Pith. sign in

REVIEW 16 cited by

Building Guardrails for Large Language Models

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 2402.01822 v2 pith:L6I2QLL4 submitted 2024-02-02 cs.CL cs.AI

Building Guardrails for Large Language Models

classification cs.CL cs.AI
keywords guardrailsllmsbuildinglanguagelargemodelsrequirementsrisks
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

As Large Language Models (LLMs) become more integrated into our daily lives, it is crucial to identify and mitigate their risks, especially when the risks can have profound impacts on human users and societies. Guardrails, which filter the inputs or outputs of LLMs, have emerged as a core safeguarding technology. This position paper takes a deep look at current open-source solutions (Llama Guard, Nvidia NeMo, Guardrails AI), and discusses the challenges and the road towards building more complete solutions. Drawing on robust evidence from previous research, we advocate for a systematic approach to construct guardrails for LLMs, based on comprehensive consideration of diverse contexts across various LLMs applications. We propose employing socio-technical methods through collaboration with a multi-disciplinary team to pinpoint precise technical requirements, exploring advanced neural-symbolic implementations to embrace the complexity of the requirements, and developing verification and testing to ensure the utmost quality of the final product.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 16 Pith papers

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

  1. Robotics-Inspired Guardrails for Foundation Models in Socially Sensitive Domains

    cs.AI 2026-05 unverdicted novelty 7.0

    Introduces the Grounded Observer framework that applies robotics-inspired formal constructs for runtime constraint enforcement on foundation model interaction trajectories in socially sensitive domains.

  2. Skill Drift Is Contract Violation: Proactive Maintenance for LLM Agent Skill Libraries

    cs.SE 2026-05 conditional novelty 7.0

    SkillGuard extracts executable environment contracts from LLM skill documents to detect only relevant drifts, reporting zero false positives on 599 cases, 100% precision in known-drift tests, and raising one-round rep...

  3. Misrouter: Exploiting Routing Mechanisms for Input-Only Attacks on Mixture-of-Experts LLMs

    cs.CR 2026-05 unverdicted novelty 7.0

    Misrouter enables input-only attacks on MoE LLMs by optimizing queries on open-source surrogates to route toward weakly aligned experts and transferring them to public APIs.

  4. Playing Along: Learning a Double-Agent Defender for Belief Steering via Theory of Mind

    cs.CL 2026-04 unverdicted novelty 7.0

    RL-trained AI double agents using combined ToM and fooling rewards outperform prompted frontier models on a new belief-steering task and show bidirectional emergence between the two skills.

  5. Playing Along: Learning a Double-Agent Defender for Belief Steering via Theory of Mind

    cs.CL 2026-04 conditional novelty 7.0

    In a new ToM-based privacy game, RL training to fool an attacker also improves the defender's theory-of-mind accuracy, and training for ToM also improves fooling; a combined reward gives the best results.

  6. Asymmetric Communication: Large Language Models and Language Games

    cs.CY 2026-07 conditional novelty 6.5

    Human–LLM exchange is asymmetric communication: model outputs circulate without commitments, so AGI, hallucination, agency, sentience, and alignment are receiver-side category mistakes, and alignment is institutional ...

  7. Understanding Annotator Safety Policy with Interpretability

    cs.AI 2026-05 unverdicted novelty 6.0

    Annotator Policy Models learn safety policies from labeling behavior alone, accurately predicting responses and revealing sources of disagreement like policy ambiguity and value pluralism.

  8. LocalAlign: Enabling Generalizable Prompt Injection Defense via Generation of Near-Target Adversarial Examples for Alignment Training

    cs.CR 2026-05 unverdicted novelty 6.0

    LocalAlign generates near-target adversarial examples via prompting and applies margin-aware alignment training to enforce tighter boundaries against prompt injection attacks.

  9. Agent-Sentry: Bounding LLM Agents via Execution Provenance

    cs.CR 2026-03 unverdicted novelty 6.0

    Agent-Sentry bounds LLM agent executions via structural provenance classification, sensitive-value allowlists, and selective LLM judgment, blocking 94.3% of injections while allowing 95.1% of benign actions on AgentDo...

  10. StarCoder 2 and The Stack v2: The Next Generation

    cs.SE 2024-02 accept novelty 6.0

    StarCoder2-15B matches or beats CodeLlama-34B on code tasks despite being smaller, and StarCoder2-3B outperforms prior 15B models, with open weights and exact training data identifiers released.

  11. From Prompts to Contracts: Harness Engineering for Auditable Enterprise LLM Agents

    cs.AI 2026-07 conditional novelty 5.5

    A code-owned harness enforces source, routing, trace, hygiene, and recommendation contracts for enterprise LLM agents; prompt-only fails and bolt-on guardrails over-refuse.

  12. Safeguards for Speech2Speech LLM-Assistants: A Case Study in Automotive Applications

    cs.AI 2026-07 conditional novelty 5.0

    Across three speech-to-speech APIs, tool-based guardrails add 0.6-1.4s and fail a 0.4s conversational-delay target; only Amazon Nova Sonic's transcript-based guardrails stayed under it for both input types.

  13. Neuro-Symbolic Verification of LLM Outputs for Data-Sensitive Domains (extended preprint)

    cs.AI 2026-05 unverdicted novelty 5.0

    Neuro-symbolic pipeline using formal logic and semantic embeddings detects hallucinations in LLM medical reports at 83%+ for entities and 72% for fabrications while cutting creation time 30%.

  14. Fence: Specialized SLM Guardrails for LLM Applications

    cs.AI 2026-05 conditional novelty 5.0

    Small language models trained on GAN-style adversarial synthetic data outperform prompted large models on use-case-specific guardrails for off-topic and prompt-injection detection.

  15. Trustworthy Agent Network: Trust in Agent Networks Must Be Baked In, Not Bolted On

    cs.AI 2026-05 unverdicted novelty 4.0

    Argues that trustworthiness in Agent-to-Agent networks requires a new conceptual framework with four design pillars baked in from the beginning, as retrofitting existing single-agent methods is insufficient.

  16. LLM-Powered AI Agent Systems and Their Applications in Industry

    cs.AI 2025-05 unverdicted novelty 2.0

    A survey categorizing LLM-powered agent systems into software-based, physical, and hybrid types, covering industrial applications and challenges such as latency and security.