Pith. sign in

REVIEW 18 cited by

A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deployment

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2504.15585 v4 pith:NY7F2D2L submitted 2025-04-22 cs.CR cs.AIcs.CLcs.LG

classification cs.CRcs.AIcs.CLcs.LG
keywords safetycomprehensivedeploymentllmsdataentirelifecycleresearch
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

The remarkable success of Large Language Models (LLMs) has illuminated a promising pathway toward achieving Artificial General Intelligence for both academic and industrial communities, owing to their unprecedented performance across various applications. As LLMs continue to gain prominence in both research and commercial domains, their security and safety implications have become a growing concern, not only for researchers and corporations but also for every nation. Currently, existing surveys on LLM safety primarily focus on specific stages of the LLM lifecycle, e.g., deployment phase or fine-tuning phase, lacking a comprehensive understanding of the entire "lifechain" of LLMs. To address this gap, this paper introduces, for the first time, the concept of "full-stack" safety to systematically consider safety issues throughout the entire process of LLM training, deployment, and eventual commercialization. Compared to the off-the-shelf LLM safety surveys, our work demonstrates several distinctive advantages: (I) Comprehensive Perspective. We define the complete LLM lifecycle as encompassing data preparation, pre-training, post-training, deployment and final commercialization. To our knowledge, this represents the first safety survey to encompass the entire lifecycle of LLMs. (II) Extensive Literature Support. Our research is grounded in an exhaustive review of over 800+ papers, ensuring comprehensive coverage and systematic organization of security issues within a more holistic understanding. (III) Unique Insights. Through systematic literature analysis, we have developed reliable roadmaps and perspectives for each chapter. Our work identifies promising research directions, including safety in data generation, alignment techniques, model editing, and LLM-based agent systems. These insights provide valuable guidance for researchers pursuing future work in this field.

Discussion (0). Sign in to comment.

Forward citations

Cited by 18 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Persistent Sparse Autoencoders: Learning Feature Timescales in Language Models

    cs.LG 2026-07 conditional novelty 6.0 of 10

    Persistent SAEs learn per-feature persistence coefficients from reconstruction, splitting features into fast local detectors and slow topic-tracking states that retain prompt-injection signals over long contexts.

  2. Temporary Authority, Permanent Effects: Commit-Time Authorization for LLM Agents

    cs.CR 2026-07 conditional novelty 6.0 of 10

    Under controlled authority invalidation, agents reach visible goals in 262/270 runs but only 55 remain authorized; CommitGuard blocks stale durable effects when runtimes emit boundary signals.

  3. Breaking the Rounding Trap: Securing LLMs against Quantization-Conditioned Backdoors

    cs.CR 2026-06 unverdicted novelty 6.0 of 10

    QuantGuard uses differentiable rounding control to break quantization-boundary backdoors in LLMs, reducing post-quantization attack success to clean-model levels across six models and INT8/FP4/NF4.

  4. ProbeLLM: Automating Principled Diagnosis of LLM Failures

    cs.CL 2026-02 conditional novelty 6.0 of 10

    An automated LLM probing pipeline using hierarchical Monte Carlo Tree Search, tool-based verification, and failure clustering to discover structured failure modes beyond static benchmarks.

  5. The Alignment Curse: Modality Alignment Supercharges Audio Attacks via Text Transfer

    cs.LG 2026-01 conditional novelty 6.0 of 10

    Text jailbreak prompts converted to audio match or beat dedicated audio jailbreaks on omni-models, and transfer success tracks how tightly the model aligns text and audio representations.

  6. Check Yourself Before You Wreck Yourself: Selectively Quitting Improves LLM Agent Safety

    cs.CL 2025-10 conditional novelty 6.0 of 10

    Explicitly instructing LLM agents to quit uncertain or risky tasks improves their safety scores by 0.39 on average in ToolEmu simulations, with a negligible average helpfulness decrease of 0.03.

  7. Uncovering Vulnerabilities of LLM-Assisted Cyber Threat Intelligence

    cs.CR 2025-09 conditional novelty 6.0 of 10

    LLMs assisting cyber threat intelligence fail mainly due to spurious correlations, contradictory knowledge, and constrained generalization that stem from the threat landscape itself.

  8. KubeGuard: LLM-Assisted Kubernetes Hardening via Configuration Files and Runtime Logs Analysis

    cs.CR 2025-09 conditional novelty 6.0 of 10

    KubeGuard generates and refines Kubernetes Roles, NetworkPolicies, and Deployments from aggregated audit, network, and provenance logs using prompt-chained LLMs, achieving F1 up to 0.96 with GPT-4o.

  9. Adaptive Backtracking for Privacy Protection in Large Language Models

    cs.CR 2025-08 conditional novelty 6.0 of 10

    A training-free backtracking defense that rewrites RAG output at the first sign of privacy leakage improves privacy utility scores by up to 15% over sanitization and prompting baselines on a new healthcare and finance...

  10. SafeMobile: Chain-level Jailbreak Detection and Automated Evaluation for Multimodal Mobile Agents

    cs.AI 2025-07 conditional novelty 6.0 of 10

    A history-aware guard model with an LLM judge is reported to cut jailbreak success on mobile agent tasks from 86.1% to 8.4% while keeping task completion unchanged at 77.8%.

  11. Agentic Services Computing

    cs.SE 2025-09 conditional novelty 5.0 of 10

    A position and survey paper that defines Agentic Services Computing, a lifecycle-based framework for engineering LLM agents as governed, first-class services.

  12. Introducing the Swiss Food Knowledge Graph: AI for Context-Aware Nutrition Recommendation

    cs.AI 2025-07 conditional novelty 5.0 of 10

    The paper introduces SwissFKG, a knowledge graph integrating Swiss recipes, nutrients, allergens, and dietary guidelines, populated via an LLM pipeline and used for a Graph-RAG question answering demo.

  13. We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems

    cs.LG 2025-06 conditional novelty 5.0 of 10

    MCP-powered LLM agents are vulnerable to prompt injection from third-party services, and simple detection or filtering defenses do not reliably stop these attacks.

  14. Securing Multi-Agent Systems Against Corruptions via Node Contribution Backpropagation

    cs.CR 2025-10 conditional novelty 4.0 of 10

    A backward-propagation scoring scheme over a signed temporal DAG can identify malicious agents in LLM multi-agent systems and cut their communications, improving defended accuracy by 3–7 percentage points in the autho...

  15. Large Language Models for Next-Generation Wireless Network Management: A Survey and Tutorial

    cs.NI 2025-09 conditional novelty 4.0 of 10

    A survey and tutorial that organizes LLM-enabled wireless network optimization into formulation, solution, and verification stages, with case studies drawn from the authors' own prior papers.

  16. A Comprehensive Survey on Trustworthiness in Reasoning with Large Language Models

    cs.CL 2025-09 conditional novelty 4.0 of 10

    A structured literature survey concluding that reasoning capabilities do not automatically make LLMs more trustworthy and can introduce new vulnerabilities in safety, robustness, and privacy.

  17. Collaborative Inference and Learning between Edge SLMs and Cloud LLMs: A Survey of Algorithms, Execution, and Open Challenges

    cs.DC 2025-07 conditional novelty 4.0 of 10

    A survey that builds a taxonomy of edge-cloud LLM-SLM collaboration for inference and training, claiming to be the first to unify both phases.

  18. Graph-Augmented Large Language Model Agents: Current Progress and Future Prospects

    cs.AI 2025-07 conditional novelty 3.0 of 10

    A survey that categorizes Graph-augmented LLM Agent research into planning, memory, tool management, and multi-agent design, and outlines open directions.

Pith tools