arXiv Machine Learning August 4, 2026 By Ruilin Xu, Junyi Li, Pengfei Chen, Zongxuan Xie

TELLER: Non-intrusive Cross-Layer Root-Cause Analysis for LLM Inference

Read the original on arXiv Machine Learning →

arXiv:2608. 01975v1 Announce Type: cross Abstract: Large language model (LLM) inference has evolved from an offline workload into a continuously operated software service, yet root-cause analysis remains difficult because a single request spans the inference engine, Python/C++ backend, host CUDA APIs, GPU kernels, and distributed communication.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv Machine Learning.

arXiv AI

Jun 3

StepFinder: A Temporal Semantic Framework for Failure Attribution in Multi-Agent Systems

arXiv:2606. 03467v1 Announce Type: new Abstract: LLM-based multi-agent systems exhibit remarkable collaborative capabilities in complex multi-step tasks.

By Taiyu Zhu, Yifan Wu, Weilin Jin, Ying Li, Gang Huang

llms agents benchmarks safety

arXiv AI

Aug 11

TRACE: TRajectory Attribution for Automated Context Engineering

arXiv:2608. 09153v1 Announce Type: new Abstract: Production AI agents fail when their context sources -- system prompts, knowledge bases, tool descriptions, and procedural skills -- contain errors or gaps.

By Yikai Zhao, Pradeep Kumar Misra, Saurabh Pandey

agents fine-tuning benchmarks

arXiv AI

Jul 7

MemTrace: Tracing and Attributing Errors in Large Language Model Memory Systems

arXiv:2605. 28732v2 Announce Type: replace-cross Abstract: Memory is essential for enabling large language models to support long-horizon reasoning, yet existing memory systems remain unreliable and difficult to debug.

By Xinle Deng, Ruobin Zhong, Hujin Peng, Xiaoben Lu, Yanzhe Wu, Guang Li, Buqiang Xu, Yunzhi Yao, Jizhan Fang, Haoliang Cao, Junjie Guo, Yuan Yuan, Ziqing Ma, Yuanqiang Yu, Rui Hu, Baohua Dong, Hangcheng Zhu, Ningyu Zhang

llms rag benchmarks

arXiv AI

Jun 16

Knowledge-Based Zero-Replay Debugging of Multi-Agent LLM Traces

arXiv:2606. 14805v1 Announce Type: cross Abstract: Reliable operation of multi-agent large language model (LLM) systems depends on debugging long execution traces, where the few causally decisive events are buried in unstructured logs of messages, routes, memory writes, and tool calls.

By Dong Ho Kang, Hyeonjeong Cha, Daein Weon

llms agents

arXiv AI

Jun 30

Mechanistically Eliciting Latent Behaviors in Language Models

arXiv:2606. 29604v1 Announce Type: cross Abstract: We aim to discover diverse, generalizable perturbations of LLM internals that can surface hidden behavioral modes.

By Andrew Mack, Nina Panickssery, Alexander Matt Turner

llms safety

arXiv AI

Jun 24

VeriPilot: An LLM-Powered Verilog Debugging Framework

arXiv:2606. 23759v1 Announce Type: cross Abstract: Verilog debugging remains one of the most time-consuming stages in digital circuit design.

By Yihan Wang, Cheng Liu, Jiazheng Zhang, Lei Zhang, Long Cheng, Xiaowei Li, Huawei Li

llms benchmarks

Related stories

StepFinder: A Temporal Semantic Framework for Failure Attribution in Multi-Agent Systems

TRACE: TRajectory Attribution for Automated Context Engineering

MemTrace: Tracing and Attributing Errors in Large Language Model Memory Systems

Knowledge-Based Zero-Replay Debugging of Multi-Agent LLM Traces

Mechanistically Eliciting Latent Behaviors in Language Models

VeriPilot: An LLM-Powered Verilog Debugging Framework