arXiv subjects

Hyunjoon Kim

Publications and source records attributed to Hyunjoon Kim.

10 recordsLinked to original sources

arXiv2026

Vivace: Exact Temporal OLAP over Interval Histories via Independent Serverless Execution

Temporal online analytical processing (OLAP) analyzes past states of data whose values change over time. Such histories are naturally stored as interval histories, in which each row records the period during which a value remained valid. Because temporal analyses typically arrive in infrequent, intermittent bursts, serverless execution that launches functions only at query time offers a cost advantage over always-on clusters. Splitting a computation that a single process performs as a whole across independent serverless functions, however, breaks correctness in two ways. A function may not receive the rows that determine the state of its time range, and naively summing partial results yields incorrect answers for duration-weighted and cumulative-threshold queries. Existing SQL engines and serverless analytics do not address both problems together. This paper presents Vivace, a serverless system for exact temporal OLAP over interval histories. Vivace resolves the two problems in separate stages. Before any query arrives, a pre-query layout step partitions the interval history, replicating boundary-crossing intervals so each function computes its range completely from a single file. At query time, a merge step combines partial results under operator-specific rules. Associative aggregates merge intermediate values, and ranking re-orders candidates within each time range. We prove that this partitioned execution matches single-process computation up to canonical form. Evaluated on AWS Lambda with real-world datasets, Vivace reduces latency and monetary cost by up to 82% and 84%, respectively, against an equivalent SQL baseline that queries the history directly, demonstrating robust generality and efficiency.

cs.DB↗

arXiv2026

Trajectory-Aware Clinical Risk Prediction via Severity-Grounded Knowledge Graphs and Retrieval-Augmented Generation

While Electronic Health Records (EHRs) offer a wealth of clinical data, effectively augmenting a patient's records with heterogeneous external knowledge to predict the patient's clinical risk remains a significant challenge. Existing methods fail to capture disease severity, treatment responses, and nuanced clinical progression, due to data sparsity and the underutilization of unstructured clinical notes. To address these challenges, we propose TRACER (a trajectory-aware and clinically grounded prediction framework) that (1) constructs a medical knowledge graph enriched with severity information from medical literature, (2) retrieves clinically relevant, severity-weighted paths of a patient's progression from the knowledge graph, (3) extracts clinically relevant events from unstructured clinical notes, and (4) augments patient context with similar peer cases. Experiments on the MIMIC-III and MIMIC-IV datasets demonstrate large gains over state-of-the-art baselines, with up to 28.5% increase in Macro F1 score for the mortality prediction task, and 19.7% increase for the readmission prediction task.

cs.AI↗

arXiv2026

DFLOP: A Data-driven Framework for Multimodal LLM Training Pipeline Optimization

Multimodal Large Language Models (MLLMs) have achieved remarkable advances by integrating text, image, and audio understanding within a unified architecture. However, existing distributed training frameworks remain fundamentally data-blind: they parallelize computation without accounting for variations in input data characteristics. This data unawareness leads to severe computation skew across stages and microbatches, where heterogeneous multimodal inputs incur different processing costs. Consequently, GPU resources are unevenly utilized, synchronization delays accumulate, and overall training efficiency degrades. To address this limitation, we present DFLOP, a data-driven framework for multimodal LLM training pipeline optimization. DFLOP continuously profiles runtime behavior to capture data-induced computation variance and employs predictive scheduling to balance workloads across stages and microbatches. By coupling data characteristics with execution planning, DFLOP substantially improves GPU utilization and throughput. Extensive experiments on large-scale multimodal benchmarks show that DFLOP achieves up to 3.6x faster training compared to state-of-the-art distributed training frameworks.

cs.DC↗

arXiv2025

Cooperative Retrieval-Augmented Generation for Question Answering: Mutual Information Exchange and Ranking by Contrasting Layers

Since large language models (LLMs) have a tendency to generate factually inaccurate output, retrieval-augmented generation (RAG) has gained significant attention as a key means to mitigate this downside of harnessing only LLMs. However, existing RAG methods for simple and multi-hop question answering (QA) are still prone to incorrect retrievals and hallucinations. To address these limitations, we propose CoopRAG, a novel RAG framework for the question answering task in which a retriever and an LLM work cooperatively with each other by exchanging informative knowledge, and the earlier and later layers of the retriever model work cooperatively with each other to accurately rank the retrieved documents relevant to a given query. In this framework, we (i) unroll a question into sub-questions and a reasoning chain in which uncertain positions are masked, (ii) retrieve the documents relevant to the question augmented with the sub-questions and the reasoning chain, (iii) rerank the documents by contrasting layers of the retriever, and (iv) reconstruct the reasoning chain by filling the masked positions via the LLM. Our experiments demonstrate that CoopRAG consistently outperforms state-of-the-art QA methods on three multi-hop QA datasets as well as a simple QA dataset in terms of both the retrieval and QA performances. Our code is available.

cs.CL↗

arXiv2025

Efficient Defective Clique Enumeration and Search with Worst-Case Optimal Search Space

A $k$-defective clique is a relaxation of the traditional clique definition, allowing up to $k$ missing edges. This relaxation is crucial in various real-world applications such as link prediction, community detection, and social network analysis. Although the problems of enumerating maximal $k$-defective cliques and searching a maximum $k$-defective clique have been extensively studied, existing algorithms suffer from limitations such as the combinatorial explosion of small partial solutions and sub-optimal search spaces. To address these limitations, we propose a novel clique-first branch-and-bound framework that first generates cliques and then adds missing edges. Furthermore, we introduce a new pivoting technique that achieves a search space size of $\mathcal{O}(3^{\frac{n}{3}} \cdot n^k)$, where $n$ is the number of vertices in the input graph. We prove that the worst-case number of maximal $k$-defective cliques is $\Omega(3^{\frac{n}{3}} \cdot n^k)$ when $k$ is a constant, establishing that our algorithm's search space is worst-case optimal. Leveraging the diameter-two property of defective cliques, we further reduce the search space size to $\mathcal{O}(n \cdot 3^{\frac{\delta}{3}} \cdot (\delta \Delta)^k)$, where $\delta$ is the degeneracy and $\Delta$ is the maximum degree of the input graph. We also propose an efficient framework for maximum $k$-defective clique search based on our branch-and-bound, together with practical techniques to reduce the search space. Experiments on real-world benchmark datasets with more than 1 million edges demonstrate that each of our proposed algorithms for maximal $k$-defective clique enumeration and maximum $k$-defective clique search outperforms the respective state-of-the-art algorithms by up to four orders of magnitude in terms of processing time.

cs.DS↗

arXiv2025

Exqutor: Extended Query Optimizer for Vector-augmented Analytical Queries

Vector similarity search is becoming increasingly important for data science pipelines, particularly in Retrieval-Augmented Generation (RAG), where it enhances large language model inference by enabling efficient retrieval of relevant external knowledge. As RAG expands with table-augmented generation to incorporate structured data, workloads integrating table and vector search are becoming more prevalent. However, efficiently executing such queries remains challenging due to inaccurate cardinality estimation for vector search components, leading to suboptimal query plans. In this paper, we propose Exqutor, an extended query optimizer for vector-augmented analytical queries. Exqutor is a pluggable cardinality estimation framework designed to address this issue, leveraging exact cardinality query optimization techniques to enhance estimation accuracy when vector indexes (e.g., HNSW, IVF) are available. In scenarios lacking these indexes, we employ a sampling-based approach with adaptive sampling size adjustment, dynamically tuning the sample size to balance estimation accuracy and sampling overhead. This allows Exqutor to efficiently approximate vector search cardinalities while minimizing computational costs. We integrate our framework into pgvector, VBASE, and DuckDB, demonstrating performance improvements of up to four orders of magnitude on vector-augmented analytical queries.

cs.DB↗

arXiv2025

ProgRAG: Hallucination-Resistant Progressive Retrieval and Reasoning over Knowledge Graphs

Large Language Models (LLMs) demonstrate strong reasoning capabilities but struggle with hallucinations and limited transparency. Recently, KG-enhanced LLMs that integrate knowledge graphs (KGs) have been shown to improve reasoning performance, particularly for complex, knowledge-intensive tasks. However, these methods still face significant challenges, including inaccurate retrieval and reasoning failures, often exacerbated by long input contexts that obscure relevant information or by context constructions that struggle to capture the richer logical directions required by different question types. Furthermore, many of these approaches rely on LLMs to directly retrieve evidence from KGs, and to self-assess the sufficiency of this evidence, which often results in premature or incorrect reasoning. To address the retrieval and reasoning failures, we propose ProgRAG, a multi-hop knowledge graph question answering (KGQA) framework that decomposes complex questions into sub-questions, and progressively extends partial reasoning paths by answering each sub-question. At each step, external retrievers gather candidate evidence, which is then refined through uncertainty-aware pruning by the LLM. Finally, the context for LLM reasoning is optimized by organizing and rearranging the partial reasoning paths obtained from the sub-question answers. Experiments on three well-known datasets demonstrate that ProgRAG outperforms existing baselines in multi-hop KGQA, offering improved reliability and reasoning quality.

cs.AI↗

arXiv2024

Embedded FPGA Developments in 130nm and 28nm CMOS for Machine Learning in Particle Detector Readout

Embedded field programmable gate array (eFPGA) technology allows the implementation of reconfigurable logic within the design of an application-specific integrated circuit (ASIC). This approach offers the low power and efficiency of an ASIC along with the ease of FPGA configuration, particularly beneficial for the use case of machine learning in the data pipeline of next-generation collider experiments. An open-source framework called "FABulous" was used to design eFPGAs using 130 nm and 28 nm CMOS technology nodes, which were subsequently fabricated and verified through testing. The capability of an eFPGA to act as a front-end readout chip was assessed using simulation of high energy particles passing through a silicon pixel sensor. A machine learning-based classifier, designed for reduction of sensor data at the source, was synthesized and configured onto the eFPGA. A successful proof-of-concept was demonstrated through reproduction of the expected algorithm result on the eFPGA with perfect accuracy. Further development of the eFPGA technology and its application to collider detector readout is discussed.

cs.AR↗

arXiv2024

Subgraph-Aware Training of Language Models for Knowledge Graph Completion Using Structure-Aware Contrastive Learning

Fine-tuning pre-trained language models (PLMs) has recently shown a potential to improve knowledge graph completion (KGC). However, most PLM-based methods focus solely on encoding textual information, neglecting the long-tailed nature of knowledge graphs and their various topological structures, e.g., subgraphs, shortest paths, and degrees. We claim that this is a major obstacle to achieving higher accuracy of PLMs for KGC. To this end, we propose a Subgraph-Aware Training framework for KGC (SATKGC) with two ideas: (i) subgraph-aware mini-batching to encourage hard negative sampling and to mitigate an imbalance in the frequency of entity occurrences during training, and (ii) new contrastive learning to focus more on harder in-batch negative triples and harder positive triples in terms of the structural properties of the knowledge graph. To the best of our knowledge, this is the first study to comprehensively incorporate the structural inductive bias of the knowledge graph into fine-tuning PLMs. Extensive experiments on three KGC benchmarks demonstrate the superiority of SATKGC. Our code is available.

cs.CL↗

arXiv2016

FM-index of Alignment with Gaps

Recently, a compressed index for similar strings, called the FM-index of alignment (FMA), has been proposed with the functionalities of pattern search and random access. The FMA is quite efficient in space requirement and pattern search time, but it is applicable only for an alignment of similar strings without gaps. In this paper we propose the FM-index of alignment with gaps, a realistic index for similar strings, which allows gaps in their alignment. For this, we design a new version of the suffix array of alignment by using alignment transformation and a new definition of the alignment-suffix. The new suffix array of alignment enables us to support the LF-mapping and backward search, the key functionalities of the FM-index, regardless of gap existence in the alignment. We experimentally compared our index with RLCSA due to Makinen et al. on 100 genome sequences from the 1000 Genomes Project. The index size of our index is less than one third of that of RLCSA.

cs.DS↗