Model Debugging

Follow this topic Ask about this topic

Momentum

0 papers in the last four weeks, with none the four weeks before. 0.0% of all new papers.

Jul 13Week of Sep 28

Latest papers 14

All topics

Don't Regenerate, Debug: A Domain-Specific Agent for Repairing Near-Miss Hardware Operators
Aug 3, 2026Yansong Sun, Shenxiu Wu, Siyuan Chen +6Model Debugging Repair
Department of Computer Science, City University of Hong Kong · Application Software Engineering Lab, Huawei Technologies Ltd.
CUADebug: Diagnosing and Repairing Computer-Use Agent Failures
Jul 31, 2026Weijia Zhang, Kunlun Zhu, Zeyi Liu +8Model Debugging Computer-Use Agents
University of Illinois Urbana-Champaign · 2Yale University
AgentDebugX: An Open-Source Toolkit for Failure Observability, Attribution, and Recovery in LLM Agents
Jul 21, 2026Kunlun Zhu, Xuyan Ye, Zhiguang Han +9Model Debugging Large Language Model Agents
University of Illinois Urbana-Champaign · University of Toronto · Google +1
KnowledgeDebugger -- an Exploration Tool for Knowledge Localization and Editing in Transformers
Jul 1, 2026Eric Benz, Lennart Stöpler, Nikolai Bolik +1Multimodal Knowledge Editing Cross-Task Knowledge Transfer
Heidelberg University
VeriPilot: An LLM-Powered Verilog Debugging Framework
Jun 22, 2026Yihan Wang, Cheng Liu, Jiazheng Zhang +4Model Debugging High-Level Synthesis
State Key Lab of Processors, Institute of Computing Technology, Chinese Academy of Sciences, and the University of Chinese Academy of Sciences, Beijing, China · North China Electric Power University, Beijing 102206, China
DDOR: Delta Debugging for Explainable Overrefusal Testing and Repair
Jun 2, 2026Qinyan Zhou, Peixin Zhang, Jun Sun +2Large Language Model Safety Refusals
Southeast University, China · School of Computing and Information Systems, Singapore Management University, Singapore · Zhejiang University, China +1
ModuLoop : Low-Level Code Generation using Modular Synthesizer and Closed-Loop Debugger for Robotic Control
Jun 2, 2026Gina Yoon, Sumin Lee, Joo Yong SimCode Generation Real-World Manipulation Tasks
Department of Mechanical Systems Engineering, Sookmyung Women’s University, Cheongpa-ro 47-gil 100, Yongsan-gu, Seoul, 04310, South Korea
LogDx-CI: Benchmarking Log Reduction Tools for LLM Root-Cause Diagnosis
May 26, 2026Bowen QinModel Debugging Code Quality
National University of Singapore
DiagEval: Trajectory-Conditioned Diagnosis for Reliable Software Evaluation with GUI Agents
May 17, 2026Sirui Hong, Zhijie Liu, Tengfei Li +3Model Debugging Diagnosis
1DeepWisdom · 2Independent Researcher · 3The Hong Kong University of Science and Technology (Guangzhou)
Debugging the Debuggers: Failure-Anchored Structured Recovery for Software Engineering Agents
May 9, 2026Chenyu Zhao, Shenglin Zhang, Yihang Lin +7Model Debugging Diagnosis
Nankai University Tianjin, China · Tsinghua University Beijing, China · Microsoft Redmond, USA
CUDABeaver: Benchmarking LLM-Based Automated CUDA Debugging
May 8, 2026Shiyang Li, Haoyang Chen, Mattia Fazzini +1Model Debugging Cuda
University of Minnesota
VeriLLMed: Interactive Visual Debugging of Medical Large Language Models with Knowledge Graphs
Apr 25, 2026Yurui Xiang, Xingyi Mao, Rui Sheng +7Model Debugging LLM Reasoning Strategies
Seeing the Whole Elephant: A Benchmark for Failure Attribution in LLM-based Multi-Agent Systems
Apr 24, 2026Mengzhuo Chen, Junjie Wang, Fangwen Mu +4Latent Failure Patterns Model Debugging
State Key Laboratory of Complex System Modeling and Simulation Technology · Institute of Software, Chinese Academy of Sciences · University of Chinese Academy of Sciences, Beijing, China
Precise Debugging Benchmark: Is Your Model Debugging or Regenerating?
Apr 19, 2026Miaosen Chai, Wang Bill Zhu, Shangshang Wang +5Model Debugging Large Language Model Benchmarks
University of Southern California · Microsoft · University of Wisconsin–Madison +1

Aug 3, 2026, cs.SEJ/K move · Enter open · S save

Don't Regenerate, Debug: A Domain-Specific Agent for Repairing Near-Miss Hardware Operators

Yansong Sun, Shenxiu Wu, Siyuan Chen, Runlin Hou+5

Department of Computer Science, City University of Hong Kong · Application Software Engineering Lab, Huawei Technologies Ltd.

Model Debugging Repair Edge Hardware LLM Inference Optimization Generators

Open paper AskPDF arXiv

Kernel generation for hardware accelerators such as GPUs and NPUs has become a proving ground for large language models (LLMs), and state-of-the-art systems raise correctness through pipelines that couple LLMs with agentic reinforcement learning and evolutionary search. Such pipelines generate, compile, and execute large numbers of candidate kernels, discarding most of them and forgoing the opportunity to distill failures into reusable knowledge. Many discarded candidates are near-miss operators that compile and run but fail numerical validation; each embodies genuine domain knowledge and a nontrivial investment in LLM inference, cross-compilation, and hardware execution. We argue for a paradigm shift: rather than regenerate, debug. Debugging is far more constrained than generating from scratch: the search space is small and feedback is dense. We present a domain-specific debug agent that addresses three core challenges in autonomous repair: mitigating knowledge scarcity through retrieved patterns and diagnostic instrumentation, ensuring integrity through anti-cheat detection and full-coverage evaluation, and controlling cost via convergence guards and bounded iteration. Debugging serves two complementary roles: it extends the capability frontier by recovering operators that repeated regeneration fails to produce, and it lowers cost per deliverable operator. Debug Pass@1 achieves 66.7% versus Regenerate Avg Pass@1's 25.9% and Regenerate Pass@3's 40.7%, while consuming 92.8% fewer tokens per success than three-trial regeneration. Component ablations show that the knowledge base drives recovery, while integrity gates reject 12.5-33.3% of the successes the workflow itself accepted.

Momentum

Related topics

Latest papers 14