Myeongsoo Kim

Myeongsoo Kim, Eunji Kim, Minwoo Chae et al.

Diffusion models steer conditional generation with a tunable guidance scale to trade off prompt alignment and diversity. However, existing debiasing techniques are optimized for a single scale, degrading fairness when users adjust this parameter. We trace this behavior to a previously overlooked source by decomposing total bias into two components: a model bias and a guidance bias. While prior work primarily targets the former, we show that the guidance bias grows monotonically with the guidance scale, eventually dominating the high-guidance regimes users prefer. To address this, we extend Strong Demographic Parity to guidance and derive a condition under which the target distribution retains its group ratio across guidance scales. We propose StayFair, which leverages this condition to design fair guidance algorithms in both regimes. For classifier guidance, it equalizes the classifier's output distributions across groups; for classifier-free guidance, it shifts the null embedding by a prompt-dependent offset. Because StayFair modifies only the guidance step, it is orthogonal to model debiasing and can be layered onto existing fair diffusion models to extend their fairness across guidance scales. Across class-conditional and text-to-image generation, StayFair decouples fairness from the guidance scale without sacrificing image quality.

Myeongsoo Kim, Dingmin Wang, Siwei Cui et al. · amazon-science

Code agents can autonomously resolve GitHub issues, yet when they fail, current evaluation provides no visibility into where or why. Metrics such as Pass@1 collapse an entire execution into a single binary outcome, making it difficult to identify where and why the agent went wrong. To address this limitation, we introduce TRAJEVAL, a diagnostic framework that decomposes agent trajectories into three interpretable stages: search (file localization), read (function comprehension), and edit (modification targeting). For each stage, we compute precision and recall by comparing against reference patches. Analyzing 16,758 trajectories across three agent architectures and seven models, we find universal inefficiencies (all agents examine approximately 22x more functions than necessary) yet distinct failure modes: GPT-5 locates relevant code but targets edits incorrectly, while Qwen-32B fails at file discovery entirely. We validate that these diagnostics are predictive, achieving model-level Pass@1 prediction within 0.87-2.1% MAE, and actionable: real-time feedback based on trajectory signals improves two state-of-the-art models by 2.2-4.6 percentage points while reducing costs by 20-31%. These results demonstrate that our framework not only provides a more fine-grained analysis of agent behavior, but also translates diagnostic signals into tangible performance gains. More broadly, TRAJEVAL transforms agent evaluation beyond outcome-based benchmarking toward mechanism-driven diagnosis of agent success and failure.

Myeongsoo Kim, Joe Hsu, Dingmin Wang et al.

LLM-based code agents treat repositories as unstructured text, applying edits through brittle string matching that frequently fails due to formatting drift or ambiguous patterns. We propose reframing the codebase as a structured action space where agents operate on named AST entities rather than text spans. Our framework, CODESTRUCT, provides readCode for retrieving complete syntactic units and editCode for applying syntax-validated transformations to semantic program elements. Evaluated on SWE-Bench Verified across six LLMs, CODESTRUCT improves Pass@1 accuracy by 1.2-5.0% while reducing token consumption by 12-38% for most models. Models that frequently fail to produce valid patches under text-based interfaces benefit most: GPT-5-nano improves by 20.8% as empty-patch failures drop from 46.6% to 7.2%. On CodeAssistBench, we observe consistent accuracy gains (+0.8-4.4%) with cost reductions up to 33%. Our results show that structure-aware interfaces offer a more reliable foundation for code agents.

Myeongsoo Kim, Tyler Stennett, Saurabh Sinha et al.

As modern web services increasingly rely on REST APIs, their thorough testing has become crucial. Furthermore, the advent of REST API documentation languages, such as the OpenAPI Specification, has led to the emergence of many black-box REST API testing tools. However, these tools often focus on individual test elements in isolation (e.g., APIs, parameters, values), resulting in lower coverage and less effectiveness in fault detection. To address these limitations, we present AutoRestTest, the first black-box tool to adopt a dependency-embedded multi-agent approach for REST API testing that integrates multi-agent reinforcement learning (MARL) with a semantic property dependency graph (SPDG) and Large Language Models (LLMs). Our approach treats REST API testing as a separable problem, where four agents -- API, dependency, parameter, and value agents -- collaborate to optimize API exploration. LLMs handle domain-specific value generation, the SPDG model simplifies the search space for dependencies using a similarity score between API operations, and MARL dynamically optimizes the agents' behavior. Our evaluation of AutoRestTest on 12 real-world REST services shows that it outperforms the four leading black-box REST API testing tools, including those assisted by RESTGPT (which generates realistic test inputs using LLMs), in terms of code coverage, operation coverage, and fault detection. Notably, AutoRestTest is the only tool able to trigger an internal server error in the Spotify service. Our ablation study illustrates that each component of AutoRestTest -- the SPDG, the LLM, and the agent-learning mechanism -- contributes to its overall effectiveness.

Myeongsoo Kim, Saurabh Sinha, Alessandro Orso

Modern web services rely heavily on REST APIs, typically documented using the OpenAPI specification. The widespread adoption of this standard has resulted in the development of many black-box testing tools that generate tests based on OpenAPI specifications. Although Large Language Models (LLMs) have shown promising test-generation abilities, their application to REST API testing remains mostly unexplored. We present LlamaRestTest, a novel approach that employs two custom LLMs-created by fine-tuning and quantizing the Llama3-8B model using mined datasets of REST API example values and inter-parameter dependencies-to generate realistic test inputs and uncover inter-parameter dependencies during the testing process by analyzing server responses. We evaluated LlamaRestTest on 12 real-world services (including popular services such as Spotify), comparing it against RESTGPT, a GPT-powered specification-enhancement tool, as well as several state-of-the-art REST API testing tools, including RESTler, MoRest, EvoMaster, and ARAT-RL. Our results demonstrate that fine-tuning enables smaller models to outperform much larger models in detecting actionable parameter-dependency rules and generating valid inputs for REST API testing. We also evaluated different tool configurations, ranging from the base Llama3-8B model to fine-tuned versions, and explored multiple quantization techniques, including 2-bit, 4-bit, and 8-bit integer formats. Our study shows that small language models can perform as well as, or better than, large language models in REST API testing, balancing effectiveness and efficiency. Furthermore, LlamaRestTest outperforms state-of-the-art REST API testing tools in code coverage achieved and internal server errors identified, even when those tools use RESTGPT-enhanced specifications.

Tyler Stennett, Myeongsoo Kim, Saurabh Sinha et al.

As REST APIs have become widespread in modern web services, comprehensive testing of these APIs is increasingly crucial. Because of the vast search space of operations, parameters, and parameter values, along with their dependencies and constraints, current testing tools often achieve low code coverage, resulting in suboptimal fault detection. To address this limitation, we present AutoRestTest, a novel tool that integrates the Semantic Property Dependency Graph (SPDG) with Multi-Agent Reinforcement Learning (MARL) and large language models (LLMs) for effective REST API testing. AutoRestTest determines operation-dependent parameters using the SPDG and employs five specialized agents (operation, parameter, value, dependency, and header) to identify dependencies of operations and generate operation sequences, parameter combinations, and values. Through an intuitive command-line interface, users can easily configure and monitor tests with successful operation count, unique server errors detected, and time elapsed. Upon completion, AutoRestTest generates a detailed report highlighting errors detected and operations exercised. In this paper, we introduce our tool and present preliminary findings, with a demonstration video available at https://www.youtube.com/watch?v=VVus2W8rap8.

Myeongsoo Kim, Shweta Garg, Baishakhi Ray et al. · amazon-science

Programming assistants powered by large language models have transformed software development, yet most benchmarks focus narrowly on code generation tasks. Recent efforts like InfiBench and StackEval attempt to address this gap using Stack Overflow data but remain limited to single-turn interactions in isolated contexts, require significant manual curation, and fail to represent complete project environments. We introduce CodeAssistBench (CAB), the first benchmark framework for evaluating multi-turn programming assistance in realistic settings that address real-world questions about actual codebases. Unlike existing programming Q&A benchmarks, CAB automatically generates scalable datasets from question-related GitHub issues using configurable parameters (e.g., repository creation date, star count, programming languages), and includes automatic containerization of codebases for evaluation. It then evaluates models through simulated users in these containerized environments with full codebase access. Using this framework, we constructed a test set of 3,286 real-world programming questions across 231 repositories, spanning seven programming languages and diverse problem domains. Our evaluation of leading LLMs reveals a substantial capability gap: while models perform well on Stack Overflow questions with success rates of 70-83%, they resolve only up to 16.49% of CAB's recent issues. This discrepancy highlights the challenges of providing assistance in complex, project-specific contexts versus answering standalone questions.