arXiv subjects

Teeradaj Racharak

Publications and source records attributed to Teeradaj Racharak.

11 recordsLinked to original sources

arXiv2026

Oilbird: Training-Free Speculative Decoding with Keys the Verifier Already Computes

Training-free speculative decoding drafts by matching an exact suffix of the context against a pool of earlier context. That lookup misses correct drafts already in the pool, most visibly on tool-calling traffic, where a request repeats almost everything but the few values minted for it, and where one rejected token discards the correct continuation behind it. We diagnose the failure position by position across ten benchmarks and find it to be a problem of addressing rather than of coverage: on our densest tool-calling benchmark, about half of what the strongest exact-match drafter misses is present in the pool yet unreachable by exact matching. We therefore propose a second, semantic draft source: the same pool, re-keyed by the hidden state the verifier has already computed at each committed token, together with a merge that lets it ride inside an existing lexical drafter's tree. In three published drafters, at matched pool and budget, it lifts accepted length by 24-29%. Oilbird reaches 4.4x autoregressive decoding speed on API-Bank, against 3.9x for the strongest training-free baseline in our harness and 2.0x for EAGLE-3.

cs.AI↗

arXiv2026

QUARE: Quality-Aware Requirements Analysis through Multi-Agent Dialectical Negotiation

Automating requirements quality analysis remains challenging because multiple, often conflicting quality attributes must be balanced while preserving stakeholder intent. Existing Large-Language-Model (LLM) approaches predominantly rely on task-oriented decomposition or implicit aggregation, limiting their ability to systematically surface and resolve cross-quality conflicts. We present QUARE (QUality-Aware REquirements Analysis), a multi-agent framework that takes a project description as input and formulates requirements quality analysis as structured negotiation among five quality-specialized agents: Safety, Efficiency, Green, Trustworthiness, and Responsibility, coordinated by a dedicated orchestrator. QUARE introduces a dialectical negotiation protocol that explicitly exposes inter-quality conflicts and resolves them through iterative proposal, critique, and synthesis. Negotiated outcomes are transformed into structurally sound KAOS goal models via topology validation and verified against industry standards through retrieval-augmented generation (RAG). We evaluate QUARE on five benchmark systems drawn from established RE benchmarks, MARE and iReDev, and an industrial autonomous-driving specification, spanning safety-critical, financial, and information-system domains. Results show that QUARE achieves 98.2% compliance coverage, a 105% improvement over both baselines; 94.9% semantic preservation, a 2.3 percentage-point improvement over the best baseline; and high verifiability, with a score of 4.96 out of 5.0, while generating 25-43% more requirements than existing multi-agent RE frameworks. These findings suggest that, when using capable instruction-tuned models, architectural choices such as quality-dimension decomposition, explicit negotiation, and automated verification may contribute more to output quality than model scale alone.

cs.SE↗

arXiv2026

ArgRE: Formal Argumentation for Conflict Resolution in Multi-Agent Requirements Negotiation

As software systems grow in complexity, they must satisfy an increasing number of competing quality attributes, making it essential to balance them in a principled manner -- for example, a safety requirement for sensor-fusion verification may conflict with a tight planning-cycle budget. Multi-agent large language model frameworks support this balancing process by assigning specialized agents to different objectives. However, their conflict resolution is typically heuristic. Requirements are aggregated implicitly without explicit acceptance or rejection, limiting auditability in regulated domains. We present ArgRE, a multi-agent requirements negotiation system that embeds Dung-style abstract argumentation into the negotiation stage. Each proposal, critique, and refinement is modeled as an argument, conflicts are represented as directed attack relations, and the accepted set of arguments is computed under grounded and preferred semantics. The pipeline further integrates KAOS goal modeling, multi-layer verification, and standards-oriented artifact generation. Evaluation across five case studies spanning safety-critical, financial, and information-system domains shows that ArgRE provides argument-level traceability absent from existing frameworks. Independent evaluators rated its decision justifications significantly higher than those of heuristic synthesis (4.32 vs. 3.07, p < 0.001), indicating improved auditability, while semantic intent preservation remains comparable (94.9% BERTScore F1) and compliance coverage reaches 84.7% versus 47.6%--47.8% for baselines. Structural analysis further confirms that the default pairwise protocol yields acyclic graphs in which grounded and preferred semantics coincide, whereas cross-pair arbitration introduces controlled cyclicity, leading to predictable divergence between the two semantics.

cs.SE↗

arXiv2025

Generative AI for Requirements Engineering: A Systematic Literature Review

Introduction: Requirements engineering faces challenges due to the handling of increasingly complex software systems. These challenges can be addressed using generative AI. Given that GenAI based RE has not been systematically analyzed in detail, this review examines related research, focusing on trends, methodologies, challenges, and future directions. Methods: A systematic methodology for paper selection, data extraction, and feature analysis is used to comprehensively review 238 articles published from 2019 to 2025 and available from major academic databases. Results: Generative pretrained transformer models dominate current applications (67.3%), but research remains unevenly distributed across RE phases, with analysis (30.0%) and elicitation (22.1%) receiving the most attention, and management (6.8%) underexplored. Three core challenges: reproducibility (66.8%), hallucinations (63.4%), and interpretability (57.1%) form a tightly interlinked triad affecting trust and consistency. Strong correlations (35% cooccurrence) indicate these challenges must be addressed holistically. Industrial adoption remains nascent, with over 90% of studies corresponding to early stage development and only 1.3% reaching production level integration. Conclusions: Evaluation practices show maturity gaps, limited tool and dataset availability, and fragmented benchmarking approaches. Despite the transformative potential of GenAI based RE, several barriers hinder practical adoption. The strong correlations among core challenges demand specialized architectures targeting interdependencies rather than isolated solutions. The limited deployment reflects systemic bottlenecks in generalizability, data quality, and scalable evaluation methods. Successful adoption requires coordinated development across technical robustness, methodological maturity, and governance integration.

cs.SE↗

arXiv2025

Towards Human-interpretable Explanation in Code Clone Detection using LLM-based Post Hoc Explainer

Recent studies highlight various machine learning (ML)-based techniques for code clone detection, which can be integrated into developer tools such as static code analysis. With the advancements brought by ML in code understanding, ML-based code clone detectors could accurately identify and classify cloned pairs, especially semantic clones, but often operate as black boxes, providing little insight into the decision-making process. Post hoc explainers, on the other hand, aim to interpret and explain the predictions of these ML models after they are made, offering a way to understand the underlying mechanisms driving the model's decisions. However, current post hoc techniques require white-box access to the ML model or are computationally expensive, indicating a need for advanced post hoc explainers. In this paper, we propose a novel approach that leverages the in-context learning capabilities of large language models to elucidate the predictions made by the ML-based code clone detectors. We perform a study using ChatGPT-4 to explain the code clone results inferred by GraphCodeBERT. We found that our approach is promising as a post hoc explainer by giving the correct explanations up to 98% and offering good explanations 95% of the time. However, the explanations and the code line examples given by the LLM are useful in some cases. We also found that lowering the temperature to zero helps increase the accuracy of the explanation. Lastly, we list the insights that can lead to further improvements in future work. This study paves the way for future studies in using LLMs as a post hoc explainer for various software engineering tasks.

cs.SE↗

arXiv2025

Test It Before You Trust It: Applying Software Testing for Trustworthy In-context Learning

In-context learning (ICL) has emerged as a powerful capability of large language models (LLMs), enabling them to perform new tasks based on a few provided examples without explicit fine-tuning. Despite their impressive adaptability, these models remain vulnerable to subtle adversarial perturbations and exhibit unpredictable behavior when faced with linguistic variations. Inspired by software testing principles, we introduce a software testing-inspired framework, called MMT4NL, for evaluating the trustworthiness of in-context learning by utilizing adversarial perturbations and software testing techniques. It includes diverse evaluation aspects of linguistic capabilities for testing the ICL capabilities of LLMs. MMT4NL is built around the idea of crafting metamorphic adversarial examples from a test set in order to quantify and pinpoint bugs in the designed prompts of ICL. Our philosophy is to treat any LLM as software and validate its functionalities just like testing the software. Finally, we demonstrate applications of MMT4NL on the sentiment analysis and question-answering tasks. Our experiments could reveal various linguistic bugs in state-of-the-art LLMs.

cs.SE↗

arXiv2024

A Decoupling and Aggregating Framework for Joint Extraction of Entities and Relations

Named Entity Recognition and Relation Extraction are two crucial and challenging subtasks in the field of Information Extraction. Despite the successes achieved by the traditional approaches, fundamental research questions remain open. First, most recent studies use parameter sharing for a single subtask or shared features for both two subtasks, ignoring their semantic differences. Second, information interaction mainly focuses on the two subtasks, leaving the fine-grained informtion interaction among the subtask-specific features of encoding subjects, relations, and objects unexplored. Motivated by the aforementioned limitations, we propose a novel model to jointly extract entities and relations. The main novelties are as follows: (1) We propose to decouple the feature encoding process into three parts, namely encoding subjects, encoding objects, and encoding relations. Thanks to this, we are able to use fine-grained subtask-specific features. (2) We propose novel inter-aggregation and intra-aggregation strategies to enhance the information interaction and construct individual fine-grained subtask-specific features, respectively. The experimental results demonstrate that our model outperforms several previous state-of-the-art models. Extensive additional experiments further confirm the effectiveness of our model.

cs.CL↗

arXiv2024

A Mutual Inclusion Mechanism for Precise Boundary Segmentation in Medical Images

In medical imaging, accurate image segmentation is crucial for quantifying diseases, assessing prognosis, and evaluating treatment outcomes. However, existing methods lack an in-depth integration of global and local features, failing to pay special attention to abnormal regions and boundary details in medical images. To this end, we present a novel deep learning-based approach, MIPC-Net, for precise boundary segmentation in medical images. Our approach, inspired by radiologists' working patterns, features two distinct modules: (i) \textbf{Mutual Inclusion of Position and Channel Attention (MIPC) module}: To enhance the precision of boundary segmentation in medical images, we introduce the MIPC module, which enhances the focus on channel information when extracting position features and vice versa; (ii) \textbf{GL-MIPC-Residue}: To improve the restoration of medical images, we propose the GL-MIPC-Residue, a global residual connection that enhances the integration of the encoder and decoder by filtering out invalid information and restoring the most effective information lost during the feature extraction process. We evaluate the performance of the proposed model using metrics such as Dice coefficient (DSC) and Hausdorff Distance (HD) on three publicly accessible datasets: Synapse, ISIC2018-Task, and Segpc. Our ablation study shows that each module contributes to improving the quality of segmentation results. Furthermore, with the assistance of both modules, our approach outperforms state-of-the-art methods across all metrics on the benchmark datasets, notably achieving a 2.23mm reduction in HD on the Synapse dataset, strongly evidencing our model's enhanced capability for precise image boundary segmentation. Codes will be available at https://github.com/SUN-1024/MIPC-Net.

eess.IV↗

arXiv2024

Logic of Awareness in Agent's Reasoning

The aim of this study is to formally express awareness for modeling practical agent communication. The notion of awareness has been proposed as a set of propositions for each agent, to which he/she pays attention, and has contributed to avoiding \textit{logical omniscience}. However, when an agent guesses another agent's knowledge states, what matters are not propositions but are accessible possible worlds. Therefore, we introduce a partition of possible worlds connected to awareness, that is an equivalence relation, to denote \textit{indistinguishable} worlds. Our logic is called Awareness Logic with Partition ($\mathcal{ALP}$). In this paper, we first show a running example to illustrate a practical social game. Thereafter, we introduce syntax and Kripke semantics of the logic and prove its completeness. Finally, we outline an idea to incorporate some epistemic actions with dynamic operators that change the state of awareness.

cs.MA↗

arXiv2023

DA-TransUNet: Integrating Spatial and Channel Dual Attention with Transformer U-Net for Medical Image Segmentation

Accurate medical image segmentation is critical for disease quantification and treatment evaluation. While traditional Unet architectures and their transformer-integrated variants excel in automated segmentation tasks. However, they lack the ability to harness the intrinsic position and channel features of image. Existing models also struggle with parameter efficiency and computational complexity, often due to the extensive use of Transformers. To address these issues, this study proposes a novel deep medical image segmentation framework, called DA-TransUNet, aiming to integrate the Transformer and dual attention block(DA-Block) into the traditional U-shaped architecture. Unlike earlier transformer-based U-net models, DA-TransUNet utilizes Transformers and DA-Block to integrate not only global and local features, but also image-specific positional and channel features, improving the performance of medical image segmentation. By incorporating a DA-Block at the embedding layer and within each skip connection layer, we substantially enhance feature extraction capabilities and improve the efficiency of the encoder-decoder structure. DA-TransUNet demonstrates superior performance in medical image segmentation tasks, consistently outperforming state-of-the-art techniques across multiple datasets. In summary, DA-TransUNet offers a significant advancement in medical image segmentation, providing an effective and powerful alternative to existing techniques. Our architecture stands out for its ability to improve segmentation accuracy, thereby advancing the field of automated medical image diagnostics. The codes and parameters of our model will be publicly available at https://github.com/SUN-1024/DA-TransUnet.

eess.IV↗

arXiv2022

A Summary of the ALQAC 2021 Competition

We summarize the evaluation of the first Automated Legal Question Answering Competition (ALQAC 2021). The competition this year contains three tasks, which aims at processing the statute law document, which are Legal Text Information Retrieval (Task 1), Legal Text Entailment Prediction (Task 2), and Legal Text Question Answering (Task 3). The final goal of these tasks is to build a system that can automatically determine whether a particular statement is lawful. There is no limit to the approaches of the participating teams. This year, there are 5 teams participating in Task 1, 6 teams participating in Task 2, and 5 teams participating in Task 3. There are in total 36 runs submitted to the organizer. In this paper, we summarize each team's approaches, official results, and some discussion about the competition. Only results of the teams who successfully submit their approach description paper are reported in this paper.

cs.CL↗