized Synergistic Lossy-Lossless Orchestration

As high-performance computing architectures evolve, more scientific computing workflows are being deployed on advanced computing platforms such as GPUs. These workflows can produce raw data at extremely high throughputs, requiring urgent high-ratio and low-latency error-bounded data compression solutions. In this paper, we propose cuSZ-Hi, an optimized high-ratio GPU-based scientific error-bounded lossy compressor with a flexible, domain-irrelevant, and fully open-source framework design. Our novel contributions are: 1) We maximally optimize the parallelized interpolation-based data prediction scheme on GPUs, enabling the full functionalities of interpolation-based scientific data prediction that are adaptive to diverse data characteristics; 2) We thoroughly explore and investigate lossless data encoding techniques, then craft and incorporate the best-fit lossless encoding pipelines for maximizing the compression ratio of cuSZ-Hi; 3) We systematically evaluate cuSZ-Hi on benchmarking datasets together with representative baselines. Compared to existing state-of-the-art scientific lossy compressors, with comparative or better throughput than existing high-ratio scientific error-bounded lossy compressors on GPUs, cuSZ-Hi can achieve up to 249% compression ratio improvement under the same error bound, and up to 215% compression ratio improvement under the same decompression data PSNR.

cs.DC↗

arXiv2025

GPZ: GPU-Accelerated Lossy Compressor for Particle Data

Particle-based simulations and point-cloud applications generate massive, irregular datasets that challenge storage, I/O, and real-time analytics. Traditional compression techniques struggle with irregular particle distributions and GPU architectural constraints, often resulting in limited throughput and suboptimal compression ratios. In this paper, we present GPZ, a high-performance, error-bounded lossy compressor designed specifically for large-scale particle data on modern GPUs. GPZ employs a novel four-stage parallel pipeline that synergistically balances high compression efficiency with the architectural demands of massively parallel hardware. We introduce a suite of targeted optimizations for computation, memory access, and GPU occupancy that enables GPZ to achieve near-hardware-limit throughput. We conduct an extensive evaluation on three distinct GPU architectures (workstation, data center, and edge) using six large-scale, real-world scientific datasets from five distinct domains. The results demonstrate that GPZ consistently and significantly outperforms five state-of-the-art GPU compressors, delivering up to 8x higher end-to-end throughput while simultaneously achieving superior compression ratios and data quality.

cs.DC↗

arXiv2025

Systematic Evaluation of Optimization Techniques for Long-Context Language Models

Large language models (LLMs) excel across diverse natural language processing tasks but face resource demands and limited context windows. Although techniques like pruning, quantization, and token dropping can mitigate these issues, their efficacy in long-context scenarios and system evaluation remains underexplored. This paper systematically benchmarks these optimizations, characterizing memory usage, latency, and throughput, and studies how these methods impact the quality of text generation. We first analyze individual optimization methods for two LLM architectures supporting long context and then systematically evaluate combinations of these techniques to assess how this deeper analysis impacts performance metrics. We subsequently study the scalability of individual optimization methods on a larger variant with 70 billion-parameter model. Our novel insights reveal that naive combination inference optimization algorithms can adversely affect larger models due to compounded approximation errors, as compared to their smaller counterparts. Experiments show that relying solely on F1 obscures these effects by hiding precision-recall trade-offs in question answering tasks. By integrating system-level profiling with task-specific insights, this study helps LLM practitioners and researchers explore and balance efficiency, accuracy, and scalability across tasks and hardware configurations.

cs.CL↗

arXiv2025

QPET: A Versatile and Portable Quantity-of-Interest-Preservation Framework for Error-Bounded Lossy Compression

Error-bounded lossy compression has been widely adopted in many scientific domains because it can address the challenges in storing, transferring, and analyzing unprecedented amounts of scientific data. Although error-bounded lossy compression offers general data distortion control by enforcing strict error bounds on raw data, it may fail to meet the quality requirements on the results of downstream analysis, a.k.a. Quantities of Interest (QoIs), derived from raw data. This may lead to uncertainties and even misinterpretations in scientific discoveries, significantly limiting the use of lossy compression in practice. In this paper, we propose QPET, a novel, versatile, and portable framework for QoI-preserving error-bounded lossy compression, which overcomes the challenges of modeling diverse QoIs by leveraging numerical strategies. QPET features (1) high portability to multiple existing lossy compressors, (2) versatile preservation to most differentiable univariate and multivariate QoIs, and (3) significant compression improvements in QoI-preservation tasks. Experiments with six real-world datasets demonstrate that integrating QPET into state-of-the-art error-bounded lossy compressors can gain 2x to 10x compression speedups of existing QoI-preserving error-bounded lossy compression solutions, up to 1000% compression ratio improvements to general-purpose compressors, and up to 133% compression ratio improvements to existing QoI-integrated scientific compressors.

cs.DB↗

arXiv2025

To Compress or Not To Compress: Energy Trade-Offs and Benefits of Lossy Compressed I/O

Modern scientific simulations generate massive volumes of data, creating significant challenges for I/O and storage systems. Error-bounded lossy compression (EBLC) offers a solution by reducing data set sizes while preserving data quality within user-specified limits. This study provides the first comprehensive energy characterization of state-of-the-art EBLC algorithms--SZ2, SZ3, ZFP, QoZ, and SZx--across various scientific data sets, CPU generations, and parallel/serial modes. We analyze the energy consumption patterns of compression and decompression operations, as well as the energy trade-offs in data I/O scenarios. Our work demonstrates the relationships between compression ratios, runtime, energy efficiency, and data quality, highlighting the importance of considering compressors and error bounds for specific use cases. We demonstrate that EBLC can significantly reduce I/O energy consumption, with savings of up to two orders of magnitude compared to uncompressed I/O for large data sets. In multi-node HPC environments, we observe energy reductions of approximately 25% when using EBLC. We also show that EBLC can achieve compression ratios of 10-100x, potentially reducing storage device requirements by nearly two orders of magnitude. This work provides a framework for system operators and computational scientists to make informed decisions about implementing EBLC for energy-efficient data management in HPC environments.

cs.DC↗