Line-level code completion requires a critical balance between high accuracy and low latency. Existing methods suffer from a trade-off: large language models (LLMs) provide high-quality suggestions but incur high latency, while small language models (SLMs) are fast but often suboptimal. We propose MCCom (Model-Cascading-based code Completion), a framework that cascades a local SLM with a cloud-based LLM. To achieve effective cascading, MCCom leverages user actions as a novel signal to trigger the LLM only when the SLM fails, significantly reducing cloud computation costs. Furthermore, we introduce a two-stage speculative decoding strategy and an iterative retrieval mechanism to enhance collaboration between the models. We also train a 121M-parameter lightweight model, which achieves 73.8

Evaluating software engineering agents on realistic benchmarks is costly, since each task may require multi-step code exploration, modification, and test execution. Existing efficient evaluation methods select representative subsets to estimate full-benchmark performance, but are largely result-only: they fit historical pass/fail response matrices or static task semantics, discarding how agents solve problems. We propose PTA-IRT, a Privileged Trajectory-Aware Item Response Theory framework that fuses process and outcome signals. Historical execution trajectories supply process-level evidence beyond pass/fail, such as explored context, attempted edits, and solving paths, which PTA-IRT uses as privileged information for calibration subset selection and ability estimation. Under low calibration budgets, PTA-IRT consistently outperforms prior IRT baselines on score and ranking recovery across four SWE benchmarks. Code and data are publicly available at https://github.com/DeepSoftwareAnalytics/PTA-IRT.

Code generation tasks aim to automate the conversion of user requirements into executable code, significantly reducing manual development efforts and enhancing software productivity. The emergence of large language models (LLMs) has significantly advanced code generation, though their efficiency is still impacted by certain inherent architectural constraints. Each token generation necessitates a complete inference pass, requiring persistent retention of contextual information in memory and escalating resource consumption. While existing research prioritizes inference-phase optimizations such as prompt compression and model quantization, the generation phase remains underexplored. To tackle these challenges, we propose a knowledge-infused framework named ShortCoder, which optimizes code generation efficiency while preserving semantic equivalence and readability. In particular, we introduce: (1) ten syntax-level simplification rules for Python, derived from AST-preserving transformations, achieving 18.1

Recent large language models (LLMs) have shown strong performance on software engineering tasks, yet most existing benchmarks evaluate code reasoning at the function level, where all relevant information is localized. This setting fails to reflect real-world development, which requires reasoning across multiple files and complex dependency structures. We introduce RepoReasoner, a benchmark for evaluating repository-level code reasoning. It assesses two complementary abilities: Output Prediction, which measures fine-grained, stateful execution reasoning across files, and Call Chain Prediction, which evaluates high-level architectural dependency understanding under noisy context. Our benchmark is constructed through a multi-stage pipeline that leverages dynamic tracing of pytest executions to obtain ground-truth call chains, along with LLM-based I/O rewriting to reduce memorization effects. We evaluate seven state-of-the-art LLMs. Even under oracle context, the best-performing model achieves only 69.1

The Internet is transitioning from Web3 toward Web4, where autonomous agents serve as independent economic actors. These agents can now hold crypto wallets, execute on-chain trades, and pay for external API calls. This transition calls for a new infrastructure stack capable of supporting key agent operations, including agent-to-tool interaction, agent-to-agent payments, and verifiable agent identity, represented by emerging protocols such as the Model Context Protocol, x402, and EIP-8004. Despite growing industrial interest in these protocols, the real-world Web4 agent ecosystem remains largely underexplored. To bridge this gap, we conduct the first large-scale empirical study of the Web4 ecosystem. Specifically, our study targets three interconnected questions: how Web4 agents are deployed and used in practice; what engineering challenges developers face when building Web4 agents; how current project communities respond to these challenges. To answer these questions, we analyze 99,448 multi-chain identity registrations, 317,596,323 transaction logs, the source code of 341 MCP projects, and 349 filtered GitHub issues. Our findings reveal that autonomous agents have established a highly active machine-to-machine payment economy, processing millions of daily transactions. However, this growth is built on immature infrastructure, including identity/authorization practice, cross-environment operation, and payment interoperability. Our follow-up analysis shows that community responses are visible but unevenly distributed across repositories, and payment interoperability remains the most persistent unresolved bottleneck. Overall, this study reveals a critical gap between the rapid growth of the Web4 agent economy and its fragile underlying infrastructure, highlighting future directions for building a more secure Web4 agent ecosystem.

Low-code programming (LCP) refers to programming using models at higher levels of abstraction, resulting in less manual and more efficient programming, and reduced learning effort for amateur developers. Many LCP tools have rapidly evolved and have benefited from the concepts of visual programming languages (VPLs) and programming by demonstration (PBD). With the huge increase in interest in using large language models (LLMs) in software engineering, LLM-based LCP has begun to become increasingly important. However, the technical principles and application scenarios of traditional approaches to LCP and LLM-based LCP are significantly different. Understanding these key differences and characteristics in the application of the two approaches to LCP by users is crucial for LCP providers in improving existing and developing new LCP tools and in better assisting users in choosing the appropriate LCP technology. We conducted an empirical study of both traditional LCP and LLM-based LCP. We analyzed developers' discussions on Stack Overflow (SO) over the past three years and then explored the similarities and differences between traditional LCP and LLM-based LCP features and developer feedback. Our findings reveal that while traditional LCP and LLM-based LCP share common primary usage scenarios, they significantly differ in scope, limitations, and usage throughout the software development lifecycle, particularly during the implementation phase. We also examine how LLMs impact and integrate with LCP, discussing the latest technological developments in LLM-based LCP, such as its integration with VPLs and the application of LLM Agents in software engineering.

With the rapid development of decentralized applications, many malicious actors exploit smart contract vulnerabilities for launching attacks. Moreover, as smart contracts utilize more state variables to support complex functionalities, some vulnerabilities require specific states to trigger (marked as vulnerable states), bringing new challenges to the vulnerability detection task. Although many smart contract fuzzers have been proposed for this task, they face limitations due to their inability to efficiently explore smart contract states. To address this challenge, we propose a novel fuzzer, Odyssey, with fine-grained state modeling and exploration, which increases the probability of reaching vulnerable states. We improve the efficacy of the fuzzer with two key mechanisms: (1) modeling an essential state space consisting of the variables related to sensitive operations to compress the exploration scope; (2) designing state-aware exploration strategies to identify test seeds that cover new state scope or cause new state transitions, to improve the efficiency of exploration. To evaluate the performance in vulnerability detection, we adopt Odyssey to a labeled benchmark consisting of 130 vulnerable contracts. Odyssey detects at least 70% more vulnerabilities than other fuzzers. Moreover, we evaluate Odyssey on a dataset that consists of 143 DApps (involving 437 contracts) from real-world security incidents. The experimental results demonstrate that state-aware feedback enhances the ability of Odyssey in state exploration by achieving 19% higher state coverage. Meanwhile, Odyssey totally finds 15 exploits of vulnerabilities from real-world attacks, showing its advantage in detecting real-world vulnerabilities.

Code intelligence is an emerging domain in software engineering, aiming to improve the effectiveness and efficiency of various code-related tasks. Recent research suggests that incorporating contextual information beyond the basic original task inputs (i.e., source code) can substantially enhance model performance. Such contextual signals may be obtained directly or indirectly from sources such as API documentation or intermediate representations like abstract syntax trees can significantly improve the effectiveness of code intelligence. Despite growing academic interest, there is a lack of systematic analysis of context in code intelligence. To address this gap, we conduct an extensive literature review of 146 relevant studies published between September 2007 and August 2024. Our investigation yields four main contributions. (1) A quantitative analysis of the research landscape, including publication trends, venues, and the explored domains; (2) A novel taxonomy of context types used in code intelligence; (3) A task-oriented analysis investigating context integration strategies across diverse code intelligence tasks; (4) A critical evaluation of evaluation methodologies for context-aware methods. Based on these findings, we identify fundamental challenges in context utilization in current code intelligence systems and propose a research roadmap that outlines key opportunities for future research.

Semantic code search, retrieving code that matches a given natural language query, is an important task to improve productivity in software engineering. Existing code search datasets face limitations: they rely on human annotators who assess code primarily through semantic understanding rather than functional verification, leading to potential inaccuracies and scalability issues. Additionally, current evaluation metrics often overlook the multichoice nature of code search. This paper introduces CoSQA+, pairing high-quality queries from CoSQA with multiple suitable codes. We develop an automated pipeline featuring multiple model-based candidate selections and the novel test-driven agent annotation system. Among a single Large Language Model (LLM) annotator and Python expert annotators (without test-based verification), agents leverage test-based verification and achieve the highest accuracy of 93.9%. Through extensive experiments, CoSQA+ has demonstrated superior quality over CoSQA. Models trained on CoSQA+ exhibit improved performance. We publicly release both CoSQA+-all, which contains 412,080 agent-annotated pairs, and CoSQA+-verified, which contains 1,000 human-verified pairs.

The repository-level code generation task requires synthesizing code that satisfies task requirements while remaining consistent with the target repository context. Since real-world repositories often exceed the input length limits of LLMs, existing approaches commonly adopt retrieval-augmented generation (RAG) to provide repository-specific context. Despite improving repository-context retrieval, existing methods typically provide context as task-level support, without explicitly identifying the critical tokens that require fine-grained repository context during generation. During the autoregressive generation process of LLMs, errors often concentrate at a small number of decisive positions: once such tokens are generated incorrectly, subsequent code may follow an incorrect semantic path and eventually lead to functional failure. We refer to these positions as "critical tokens". In this paper, we propose ACToR, an adaptive critical token-aware retrieval framework for repository-level code generation. ACToR identifies critical tokens during generation and triggers targeted retrieval on demand to provide repository context at these decisive positions. In addition, we design a position-aware weighting method for dense retrievers to prioritize context that is more informative for generation. We evaluate ACToR on two representative repository-level benchmarks, RepoExec and CoderEval. Experimental results show that ACToR consistently outperforms state-of-the-art methods, achieving relative improvements of 8.4

The Model Context Protocol (MCP) is an emerging standard designed to enable seamless interaction between Large Language Model (LLM) applications and external tools or resources. Within a short period, thousands of MCP services have already been developed and deployed. However, the client-server integration architecture inherent in MCP may expand the attack surface against LLM Agent systems, introducing new vulnerabilities that allow attackers to exploit by designing malicious MCP servers. In this paper, we present the first systematic study of attack vectors targeting the MCP ecosystem. Our analysis identifies four categories of attacks, i.e., Tool Poisoning Attacks, Puppet Attacks, Rug Pull Attacks, and Exploitation via Malicious External Resources. To evaluate the feasibility of these attacks, we conduct experiments following the typical steps of launching an attack through malicious MCP servers: upload-download-attack. Specifically, we first construct malicious MCP servers and successfully upload them to three widely used MCP aggregation platforms. The results indicate that current audit mechanisms are insufficient to identify and prevent the proposed attack methods. Next, through a user study and interview with 20 participants, we demonstrate that users struggle to identify malicious MCP servers and often unknowingly install them from aggregator platforms. Finally, we demonstrate that these attacks can trigger harmful behaviors within the user's local environment-such as accessing private files or controlling devices to transfer digital assets-by deploying a proof-of-concept (PoC) framework against five leading LLMs. Additionally, based on interview results, we discuss four key challenges faced by the current security ecosystem surrounding MCP servers. These findings underscore the urgent need for robust security mechanisms to defend against malicious MCP servers.

Smart contract vulnerabilities have led to billions of dollars in economic losses. Among these, improper Access Control, which allows unauthorized users to execute restricted functions, is particularly prevalent and has caused significant financial damage. Smart contract repositories contain source code, documentation, configuration files, and other artifacts necessary for building and deploying smart contracts. GitHub hosts numerous open-source repositories of this kind, which serve as intermediate artifacts in development and require compilation and packaging to produce deployable contracts. Third-party developers often reference, reuse, or fork code from these repositories during custom development. However, if the referenced code contains vulnerabilities, it can introduce significant security risks. Existing tools for detecting smart contract vulnerabilities are limited in their ability to handle such complex repositories, as they typically require the target contract to be compilable to generate an abstract representation of the program for further analysis. This paper presents TRACE, a tool designed to secure non-compilable smart contract repositories against access control vulnerabilities. TRACE employs LLMs to locate sensitive functions involving critical operations (e.g., transfer) within the contract and subsequently completes function snippets into a fully compilable contract. TRACE constructs a function call graph from the abstract syntax tree (AST) of the completed contract. It uses the control flow graph (CFG) of each function as node information. The nodes of the sensitive functions are then analyzed to detect Access Control vulnerabilities. Experimental results demonstrate that TRACE outperforms state-of-the-art tools on an open-sourced CVE dataset, detecting 14 out of 15 CVEs. In addition, it achieves 89.2% precision on 5,000 recent on-chain contracts, far exceeding the best existing tool at 76.9%. On 83 real-world repositories, TRACE achieves 87.0% precision, significantly surpassing DeepSeek-R1’s 14.3%.

Polymarket has emerged as a prominent prediction market platform and one of the fastest-growing applications in DeFi. To achieve low-latency trading, it adopts a hybrid architecture that matches orders off-chain but settles them on-chain for final execution. This design creates a consistency gap we call Ghost Fills: an order that is successfully matched off-chain may later fail during on-chain settlement. To understand the security implications of this gap, we investigate such failed settlements by building GHOSTHUNTER, which reconstructs them from on-chain traces and attributes to concrete attack patterns. Across 1,952,440 reverted match-order transactions, we find that attackers exploit the time gap between matching and settlement to invalidate already matched orders before they are finalized on-chain. We then identify four attack vectors from these incidents: nonce bump, balance drain, allowance revoke, and proxy trap, realized via 35 evolving variants. These vectors allow attackers to selectively revert 980,133 filled orders, enabling risk-free prediction, arbitrage-bot hunting, and liquidity reward manipulation, realizing at least \$1.49M in profit, which places \$1.78 B USD at risk and 2.17 M POL (about \$212 K) paid by operator. During peak hours, more than 24.3% of all filled orders reverted, causing de facto DoS attacks. We also find that code derived from the flawed contract still appears in 167 independent contracts across 10 chains holding at least \$23 M in user funds, extending the impact beyond Polymarket. We have disclosed our evidence to affected parties, and the issue has been partially mitigated.

Solidity has undergone 116 version iterations between August 2015 and February 2026, during which compiler updates have introduced behavioral changes, including issues later fixed in subsequent releases. Contracts compiled under specific versions may exhibit version-dependent execution behaviors, particularly in low-level code. These differences are often difficult for developers and users to recognize, creating opportunities for adversaries to exploit legacy compiler behaviors and deploy contracts with potentially deceptive outcomes. We define this issue as the Compiler Version Discrepancy (CVD) risk, where attackers leverage compiler-version-dependent behaviors to produce misleading or unfair outcomes while contracts appear functionally benign. We summarize five representative CVD risk patterns from real compiler inconsistencies.To mitigate this risk, we develop the CompileGuard detection tool. It combines Abstract Syntax Tree (AST) analysis, taint analysis, and symbolic execution with Control Flow Graph (CFG) analysis to identify version-sensitive code patterns. Evaluation on 227 smart contracts shows CompileGuard achieves an overall F1 score of 95.22%. A user study with 21 blockchain practitioners shows contracts exploiting CVD risks can mislead users, while detection reports enable all participants to correctly identify risk-inducing behaviors. These results highlight the practical exploitability of CVD risks and the effectiveness of automated detection in preventing such deception.

While Large Language Models have greatly advanced automated issue resolution, existing agent-based methods exhibit a fundamental limitation in their insufficient exploration of repair strategies. This insufficiency manifests in two key aspects. First, the exploration of multiple potential edit locations is limited. Second, the exploration of repair attempts at each location is also insufficient. To address these challenges, we present PhoenixRepair, a multi-agent framework that systematically explores multiple candidate edit locations and performs iterative reflection and refinement on patch generation, thereby expanding the search space of repair strategies. Our framework begins with multi-location sampling, optionally augmented with graph-based localization information for difficult tasks, followed by iterative reflection and refinement to generate better patches, culminating in final-round generation guided by distilled insights from all historical attempts. Experiments on SWE-bench-Verified demonstrate that PhoenixRepair achieves the largest relative improvement of 7.8% over SWE-agent under DeepSeek-V3.1, and attains the highest resolved rate of 76.0% Pass@1 under MiniMax-M2.5. Meanwhile, it achieves higher fault localization accuracy than existing approaches. Our code is available at https://github.com/DeepSoftwareAnalytics/PhoenixRepair.

Labelled blockchain address datasets are critical for enabling machine learning-based analytics such as address classification, risk detection, and user profiling. Existing datasets, however, suffer from limited scale, coarse-grained labels, and narrow blockchain coverage, which constrain model performance and reproducibility. To address these limitations, we construct MBAL (Multi-Blockchain Address Label), a publicly available dataset comprising 10 million labeled addresses spanning 63 address categories across four major blockchains: Bitcoin, Ethereum, Binance Smart Chain, and Polygon. MBAL integrates over 4.7 million validated labels and 5.3 million heuristically inferred labels, curated in collaboration with domain experts and covering 933 unique blockchain entities. We conduct comprehensive evaluation using eight prediction models on MBAL and the existing BABD-13 dataset. Except for Naive Bayes, all models achieve higher accuracy on MBAL, demonstrating the benefits of its larger scale, richer features, and higher label quality. MBAL further enables diverse downstream applications, illustrated through two case studies: (i) high-accuracy address label prediction on Ethereum using enriched features and XGBoost, achieving average accuracy above 80%, and (ii) graph-based risk estimation for unlabeled addresses via transaction propagation from blacklisted nodes, highlighting dominant two-hop propagation patterns.

Code Large Language Models face critical Time-To-First-Token (TTFT) latency challenges when handling long code completion due to the quadratic complexity ( O(n 2 ) ) of attention mechanisms. While existing sparse attention methods attempt to address this issue, they suffer from three key limitations: (1) general sparse patterns cause excessive accuracy degradation without considering code structure, (2) code-specific methods achieve only logical sparsity without actual computational speedup, and (3) limited adaptation to complex scenarios such as repository-level completion. We propose **SabreCoder**, a training-free **S**tructure-**a**ware **b**lock-spa**r**s**e** attention mechanism that bridges the gap between logical and computational sparsity. SabreCoder parses code into semantic chunks, constructs chunk-level sparse patterns through dependency analysis and similarity matching, and maps them to GPU-friendly block-sparse formats. Extensive experiments on LCC and CrossCodeEval benchmarks demonstrate that SabreCoder reduces TTFT by 45-55% while maintaining accuracy within 3% of dense attention.

Artificial intelligence (AI) agents are systems capable of perceiving their environment, autonomously planning and executing tasks. Recent advancements in Large Language Models (LLMs) have introduced a transformative paradigm for AI agents, enabling them to interact with external resources and tools through prompt techniques. This advancement has significantly extended the capabilities of LLMs, positioning LLM-based AI Agents as an important research area. In such agents, the workflow integrates developer-written code, which manages framework construction and logic control, with LLM-generated natural language that enhances dynamic decision-making and interaction. However, inconsistencies between LLM outputs and developer logic can lead to defects, such as tool invocation failures. These issues introduce specific risks, leading to various defects in LLM-based AI Agents, including service interruptions and incorrect output. Despite the importance of these issues, there is a lack of systematic work that focuses on analyzing LLM-based AI Agents to uncover defects in their code. To address this gap, we present the first study focused on identifying and detecting defects in LLM Agents. We collected and analyzed 14,754 relevant developer reports from StackOverflow and GitHub. We further filtered 2,604 valid posts to define and classify eight types of agent code defects. Then, we designed a static analysis tool, named Agentable, to detect these defects. Agentable leverages Code Property Graphs (CPGs) and LLMs to analyze Agent workflows by efficiently identifying specific code patterns and analyzing natural language descriptions. To evaluate Agentable, we constructed two datasets: AgentSet, which consists of 84 real-world Agent projects, and AgentTest, which contains 78 Agent projects specifically designed to include various types of defects. Our evaluation shows that Agentable achieves a precision of $88.79\%$88.79% on the real-world agent dataset and a recall of $91.03\%$91.03% on the manually labeled defect dataset. Furthermore, our analysis identifies 889 defects in real-world agent projects, highlighting the prevalence of these issues in practice.

Large Language Models (LLMs) have demonstrated remarkable capabilities in code generation, but their proficiency in producing secure code remains a critical, under-explored area. Existing benchmarks often fall short by relying on synthetic vulnerabilities or evaluating functional correctness in isolation, failing to capture the complex interplay between functionality and security found in real-world software. To address this gap, we introduce RealSec-bench, a new benchmark for secure code generation meticulously constructed from real-world, high-risk Java repositories. Our methodology employs a multi-stage pipeline that combines systematic SAST scanning with CodeQL, LLM-based false positive elimination, and rigorous human expert validation. The resulting benchmark contains 105 instances grounded in real-word repository contexts, spanning 19 Common Weakness Enumeration (CWE) types and exhibiting a wide diversity of data flow complexities, including vulnerabilities with up to 34-hop inter-procedural dependencies. Using RealSec-bench, we conduct an extensive empirical study on 5 popular LLMs. We introduce a novel composite metric, SecurePass@K, to assess both functional correctness and security simultaneously. We find that while Retrieval-Augmented Generation (RAG) techniques can improve functional correctness, they provide negligible benefits to security. Furthermore, explicitly prompting models with general security guidelines often leads to compilation failures, harming functional correctness without reliably preventing vulnerabilities. Our work highlights the gap between functional and secure code generation in current LLMs.