We propose a source code search system named CHICOT (Code search with HIgh level COnText) to assist developers in reusing existing code. While previous studies have examined code search on the basis of code-level, fine-grained specifications such as functionality, logic, or implementation, CHICOT addresses a unique mission: code search with high-level contextual information, such as the purpose or domain of a developer's project. It achieves this feature by first extracting the context information from codebases and then considering this context during the search. It provides a VSCode plugin for daily coding assistance, and the built-in crawler ensures up-to-date code suggestions. The case study attests to the utility of CHICOT in real-world scenarios.

Large language models (LLMs) have proficiently solved a broad range of tasks with their rich knowledge but often struggle with logical reasoning. To foster the research on logical reasoning, many benchmarks have been proposed so far. However, most of these benchmarks are limited to English, hindering the evaluation of LLMs specialized for each language. To address this, we propose JFLD (Japanese Formal Logic Deduction), a deductive reasoning benchmark for Japanese. JFLD assess whether LLMs can generate logical steps to (dis-)prove a given hypothesis based on a given set of facts. Its key features are assessing pure logical reasoning abilities isolated from knowledge and assessing various reasoning rules. We evaluate various Japanese LLMs and see that they are still poor at logical reasoning, thus highlighting a substantial need for future research.

Large language models (LLMs) are capable of solving a wide range of tasks, yet they have struggled with reasoning. To address this, we propose $\textbf{Additional Logic Training (ALT)}$, which aims to enhance LLMs' reasoning capabilities by program-generated logical reasoning samples. We first establish principles for designing high-quality samples by integrating symbolic logic theory and previous empirical insights. Then, based on these principles, we construct a synthetic corpus named $\textbf{Formal} \ \textbf{Logic} \ \textbf{\textit{D}eduction} \ \textbf{\textit{D}iverse}$ (FLD$ _{\times2}$), comprising numerous samples of multi-step deduction with unknown facts, diverse reasoning rules, diverse linguistic expressions, and challenging distractors. Finally, we empirically show that ALT on FLD$ _{\times2}$ substantially enhances the reasoning capabilities of state-of-the-art LLMs, including LLaMA-3.1-70B. Improvements include gains of up to 30 points on logical reasoning benchmarks, up to 10 points on math and coding benchmarks, and 5 points on the benchmark suite BBH.

Writing a readme is a crucial aspect of software development as it plays a vital role in managing and reusing program code. Though it is a pain point for many developers, automatically creating one remains a challenge even with the recent advancements in large language models (LLMs), because it requires generating an abstract description from thousands of lines of code. In this demo paper, we show that LLMs are capable of generating a coherent and factually correct readmes if we can identify a code fragment that is representative of the repository. Building upon this finding, we developed LARCH (LLM-based Automatic Readme Creation with Heuristics) which leverages representative code identification with heuristics and weak supervision. Through human and automated evaluations, we illustrate that LARCH can generate coherent and factually correct readmes in the majority of cases, outperforming a baseline that does not rely on representative code identification. We have made LARCH open-source and provided a cross-platform Visual Studio Code interface and command-line interface, accessible at https://github.com/hitachi-nlp/larch. A demo video showcasing LARCH's capabilities is available at https://youtu.be/ZUKkh5ED-O4.

We study a synthetic corpus based approach for language models (LMs) to acquire logical deductive reasoning ability. The previous studies generated deduction examples using specific sets of deduction rules. However, these rules were limited or otherwise arbitrary. This can limit the generalizability of acquired deductive reasoning ability. We rethink this and adopt a well-grounded set of deduction rules based on formal logic theory, which can derive any other deduction rules when combined in a multistep way. We empirically verify that LMs trained on the proposed corpora, which we name $\textbf{FLD}$ ($\textbf{F}$ormal $\textbf{L}$ogic $\textbf{D}$eduction), acquire more generalizable deductive reasoning ability. Furthermore, we identify the aspects of deductive reasoning ability on which deduction corpora can enhance LMs and those on which they cannot. Finally, on the basis of these results, we discuss the future directions for applying deduction corpora or other approaches for each aspect. We release the code, data, and models.

One of the challenges in text generation is to control text generation as intended by the user.Previous studies proposed specifying the keywords that should be included in the generated text.However, this approach is insufficient to generate text that reflect the user's intent.For example, placing an important keyword at the beginning of the text would help attract the reader's attention; however, existing methods do not enable such flexible control.In this paper, we tackle a novel task of controlling not only keywords but also the position of each keyword in the text generation.To this end, we propose a task-independent method that uses special tokens to control the relative position of keywords.Experimental results on summarization and story generation tasks show that the proposed method can control keywords and their positions.The experimental results also demonstrate that controlling the keyword positions can generate summary texts that are closer to the user's intent than baseline.

Machine reading comprehension (MRC) is a task for question answering that finds answers to questions from documents of knowledge.Most studies on the domain adaptation of MRC require documents describing knowledge of the target domain.However, it is sometimes difficult to prepare such documents.The goal of this study was to transfer an MRC model to another domain without documents in an unsupervised manner.Therefore, unlike previous studies, we propose a domain-adaptation framework of MRC under the assumption that the only available data in the target domain are human conversations between a user asking questions and an expert answering the questions.The framework consists of three processes: (1) training an MRC model on the source domain, (2) converting conversations into documents using document generation (DG), a task we developed for retrieving important information from several human conversations and converting it to an abstractive document text, and (3) transferring the MRC model to the target domain with unsupervised domain adaptation.To the best of our knowledge, our research is the first to use conversation data to train MRC models in an unsupervised manner.We show that the MRC model successfully obtains question-answering ability from conversations in the target domain.

This paper proposes an online mixture modeling methodology in which individual components can have different marginal distributions and dependency structures. Mixture models have been widely studied and applied to various application areas, including density estimation, fraud/failure detection, image segmentation, etc. Previous research has been almost exclusively focused on mixture models having components of a single type (e.g., a Gaussian mixture model.) However, recent growing needs for complicated data modeling necessitate the use of more flexible mixture models (e.g., a mixture of a lognormal distribution for medical costs and a Gaussian distribution for blood pressure, for medical analytics.) Our key ideas include: 1) separating marginal distributions and their dependencies using copulas and 2) online extension of a recently-developed "expectation minimization of description length," which enable us to efficiently learn types of both marginal distributions and copulas as well as their parameters. The proposed method provides not only good performance in applications, but also scalable, automatic model selection, which greatly reduces the intensive modeling costs in data mining processes. We show that the proposed method outperforms state-of-the-art methods in application to density estimation and to anomaly detection.

Many multivariate Gaussianity-based techniques for identifying causal networks of observed variables have been proposed. These methods have several problems such that they cannot uniquely identify the causal networks without any prior knowledge. To alleviate this problem, a non-Gaussianity-based identification method LiNGAM was proposed. Though the LiNGAM potentially identifies a unique causal network without using any prior knowledge, it needs to properly examine independence assumptions of the causal network and search the correct causal network by using finite observed data points only. On another front, a kernel based independence measure that evaluates the independence more strictly was recently proposed. In addition, some advanced generic search algorithms including beam search have been extensively studied in the past. In this paper, we propose some variants of the LiNGAM method which introduce the kernel based method and the beam search enabling more accurate causal network identification. Furthermore, we experimentally characterize the LiNGAM and its variants in terms of accuracy and robustness of their identification.