arXiv subjects

Cuiping Li

Publications and source records attributed to Cuiping Li.

At least 19 recordsLinked to original sources

arXiv2026

ACTS-SQL: Agentic and Critic-Oriented Tree-Structured SQL Correctness with Large Language Models

Large Language Models (LLMs) have been increasingly adopted in Text-to-SQL systems, yet SQL errors remain a major obstacle in real-world Text-to-SQL inference pipelines. Existing SQL correction approaches either rely on large-scale, high-quality training data with substantial overhead, or adopt single-path agentic workflows that are brittle to early mistakes and prone to error propagation. To develop a practical SQL correctness system for industrial scenarios, we present a training-free framework that formulates SQL correction as a plan-guided, tree-structured debugging process. By maintaining multiple correction strategies and enabling backtracking, the framework mitigates error accumulation during iterative refinement. We further integrate execution-based verification and clause-level diagnostic tools to support strategy pruning and precise error localization. We evaluate the system on the BIRD-Critic benchmark and observe consistent accuracy gains over strong LLM backbones and representative agent-based baselines, achieving a 9.42% improvement over the previous state-of-the-art method. The framework is also deployed in the Torch Log Service (TLS) of Volcano Engine to support an online Text-to-TLS API. In production, it improves execution accuracy from 36.77% to 53.61% on real user queries with a representative strong LLM backbone (GPT-5). These results demonstrate the effectiveness and stability of our approach in real-world deployments.

cs.AI↗

arXiv2026

IDSTune: A Multi-Agent Collaborative Framework for Integrated Database System Tuning

Database tuning is critical for achieving high performance in modern database management systems (DBMSs). Existing methods typically optimize a single component---knobs, indexes, or materialized views---without accounting for their interdependencies. This limitation arises because these components require different tuning strategies and are difficult to integrate within a unified framework. As a result, directly extending a method to multiple components or simply combining separate methods often fails to capture cross-component collaboration and shared tuning signals. Moreover, existing methods are insufficient for handling diverse workloads, evolving data, and dynamic query patterns. To address these limitations, we propose IDSTune, an integrated tuning framework that jointly optimizes multiple configuration components through LLM-driven multi-agent collaboration. IDSTune operates in two phases: (i) workload compression, which extracts and selects task-relevant features, and (ii) configuration recommendation, where specialized agents collaboratively generate and refine configurations for knobs, indexes, and materialized views under the supervision of a centralized coordinator. By incorporating feedback and external knowledge retrieval, IDSTune achieves efficient and globally consistent tuning. Extensive experiments show that IDSTune achieves up to 38% performance improvement and 57% faster tuning, with strong adaptability across diverse scenarios.

cs.DB↗

arXiv2026

MemEmo: Evaluating Emotion in Memory Systems of Agents

Memory systems address the challenge of context loss in Large Language Model during prolonged interactions. However, compared to human cognition, the efficacy of these systems in processing emotion-related information remains inconclusive. To address this gap, we propose an emotion-enhanced memory evaluation benchmark to assess the performance of mainstream and state-of-the-art memory systems in handling affective information. We developed the \textbf{H}uman-\textbf{L}ike \textbf{M}emory \textbf{E}motion (\textbf{HLME}) dataset, which evaluates memory systems across three dimensions: emotional information extraction, emotional memory updating, and emotional memory question answering. Experimental results indicate that none of the evaluated systems achieve robust performance across all three tasks. Our findings provide an objective perspective on the current deficiencies of memory systems in processing emotional memories and suggest a new trajectory for future research and system optimization.

cs.CL↗

arXiv2026

TableCache: Primary Foreign Key Guided KV Cache Precomputation for Low Latency Text-to-SQL

In Text-to-SQL tasks, existing LLM-based methods often include extensive database schemas in prompts, leading to long context lengths and increased prefilling latency. While user queries typically focus on recurrent table sets-offering an opportunity for KV cache sharing across queries-current inference engines, such as SGLang and vLLM, generate redundant prefix cache copies when processing user queries with varying table orders. To address this inefficiency, we propose precomputing table representations as KV caches offline and querying the required ones online. A key aspect of our approach is the computation of table caches while preserving primary foreign key relationships between tables. Additionally, we construct a Table Trie structure to facilitate efficient KV cache lookups during inference. To enhance cache performance, we introduce a cache management system with a query reranking strategy to improve cache hit rates and a computation loading pipeline for parallelizing model inference and cache loading. Experimental results show that our proposed TableCache achieves up to a 3.62x speedup in Time to First Token (TTFT) with negligible performance degradation.

cs.CL↗

arXiv2025

LQRS: Learned Query Re-optimization Framework for Spark SQL

The query optimizer is a fundamental component of database management systems. Recent studies have shown that learned query optimizers outperform traditional cost-based query optimizers. However, they fail to exploit valuable runtime observations generated during query execution to dynamically re-optimize the plan, thereby limiting further improvements in query performance. To address this issue, we propose learned query re-optimization, which allows optimization decisions to be deferred to execution time and guided by actual runtime observations. We realize this idea through LQRS, a learned query re-optimization framework that builds upon Spark SQL, exploiting runtime observations for dynamic plan refinement. Specifically, LQRS employs a curriculum reinforcement learning strategy and jointly supports pre-execution and in-execution optimization, allowing knowledge learned during execution to directly benefit pre-execution planning. Furthermore, we design a plug-and-play planner extension built upon the extensibility interfaces of Spark SQL, enabling online plan modification. Experiments on Spark SQL demonstrate that LQRS reduces end-to-end execution time by up to 90% compared to other learned query optimizers and query re-optimization methods.

cs.DB↗

arXiv2025

Unsupervised Learning for Class Distribution Mismatch

Class distribution mismatch (CDM) refers to the discrepancy between class distributions in training data and target tasks. Previous methods address this by designing classifiers to categorize classes known during training, while grouping unknown or new classes into an "other" category. However, they focus on semi-supervised scenarios and heavily rely on labeled data, limiting their applicability and performance. To address this, we propose Unsupervised Learning for Class Distribution Mismatch (UCDM), which constructs positive-negative pairs from unlabeled data for classifier training. Our approach randomly samples images and uses a diffusion model to add or erase semantic classes, synthesizing diverse training pairs. Additionally, we introduce a confidence-based labeling mechanism that iteratively assigns pseudo-labels to valuable real-world data and incorporates them into the training process. Extensive experiments on three datasets demonstrate UCDM's superiority over previous semi-supervised methods. Specifically, with a 60% mismatch proportion on Tiny-ImageNet dataset, our approach, without relying on labeled data, surpasses OpenMatch (with 40 labels per class) by 35.1%, 63.7%, and 72.5% in classifying known, unknown, and new classes.

cs.CV↗

arXiv2025

QUAD: Quantization and Parameter-Efficient Tuning of LLM with Activation Decomposition

Large Language Models (LLMs) excel in diverse applications but suffer inefficiency due to massive scale. While quantization reduces computational costs, existing methods degrade accuracy in medium-sized LLMs (e.g., Llama-3-8B) due to activation outliers. To address this, we propose QUAD (Quantization with Activation Decomposition), a framework leveraging Singular Value Decomposition (SVD) to suppress activation outliers for effective 4-bit quantization. QUAD estimates activation singular vectors offline using calibration data to construct an orthogonal transformation matrix P, shifting outliers to additional dimensions in full precision while quantizing rest components to 4-bit. Additionally, QUAD enables parameter-efficient fine-tuning via adaptable full-precision outlier weights, narrowing the accuracy gap between quantized and full-precision models. Experiments demonstrate that QUAD achieves 94% ~ 96% accuracy under W4A4 quantization and 98% accuracy with W4A4/A8 and parameter-efficient fine-tuning for Llama-3 and Qwen-2.5 models. Our code is available at \href{https://github.com/hyx1999/Quad}{repository}.

cs.LG↗

arXiv2025

A Query Optimization Method Utilizing Large Language Models

Query optimization is a critical task in database systems, focused on determining the most efficient way to execute a query from an enormous set of possible strategies. Traditional approaches rely on heuristic search methods and cost predictions, but these often struggle with the complexity of the search space and inaccuracies in performance estimation, leading to suboptimal plan choices. This paper presents LLMOpt, a novel framework that leverages Large Language Models (LLMs) to address these challenges through two innovative components: (1) LLM for Plan Candidate Generation (LLMOpt(G)), which eliminates heuristic search by utilizing the reasoning abilities of LLMs to directly generate high-quality query plans, and (2) LLM for Plan Candidate Selection (LLMOpt(S)), a list-wise cost model that compares candidates globally to enhance selection accuracy. To adapt LLMs for query optimization, we propose fine-tuning pre-trained models using optimization data collected offline. Experimental results on the JOB, JOB-EXT, and Stack benchmarks show that LLMOpt(G) and LLMOpt(S) outperform state-of-the-art methods, including PostgreSQL, BAO, and HybridQO. Notably, LLMOpt(S) achieves the best practical performance, striking a balance between plan quality and inference efficiency.

cs.DB↗

arXiv2025

LLMIA: An Out-of-the-Box Index Advisor via In-Context Learning with LLMs

Index recommendation is crucial for optimizing database performance. However, existing heuristic- and learning-based methods often rely on inefficient exhaustive search and estimated costs, leading to low efficiency (due to the vast search space) and unsatisfactory actual latency (due to inaccurate estimations). Inspired by the refinement strategies of experienced DBAs-who efficiently identify and iteratively refine indexes with database feedback-we present LLMIA, an out-of-the-box, tuning-free index advisor leveraging large language models (LLMs) through in-context learning for index recommendation. LLMIA injects database expertise into the LLM using a high-quality demonstration pool and comprehensive workload feature extraction, while iteratively incorporating database feedback to guide the index refinement. This design enables LLMIA to emulate the decision-making process of expert DBAs: efficiently recommending and refining indexes for various workloads within just a few interactions with the DBMS. We validate LLMIA with extensive experiments on five standard OLAP benchmarks (TPC-H with different scales, JOB, TPC-DS, SSB), where it consistently outperforms or matches 12 baselines by producing superior index recommendations with minimal database interactions. Additionally, LLMIA demonstrates robust generalization on two real-world commercial workloads, delivering high-quality recommendations without the need for additional adaptation or retraining, highlighting its out-of-the-box capability.

cs.DB↗

arXiv2025

OmniSQL: Synthesizing High-quality Text-to-SQL Data at Scale

Text-to-SQL, the task of translating natural language questions into SQL queries, plays a crucial role in enabling non-experts to interact with databases. While recent advancements in large language models (LLMs) have significantly enhanced text-to-SQL performance, existing approaches face notable limitations in real-world text-to-SQL applications. Prompting-based methods often depend on closed-source LLMs, which are expensive, raise privacy concerns, and lack customization. Fine-tuning-based methods, on the other hand, suffer from poor generalizability due to the limited coverage of publicly available training data. To overcome these challenges, we propose a novel and scalable text-to-SQL data synthesis framework for automatically synthesizing large-scale, high-quality, and diverse datasets without extensive human intervention. Using this framework, we introduce SynSQL-2.5M, the first million-scale text-to-SQL dataset, containing 2.5 million samples spanning over 16,000 synthetic databases. Each sample includes a database, SQL query, natural language question, and chain-of-thought (CoT) solution. Leveraging SynSQL-2.5M, we develop OmniSQL, a powerful open-source text-to-SQL model available in three sizes: 7B, 14B, and 32B. Extensive evaluations across nine datasets demonstrate that OmniSQL achieves state-of-the-art performance, matching or surpassing leading closed-source and open-source LLMs, including GPT-4o and DeepSeek-V3, despite its smaller size. We release all code, datasets, and models to support further research.

cs.CL↗

arXiv2025

LoRS: Efficient Low-Rank Adaptation for Sparse Large Language Model

Existing low-rank adaptation (LoRA) methods face challenges on sparse large language models (LLMs) due to the inability to maintain sparsity. Recent works introduced methods that maintain sparsity by augmenting LoRA techniques with additional masking mechanisms. Despite these successes, such approaches suffer from an increased memory and computation overhead, which affects efficiency of LoRA methods. In response to this limitation, we introduce LoRS, an innovative method designed to achieve both memory and computation efficiency when fine-tuning sparse LLMs. To mitigate the substantial memory and computation demands associated with preserving sparsity, our approach incorporates strategies of weight recompute and computational graph rearrangement. In addition, we also improve the effectiveness of LoRS through better adapter initialization. These innovations lead to a notable reduction in memory and computation consumption during the fine-tuning phase, all while achieving performance levels that outperform existing LoRA approaches.

cs.CL↗

arXiv2024

Personalized Clustering via Targeted Representation Learning

Clustering traditionally aims to reveal a natural grouping structure within unlabeled data. However, this structure may not always align with users' preferences. In this paper, we propose a personalized clustering method that explicitly performs targeted representation learning by interacting with users via modicum task information (e.g., $\textit{must-link}$ or $\textit{cannot-link}$ pairs) to guide the clustering direction. We query users with the most informative pairs, i.e., those pairs most hard to cluster and those most easy to miscluster, to facilitate the representation learning in terms of the clustering preference. Moreover, by exploiting attention mechanism, the targeted representation is learned and augmented. By leveraging the targeted representation and constrained contrastive loss as well, personalized clustering is obtained. Theoretically, we verify that the risk of personalized clustering is tightly bounded, guaranteeing that active queries to users do mitigate the clustering risk. Experimentally, extensive results show that our method performs well across different clustering tasks and datasets, even when only a limited number of queries are available.

cs.LG↗

arXiv2024

SAM Decoding: Speculative Decoding via Suffix Automaton

Speculative decoding (SD) has been demonstrated as an effective technique for lossless LLM inference acceleration. Retrieval-based SD methods, one kind of model-free method, have yielded promising speedup, but they often rely on incomplete retrieval resources, inefficient retrieval methods, and are constrained to certain domains. This paper presents a novel retrieval-based speculative decoding method that adapts suffix automaton (SAM) for efficient and accurate draft generation by utilizing common text corpus and dynamic text sequence. Unlike existing $n$-gram matching methods, SAM-Decoding finds the exact longest suffix match, achieving an average time complexity of O(1) per generation step of SAM update and suffix retrieval. It can also integrate with existing methods, adaptively selecting a draft generation strategy based on match length to generalize to broader domains. Extensive experiments on Spec-Bench show that our method is $18\%+$ faster than other retrieval-based SD methods. Additionally, when combined with advanced EAGLE-2, it provides an additional speedup of $3.28\%$ -- $11.13\%$ across various-sized LLM backbones. Our code is available at our \href{https://github.com/hyx1999/SAM-Decoding}{repository}.

cs.CL↗

arXiv2024

P$^2$ Law: Scaling Law for Post-Training After Model Pruning

Pruning has become a widely adopted technique for reducing the hardware requirements of large language models (LLMs). To recover model performance after pruning, post-training is commonly employed to mitigate the resulting performance degradation. While post-training benefits from larger datasets, once the dataset size is already substantial, increasing the training data provides only limited performance gains. To balance post-training cost and model performance, it is necessary to explore the optimal amount of post-training data.Through extensive experiments on the Llama-3 and Qwen-2.5 series models, pruned using various common pruning methods, we uncover the scaling \textbf{Law} for \textbf{P}ost-training after model \textbf{P}runing, referred to as the P$^2$ Law.This law identifies four key factors for predicting the pruned model's post-training loss: the model size before pruning, the number of post-training tokens, the pruning rate, and the model's loss before pruning. Moreover, P$^2$ Law can generalize to larger dataset sizes, larger model sizes, and higher pruning rates, offering valuable insights for the post-training of pruned LLMs.

cs.AI↗

arXiv2024

PCQPR: Proactive Conversational Question Planning with Reflection

Conversational Question Generation (CQG) enhances the interactivity of conversational question-answering systems in fields such as education, customer service, and entertainment. However, traditional CQG, focusing primarily on the immediate context, lacks the conversational foresight necessary to guide conversations toward specified conclusions. This limitation significantly restricts their ability to achieve conclusion-oriented conversational outcomes. In this work, we redefine the CQG task as Conclusion-driven Conversational Question Generation (CCQG) by focusing on proactivity, not merely reacting to the unfolding conversation but actively steering it towards a conclusion-oriented question-answer pair. To address this, we propose a novel approach, called Proactive Conversational Question Planning with self-Refining (PCQPR). Concretely, by integrating a planning algorithm inspired by Monte Carlo Tree Search (MCTS) with the analytical capabilities of large language models (LLMs), PCQPR predicts future conversation turns and continuously refines its questioning strategies. This iterative self-refining mechanism ensures the generation of contextually relevant questions strategically devised to reach a specified outcome. Our extensive evaluations demonstrate that PCQPR significantly surpasses existing CQG methods, marking a paradigm shift towards conclusion-oriented conversational question-answering systems.

cs.CL↗

arXiv2024

Is Large Language Model Good at Database Knob Tuning? A Comprehensive Experimental Evaluation

Knob tuning plays a crucial role in optimizing databases by adjusting knobs to enhance database performance. However, traditional tuning methods often follow a Try-Collect-Adjust approach, proving inefficient and database-specific. Moreover, these methods are often opaque, making it challenging for DBAs to grasp the underlying decision-making process. The emergence of large language models (LLMs) like GPT-4 and Claude-3 has excelled in complex natural language tasks, yet their potential in database knob tuning remains largely unexplored. This study harnesses LLMs as experienced DBAs for knob-tuning tasks with carefully designed prompts. We identify three key subtasks in the tuning system: knob pruning, model initialization, and knob recommendation, proposing LLM-driven solutions to replace conventional methods for each subtask. We conduct extensive experiments to compare LLM-driven approaches against traditional methods across the subtasks to evaluate LLMs' efficacy in the knob tuning domain. Furthermore, we explore the adaptability of LLM-based solutions in diverse evaluation settings, encompassing new benchmarks, database engines, and hardware environments. Our findings reveal that LLMs not only match or surpass traditional methods but also exhibit notable interpretability by generating responses in a coherent ``chain-of-thought'' manner. We further observe that LLMs exhibit remarkable generalizability through simple adjustments in prompts, eliminating the necessity for additional training or extensive code modifications. Drawing insights from our experimental findings, we identify several opportunities for future research aimed at advancing the utilization of LLMs in the realm of database management.

cs.DB↗

arXiv2024

A Learn-Then-Reason Model Towards Generalization in Knowledge Base Question Answering

Large-scale knowledge bases (KBs) like Freebase and Wikidata house millions of structured knowledge. Knowledge Base Question Answering (KBQA) provides a user-friendly way to access these valuable KBs via asking natural language questions. In order to improve the generalization capabilities of KBQA models, extensive research has embraced a retrieve-then-reason framework to retrieve relevant evidence for logical expression generation. These multi-stage efforts prioritize acquiring external sources but overlook the incorporation of new knowledge into their model parameters. In effect, even advanced language models and retrievers have knowledge boundaries, thereby limiting the generalization capabilities of previous KBQA models. Therefore, this paper develops KBLLaMA, which follows a learn-then-reason framework to inject new KB knowledge into a large language model for flexible end-to-end KBQA. At the core of KBLLaMA, we study (1) how to organize new knowledge about KBQA and (2) how to facilitate the learning of the organized knowledge. Extensive experiments on various KBQA generalization tasks showcase the state-of-the-art performance of KBLLaMA. Especially on the general benchmark GrailQA and domain-specific benchmark Bio-chemical, KBLLaMA respectively derives a performance gain of up to 3.8% and 9.8% compared to the baselines.

cs.CL↗

arXiv2024

Is Diffusion Model Safe? Severe Data Leakage via Gradient-Guided Diffusion Model

Gradient leakage has been identified as a potential source of privacy breaches in modern image processing systems, where the adversary can completely reconstruct the training images from leaked gradients. However, existing methods are restricted to reconstructing low-resolution images where data leakage risks of image processing systems are not sufficiently explored. In this paper, by exploiting diffusion models, we propose an innovative gradient-guided fine-tuning method and introduce a new reconstruction attack that is capable of stealing private, high-resolution images from image processing systems through leaked gradients where severe data leakage encounters. Our attack method is easy to implement and requires little prior knowledge. The experimental results indicate that current reconstruction attacks can steal images only up to a resolution of $128 \times 128$ pixels, while our attack method can successfully recover and steal images with resolutions up to $512 \times 512$ pixels. Our attack method significantly outperforms the SOTA attack baselines in terms of both pixel-wise accuracy and time efficiency of image reconstruction. Furthermore, our attack can render differential privacy ineffective to some extent.

cs.CV↗