arXiv subjects

Ciprian-Octavian Truica

Publications and source records attributed to Ciprian-Octavian Truica.

5 recordsLinked to original sources

arXiv2026

IROH: Insightful Ranking Of Humor using Multi-Stage Hybrid Retrieval with Rationale-Distilled LLM Judges for JOKER 2026 Track Task 1 English

Our team, VANGUARD, presents IROH (Insightful Ranking of Humor), a three-stage retrieval system for JOKER Task 1 English at CLEF 2026, achieving first place on the leaderboard with 0.6347 MAP. Our pipeline combines hybrid sparse-dense retrieval, cross-encoder reranking, and a LoRA-adapted Large Language Model judge ensemble. We employ Gemma 4 to generate query-aware rationales under two prompt strategies, generic and typed, and produce up to four types of structured hard negatives for training data construction. Through an ablation across three cross-encoder architectures, four dense embedders, and eight judge configurations, our key findings are threefold: (1) the rationale-distilled judge is the primary driver of ranking quality, whereas appending rationales to the first-stage index contributes negligibly; (2) structured hard negatives degrade generalisation in nearly all configurations despite inflating local validation scores; and (3) across the components we ablate, the lighter, better-calibrated model is competitive with or stronger than its larger counterpart, with the generic-rationale Qwen2.5-7B judge (0.6055 MAP) outperforming every Gemma-4-31B configuration, and the advantage of generic over typed rationales is concentrated almost entirely in the smaller model.

cs.IR↗

arXiv2026

Cross-lingual Relation Extraction with Large Language Models: Zero-Shot, Few-Shot, and Fine-Tuned Evaluation on Romanian

Relation extraction (RE) for low-resource languages is typically constrained by the lack of annotated corpora. We investigate the feasibility of cross-lingual RE for Romanian by combining automatic dataset translation with large language model (LLM) inference. We translate the SemEval-2010 Task 8 benchmark from English to Romanian using an LLM-based translation pipeline and evaluate Gemma 4 31B under zero-shot, few-shot, and QLoRA fine-tuned configurations, against four encoder baselines spanning 125M to 560M parameters: XLM- RoBERTa (base and large), Romanian BERT, and RoBERT- large. We assess two task formulations: relation classification with marked entities and end-to-end extraction. Our results show that Romanian incurs a 3 to 5 percentage point (pp) drop relative to English in prompt-only settings, that few-shot prompting provides marginal gains over zero-shot, and that QLoRA fine-tuning improves macro F1-Score by more than 22 percentage points in both languages while reducing the cross-lingual gap from 3.3 to 1.4pp. The encoder baselines come within 1-4pp of QLoRA Gemma on Romanian despite being 50-250 times smaller, with monolingual Romanian BERT at 125M parameters matching multilingual XLM-R at 278M. The case for using a 31B model for single-task RE on Romanian is therefore weak in deployment scenarios where compute matters. We release the translated dataset, evaluation code, and trained models.

cs.CL↗

arXiv2026

ACAT: A Collaborative Platform for Efficient Aspect-Based Sentiment Dataset Annotation

Aspect-Based Sentiment Analysis (ABSA) requires high-quality datasets to train reliable models. However, existing annotation tools treat output as flat files, leaving researchers to manually consolidate multi-annotator data, reconstruct relational structures, and compute reliability metrics through custom scripts. This paper introduces ACAT (Aspect-based sentiment analysis Collaborative Annotation Tool), a web-based platform natively supporting four ABSA workflows: (1) Aspect-Category Sentiment Analysis, (2) Clause-Level Segmentation, (3) Aspect-Term Sentiment Analysis with character-level position tracking, and (4) Aspect Sentiment Triplet Extraction with dual span offset preservation. Its core contribution is an automated Extract, Transform, Load (ETL) pipeline that aligns collaborative annotations and computes Inter-Annotator Agreement (IAA) metrics directly at export, yielding training-ready datasets. In a preliminary validation on 1,002 restaurant reviews with two annotators of differing expertise, ACAT achieves a median annotation time of 31.58 seconds and a raw IAA ranging from 0.78 to 0.86 across all tasks.

cs.CL↗

arXiv2021

TextBenDS: a generic Textual data Benchmark for Distributed Systems

Extracting top-k keywords and documents using weighting schemes are popular techniques employed in text mining and machine learning for different analysis and retrieval tasks. The weights are usually computed in the data preprocessing step, as they are costly to update and keep track of all the modifications performed on the dataset. Furthermore, computation errors are introduced when analyzing only subsets of the dataset. Therefore, in a Big Data context, it is crucial to lower the runtime of computing weighting schemes, without hindering the analysis process and the accuracy of the machine learning algorithms. To address this requirement for the task of top-k keywords and documents, it is customary to design benchmarks that compare weighting schemes within various configurations of distributed frameworks and database management systems. Thus, we propose a generic document-oriented benchmark for storing textual data and constructing weighting schemes (TextBenDS). Our benchmark offers a generic data model designed with a multidimensional approach for storing text documents. We also propose using aggregation queries with various complexities and selectivities for constructing term weighting schemes, that are utilized in extracting top-k keywords and documents. We evaluate the computing performance of the queries on several distributed environments set within the Apache Hadoop ecosystem. Our experimental results provide interesting insights. As an example, MongoDB proves to have the best overall performance, while Spark's execution time remains almost the same, regardless of the weighting schemes.

cs.DB↗

arXiv2018

Benchmarking Top-K Keyword and Top-K Document Processing with T${}^2$K${}^2$ and T${}^2$K${}^2$D${}^2$

Top-k keyword and top-k document extraction are very popular text analysis techniques. Top-k keywords and documents are often computed on-the-fly, but they exploit weighted vocabularies that are costly to build. To compare competing weighting schemes and database implementations, benchmarking is customary. To the best of our knowledge, no benchmark currently addresses these problems. Hence, in this paper, we present T${}^2$K${}^2$, a top-k keywords and documents benchmark, and its decision support-oriented evolution T${}^2$K${}^2$D${}^2$. Both benchmarks feature a real tweet dataset and queries with various complexities and selectivities. They help evaluate weighting schemes and database implementations in terms of computing performance. To illustrate our bench-marks' relevance and genericity, we successfully ran performance tests on the TF-IDF and Okapi BM25 weighting schemes, on one hand, and on different relational (Oracle, PostgreSQL) and document-oriented (MongoDB) database implementations, on the other hand.

cs.DB↗