Fidelity Is Not Safety: Compressed LLMs Pass Quality Guards yet Invent

A study submitted to arXiv on 30 Jul 2026 found that gently-compressed large language models (LLMs) can pass standard data-free quality guards—including perplexity, MMLU accuracy, and output-fidelity tests—yet still invent procedure steps when executing standard operating procedures as agents. The researchers, across three model families, showed that coherent low-rank (SVD) truncation induces this behavior while magnitude pruning matched to the same perplexity does not, and they propose a data-free two-axis screen (coherent-fraction and error-rate) to flag failing builds before agentic deployment.

Computer Science Computation and Language Submitted on 30 Jul 2026 Title:Fidelity Is Not Safety: Gently-Compressed LLMs Pass Every Data-Free Quality Guard Yet Invent Procedure Steps in Agentic Execution View PDF /pdf/2607.28196 HTML experimental https://arxiv.org/html/2607.28196v1 Abstract:Practitioners accept a compressed language model once it clears a stack of data-cheap quality guards: perplexity within a small factor of the original, downstream accuracy for example MMLU inside a confidence interval, and data-free output-fidelity signals that compare the compressed and original network's internal representations under random probe inputs. This stack has a blind spot. Across three model families, gently-compressed models clear every guard and then invent procedure steps that were never in the instructions when they run a standard operating procedure SOP as an agent. The effect is operator-specific: coherent low-rank SVD truncation induces it, and magnitude pruning matched to the same perplexity does not. One dissociation isolates the cause. The same compressed weights that CI-win a paired output-fidelity test CI-fail the invented-step canary. The governing axis is the coherence of the compression error times its rate; the magnitude of the damage does not predict it. The data-free fidelity probe is a fidelity oracle by construction, so it cannot see this axis. We characterize the blindspot and dissociation with paired confidence intervals on a pre-registered, powered canary across three architectures. Operator-specificity replicates on all three, and the perplexity-guard evasion appears where the model admits in-guard low-rank headroom. We then give a data-free screen: a two-axis statistic of the compression error coherent-fraction and error-rate that flags the failing builds with fixed thresholds across architectures and matches the coherence-times-rate mechanism. Perplexity, MMLU, and fidelity acceptance do not certify agent safety. Screen gently-compressed low-rank builds before agentic deployment References & Citations Loading... Bibliographic and Citation Tools Bibliographic Explorer What is the Explorer? https://info.arxiv.org/labs/showcase.html arxiv-bibliographic-explorer Connected Papers What is Connected Papers? https://www.connectedpapers.com/about Litmaps What is Litmaps? https://www.litmaps.co/ scite Smart Citations What are Smart Citations? https://www.scite.ai/ Code, Data and Media Associated with this Article alphaXiv What is alphaXiv? https://alphaxiv.org/ CatalyzeX Code Finder for Papers What is CatalyzeX? https://www.catalyzex.com DagsHub What is DagsHub? https://dagshub.com/ Gotit.pub What is GotitPub? http://gotit.pub/faq Hugging Face What is Huggingface? https://huggingface.co/huggingface ScienceCast What is ScienceCast? https://sciencecast.org/welcome Demos Recommenders and Search Tools Influence Flower What are Influence Flowers? https://influencemap.cmlab.dev/ CORE Recommender What is CORE? https://core.ac.uk/services/recommender arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs https://info.arxiv.org/labs/index.html .