相关论文: LastMerge: A language-agnostic structured tool for…

Model merging aims to combine multiple task-specific expert models into a single model while preserving generalization across diverse tasks. However, interference among experts, especially when they are trained on different objectives,…

While most forks on platforms like GitHub are short-lived and used for social collaboration, a smaller but impactful subset evolve into long-lived forks, referred to here as variants, that maintain independent development trajectories.…

Deployed language models must produce outputs that are both correct and format-compliant. We study this structured-output reliability gap using two mathematical benchmarks -- GSM8K and MATH -- as a controlled testbed: ground truth is…

In this paper, we propose StruM, a novel structured mixed-precision-based deep learning inference method, co-designed with its associated hardware accelerator (DPU), to address the escalating computational and memory demands of deep…

Executing large test suites is time and resource consuming, sometimes impossible, and such test suites typically contain many redundant test cases. Hence, test case minimization is used to remove redundant test cases that are unlikely to…

This study proposes a simple yet effective LoRA merge method to achieve LLM adaptation for low-resource language generation tasks. The LoRA merge technique, which integrates multiple LoRA modules trained on different tasks, has gained…

Programs expecting structured inputs often consist of both a syntactic analysis stage, which parses raw input, and a semantic analysis stage, which conducts checks on the parsed input and executes the core logic of the program.…

To remain useful for their users, software systems need to continuously enhance and extend their functionality. Nevertheless, in many object-oriented applications, features are not represented explicitly. The lack of modularization is known…

While most current approaches rely on further training techniques, such as fine-tuning or reinforcement learning, to enhance model capacities, model merging stands out for its ability of improving models without requiring any additional…

Concerns about benchmark leakage in large language models for code (Code LLMs) have raised issues of data contamination and inflated evaluation metrics. The diversity and inaccessibility of many training datasets make it difficult to…

Mathematical reasoning is an important research direction in the field of artificial intelligence. This article proposes a novel multi tool application framework for mathematical reasoning, aiming to achieve more comprehensive and accurate…

Large Language Models (LLMs) are frequently used for multi-faceted language generation and evaluation tasks that involve satisfying intricate user constraints or taking into account multiple aspects and criteria. However, their performance…

With little to no parallel data available for programming languages, unsupervised methods are well-suited to source code translation. However, the majority of unsupervised machine translation approaches rely on back-translation, a method…

In software practice, static analysis tools remain an integral part of detecting defects in software and there have been various tools designed to run the analysis in different programming languages like Java, C++, and Python. This paper…

Low-Rank Adaptation (LoRA) is a popular technique for parameter-efficient fine-tuning of Large Language Models (LLMs). We study how different LoRA modules can be merged to achieve skill composition -- testing the performance of the merged…

Achieving high-performing language models which include medium- and lower-resource languages remains a challenge. Massively multilingual models still underperform compared to language-specific adaptations, especially at smaller model…

We study model merging as a practical alternative to conventional adaptation strategies for code-mixed NLP. Starting from a multilingual base model, we: (i) perform continued pre-training (CPT) on unlabeled code-mixed text to obtain an…

Code translation aims to transform code between programming languages while preserving functionality, with applications in cross-platform development and software migration. Recent advances in Large Language Models (LLMs) have improved code…

Continual learning (CL) is essential for deploying large language models (LLMs) in dynamic real-world environments without the need for costly retraining. Recent model merging-based methods have attracted significant attention, but they…

Large Language Models (LLMs) are increasingly used by software engineers for code generation. However, limitations of LLMs such as irrelevant or incorrect code have highlighted the need for prompt programming (or prompt engineering) where…