cs.CR · 2026-09-02 · No. 103

Cryptography and Security, 2026-09-02.

6 new papers in cs.CR. Titles, authors, abstracts. Links to arXiv. Want this in your inbox every morning? Subscribe →

01 — The papers

6 entries
  1. 01

    Defense-as-Skill: Evolving Runtime Guard Skill for Skill-Augmented Agents

    Xiaofang Yang, Ziqi Miao, Dianbo Sui, Jing Shao, Lijun Li

    cs.CR · cs.AI

    Skill-augmented agents load reusable skills as persistent runtime context, improving task performance but also giving malicious skills a durable channel for steering future actions. Such skills may leak secrets, corrupt code, bypass approvals, or stage data for exfiltration only after a concrete user task and workspace state make the unsafe action appear useful. This makes pre-install vetting insufficient and calls for runtime,...

    arxiv.org/abs/2609.01487 · PDF

  2. 02

    When Safety Routing Breaks: Understanding Alignment Fragility under Benign Fine-Tuning

    Yitong Guo, Xiaoyi Chen, Siyuan Zhang, Xiaofeng Wang, Haixu Tang

    cs.CR · cs.AI

    Benign fine-tuning severely weakens the safety alignment of large language models (LLMs), so we study why refusal behavior is so fragile. While prior work often attributes this failure to gradient conflict, we propose a fundamentally different Fisher-geometric explanation: safety Fisher is low-rank, and alignment makes the safety geometry flatter while preserving an output-routing pathway. After 100 benign fine-tuning examples, this pathway...

    arxiv.org/abs/2609.01455 · PDF

  3. 03

    MutMem-V2: Cryptographically Authorized Mutation in Persistent Agent Memory Portable Verification and Reproducible Evidence

    Walid Saidi

    cs.CR · cs.AI

    MutMem V1 introduced retention-preserving, cryptographically authorized mutation for persistent agent memory but did not provide a complete portable verification contract or clean-install reproduction path. MutMem V2 closes that publication gap without introducing a second memory engine. It specifies exact canonical bytes, domain-separated object and bundle commitments, mandatory recall-evidence membership and ordering, external trust...

    arxiv.org/abs/2609.01235 · PDF

  4. 04

    Position Matters: Feature Inversion Attacks in ViT Split Inference with Token Reduction and Shuffling

    Stefano Leggio, Giulio Rossolini, Alessandro Biondi

    cs.CR · cs.AI

    Vision Transformers (ViTs) are increasingly used in split-inference systems, where edge devices transmit intermediate token representations to a remote cloud. In this setting, token reduction lowers computation and communication costs, while token shuffling disrupts the spatial organization of the transmitted tokens, potentially limiting information leakage. However, their privacy benefits remain unclear against feature inversion attacks,...

    arxiv.org/abs/2609.01232 · PDF

  5. 05

    Who Judges the Judges? A Chinese Safety QA Benchmark for Evaluating LLM Responses and Safety Judges

    Rui Yang, Shuang Huang, Junhua Liu, Ziqi Zhao, Qingzhong Yan, Yuhang Sun, Cong Liu, Guoping Hu, Rui Mei, Jing Shao

    cs.CR · cs.AI

    Safety benchmarks for large language models often assess the risk of a user query, although the outcome of question answering depends on whether the response violates a policy. This distinction is critical in Chinese harmful-content evaluation, where linguistic variation and adversarial transformations can obscure risky intent. We introduce C-SafeQA, a policy-grounded benchmark for response-level Chinese safety evaluation. It comprises 538...

    arxiv.org/abs/2609.01210 · PDF

  6. 06

    HiveTraceGuard-Pro: A Compact Generative Guardrail for Prompt Injection, Jailbreaks, and Adversarial Obfuscation

    Nikita Oblakov, Sabrina Sadiekh, Evgeniy Kokuykin

    cs.CR · cs.AI

    Production LLMs must handle inputs that attempt to override system instructions, bypass safety policies or elicit harmful responses. A common mitigation is a separate guardrail model. Existing reports, however, provide little evidence on Russian prompt injection or Russian surface obfuscation. We present HiveTraceGuard-Pro, a 0.6B generative guardrail LoRA-tuned from Qwen3-0.6B. It is trained on Russian and English and uses one binary scoring...

    arxiv.org/abs/2609.01046 · PDF

This edition is part of The Daily Abstract — cs.CR archive. Subscribe to receive these in your inbox each morning, automatically translated to Spanish, with reply-to-PDF: arxivdaily.ignorelist.com.

Colophon Set in Georgia, with system sans for interface chrome and a monospaced stack for code and paper identifiers. Sole accent: amber #D99C5E. Built and served on an always-free VM. The masthead is set 14% letterspaced because newspapers do that and it works.