Code summarization aims to generate natural language (NL) descriptions for code snippets to assist developers in understanding and maintaining programs. Recent advances in large language models (LLMs) have substantially improved the quality of automatic code summarization. In real-world projects, code comments are writ...
Yi-Fei Ge, Chun-Rong Fang, Zhen-Yu Chen et al.· Proceedings of the ACM on So...· 0 citations
Retrieval-Augmented Generation (RAG) enables large language models to access external knowledge during inference, but its reliance on external knowledge bases also introduces new attack surfaces. Existing data extraction attacks against RAG systems rely on either prompt injection, which heavily depends on the generator...
Yuchen Chen, Wei Cheng, Yuan Xiao et al.· 0 citations
LiDAR perception provisioning is formulated as constructing a detector that remains accurate under point-cloud corruptions while meeting constraints on inference latency and peak GPU memory, and this method improves AP under corruption by 4.03 points on average over the corresponding original detectors.
Ting-Xu Han, Zi-Jie Shen, An Guo et al.· 0 citations
Experimental results show that Jev integration substantially reduces decision cost and latency across all five agents, with task-level gains varying by workflow, and highlight the importance of preserving progress, supporting recovery, and verifying completion when integrating efficient decision models into agent workf...
Yu-Chen Ling, Sheng-Cheng Yu, Zhen-Yu Chen et al.· 2 citations
It is argued that AIS security must be assessed at the level of the whole software system, and the problems arising from this coexistence into four categories: context misuse, authorization violation, execution control, and effect integrity are organized.
A systematic empirical study of harness effects in SE agents shows that harness effectiveness depends jointly on model capability and task type, and highlights harness design as a first-class factor in SE-agent performance and provides insights for building more effective and efficient coding agents.
Hai-Chuan Hu, Quan-Jun Zhang, Sheng-Cheng Yu et al.· 4 citations
CacheReforge, which represents stale KV caches as layerwise mixed-version objects, combines per-layer adapter anchors, calibrated sensitivity, accumulated drift, and executable restart boundaries to select direct reuse, bounded recomputation, or complete affected-suffix recovery.
Yu-Hang Cao, Yan-Zhou Mu, Chun-Rong Fang et al.· 0 citations
This paper introduces Agent-Integrated Software (AIS) as a software pattern combining a conventional core, direct interaction, and a built-in agent, and Intent-Level Interaction Abstraction (IIA) as the task semantics through which users inspect and control delegated work.
Sheng-Cheng Yu, C. Fang, Zhen-Yu Chen· arXiv.org· 2 citations
The results show that reliable malicious-Skill detection requires both broader cross-source benchmark coverage and evaluation that jointly measures attack detection and benign over-flagging.
BASM is proposed, which augments each skill with explicit boundary fields, which transforms each retrieved skill from an unconditional action template into state-conditioned guidance: the agent applies the skill when its conditions hold, suppresses inapplicable tool calls when they do not, and issues targeted repairs w...
Zi-Han Lin, Zhen-Yu Chen, Jiawen Wei et al.· 1 citation
This paper proposes ARIA, an automated red-teaming framework for crafting covert and effective backdoored instructions against customized LLMs, which significantly outperforms existing attacks in evading platform-side and user-side detection, and stays effective against existing defense methods.
Yuchen Chen, Wei Cheng, Yuan Xiao et al.· Proceedings of the 41st IEEE...· 0 citations
ReProAgent is a multi-stage agent framework for reproduction test generation from issue reports that decomposes the task into four agent stages: bug localization, root cause analysis, test planning, and test generation, and generalizes across multiple backbone LLMs.
Quanjun Zhang, Yi Zheng, Ye Shang et al.· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.