Research topic

LLM-Assisted Systems Analysis Research

This topic groups work that uses large language models to query, summarize, reason over, or govern system evidence. The papers distinguish the trace or structured representation provided to the model, the reasoning task, evaluation data, and the limits of language-model assistance for technical diagnosis.

Related search terms: LLM systems analysis · large language models for tracing · LLM-assisted observability

5 papers in this topic, ordered newest first. The detailed paper records contain the evidence-grounded methods, tools, datasets, findings, and citation guidance.

Selected papers

2026 · Journal of Systems and Software

LMAT: An adaptive tracing approach based on efficient system behavior analysis using language models

Kasra Darvishi, Morteza Noferesti, Yuvraj Sehgal, Naser Ezzati-Jivan

LMAT combines multi-task language models for kernel-event and event-duration prediction with online change detection, lightweight error-vector root-cause analysis, and adaptive tracing control.

Keywords: adaptive tracing · LTTng · kernel events · system-call sequences · event-duration modeling

Read the detailed paper record · · Authoritative source

2026 · Proceedings of the 34th ACM International Conference on the Foundations of Software Engineering (FSE)

Observability and Runtime Governance for Agentic AI Systems

Naser Ezzati-Jivan, Maryam Ekhlasi

The tutorial presents an end-to-end AgentOps workflow that connects task intent and model decisions to tool calls, memory access, inter-agent communication, external side effects, and runtime-governance decisions.

Keywords: agentic AI · AgentOps · runtime governance · AI observability · software agents

Read the detailed paper record · · Authoritative source

2026 · SIGSOFT FSE Companion

Rethinking Performance Debugging: From Optimization to Collaborative Reasoning

Mahsa Panahandeh, Naser Ezzati-Jivan, Abdelwahab Hamou-Lhadj

The paper reframes performance debugging as collaborative reasoning over multiple evidence-grounded hypotheses rather than optimization for one supposedly best explanation.

Keywords: performance debugging · collaborative reasoning · AgentDebug · Reasoning Surface · hypothesis generation

Read the detailed paper record · · Authoritative source

2026 · arXiv / CoRR; arXiv page reports acceptance to ICSE 2026

TAAF: A Trace Abstraction and Analysis Framework Synergizing Knowledge Graphs and LLMs

Alireza Ezaz, Ghazal Khodabandeh, Majid Babaei, Naser Ezzati-Jivan

TAAF combines time-indexed knowledge graphs with large language models to turn large execution traces into queryable, actionable explanations.

Keywords: TAAF · TraceQA-100 · time-indexed knowledge graph · natural-language trace queries · multi-hop reasoning

Read the detailed paper record · · Authoritative source

2024 · 2024 IEEE/ACM International Conference on Software Engineering: New Ideas and Emerging Results (ICSE-NIER)

Toward Adaptive Tracing: Efficient System Behavior Analysis using Language Models

Kasra Darvishi, Morteza Noferesti, Naser Ezzati-Jivan

The paper proposes an adaptive tracing loop that combines language-model prediction of kernel-event sequences and event durations with change detection and root-cause analysis, so detailed tracing is activated only around significant behavior shifts.

Keywords: adaptive tracing · LTTng · kernel events · system-call sequences · event-duration modeling

Read the detailed paper record · · Authoritative source

Related topics