The issue-resolving task, where a model generates patches to fix real-world bugs, has emerged as a key benchmark for evaluating the capabilities of large language models (LLMs). While SWE-bench has become the dominant benchmark in this domain, it suffers from several limitations: it has not been updated since its release, is restricted to only 12 repositories, and relies heavily on manual effort for constructing test instances and setting up executable environments, significantly limiting its scalability. We present SWE-bench-Live, a live-updatable benchmark designed to address these limitations. SWE-bench-Live currently includes 1,890 tasks derived from real GitHub issues created since 2024, spanning 223 repositories. Each task is accompanied by a dedicated Docker image to ensure reproducible execution. Additionally, we introduce an automated curation pipeline that streamlines the entire process from instance creation to environment setup, removing manual bottlenecks and enabling scalability and continuous updates. We evaluate a range of state-of-the-art models and agent frameworks on SWE-bench-Live, offering detailed empirical insights into their real-world bug-fixing capabilities. By providing a fresh, diverse, and executable benchmark grounded in live repository activity, SWE-bench-Live supports reliable, large-scale assessment of code LLMs and code agents in realistic development settings.

Software version migration and program translation are an important and costly part of the lifecycle of large codebases.Traditional machine translation relies on parallel corpora for supervised translation, which is not feasible for program translation due to a dearth of aligned data.Recent unsupervised neural machine translation techniques have overcome data limitations by included techniques such as back translation and low level compiler intermediate representations (IR).These methods face significant challenges due to the noise in code snippet alignment and the diversity of IRs respectively.In this paper we propose a novel model called Code Distillation (CoDist) whereby we capture the semantic and structural equivalence of code in a language agnostic intermediate representation.Distilled code serves as a translation pivot for any programming language, leading by construction to parallel corpora which scale to all available source code by simply applying the distillation compiler.We demonstrate that our approach achieves state-of-the-art performance on CodeXGLUE and TransCoder GeeksForGeeks translation benchmarks, with an average absolute increase of 12.7% on the TransCoder GeeksforGeeks translation benchmark compare to TransCoder-ST.

Program translation, i.e. transcompilation has been attracting increasing attention from researchers due to its enormous application value.However, we observe that current program translating models still make elementary syntax errors, particularly when the source language uses syntax elements not present in the target language, which is exactly what developers are concerned about while may not be well exposed by frequently used metrics such as BLEU, CodeBLEU and Computation Accuracy.In this paper, we focus on evaluating the model's ability to address these basic syntax errors and developed an novel active defects probing suite, the Syntactic Unit Tests (SUT) and highly interpretable evaluation harness including Syntax Unit Test Accuracy (SUT Acc) metric and Syntax Element Test Score (SETS), to help diagnose and promote progress in this area.Our Syntactic Unit Test fills the gap in the community for a fine-grained evaluation dataset for program translation.Experimental analysis shows that our evaluation harness is more accurate, reliable, and in line with human judgments compared to previous metrics.

Attention mechanisms achieve promising performance in text classification. When classifying the sentence into a certain category among multiple candidates, the existing attention utilizes a unified attention weight vector to determine contribution of each word. We observe that it is not accurate enough for a single vector to capture diverse category features of a sentence. In this paper, we propose a Category Expert Attention matrix to extract sentence features from different category perspectives. Each row of the attention matrix represents the unique feature of each category. We evaluate our model on four large scale datasets. Empirical results show that our model outperforms previous baseline methods by a significant margin. Especially on the NLPCC dataset, we achieve new state-of-the-art results. We also conduct visualization experiments to interpret how category expert information analyzes sentences.