CLASP: Defending Hybrid Large Language Models Against Hidden State Poisoning Attacks

• URL: http://arxiv.org/abs/2603.12206v1
• Date: Thu, 12 Mar 2026 17:29:55 GMT
• Title: CLASP: Defending Hybrid Large Language Models Against Hidden State Poisoning Attacks
• Abstract summary: State space models (SSMs) like Mamba have gained significant traction as efficient alternatives to Transformers.<n>HiSPAs, a recently discovered vulnerability that corrupts SSM memory through adversarial strings, pose a critical threat to these architectures.<n>We introduce the CLASP model to defend against this threat.
• Score: 48.54598003197356
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: State space models (SSMs) like Mamba have gained significant traction as efficient alternatives to Transformers, achieving linear complexity while maintaining competitive performance. However, Hidden State Poisoning Attacks (HiSPAs), a recently discovered vulnerability that corrupts SSM memory through adversarial strings, pose a critical threat to these architectures and their hybrid variants. Framing the HiSPA mitigation task as a binary classification problem at the token level, we introduce the CLASP model to defend against this threat. CLASP exploits distinct patterns in Mamba's block output embeddings (BOEs) and uses an XGBoost classifier to identify malicious tokens with minimal computational overhead. We consider a realistic scenario in which both SSMs and HiSPAs are likely to be used: an LLM screening résumés to identify the best candidates for a role. Evaluated on a corpus of 2,483 résumés totaling 9.5M tokens with controlled injections, CLASP achieves 95.9% token-level F1 score and 99.3% document-level F1 score on malicious tokens detection. Crucially, the model generalizes to unseen attack patterns: under leave-one-out cross-validation, performance remains high (96.9% document-level F1), while under clustered cross-validation with structurally novel triggers, it maintains useful detection capability (91.6% average document-level F1). Operating independently of any downstream model, CLASP processes 1,032 tokens per second with under 4GB VRAM consumption, potentially making it suitable for real-world deployment as a lightweight front-line defense for SSM-based and hybrid architectures. All code and detailed results are available at https://anonymous.4open.science/r/hispikes-91C0.

Related papers

• MaskForge: Structure-Aware Adaptive Attacks for Jailbreaking Diffusion Large Language Models [53.05463623673949]
  MaskForge is a fully black-box adaptive attack that casts dLLM red-teaming as optimized search over a growing library of structural patterns.<n>It achieves an average attack success rate of 79.3%, a 17.6% improvement over the strongest competing dLLM baseline.
  arXiv  Detail & Related papers  (2026-06-01T18:10:21Z)
• Context-Aware Web Attack Detection in Open-Source SIEM Systems via MITRE ATT&CK-Enriched Behavioral Profiling [0.0]
  We present Smart-SIEM, an AI module for the open-source Wazuh SIEM platform.<n>A behavioural context vector encoding HTTP response-status distributions is used.<n>A two-stage hybrid cascade combines LightGBM for binary attack detection and XGBoost for six-class attack categorisation.
  arXiv  Detail & Related papers  (2026-05-13T10:54:36Z)
• GuardPhish: Securing Open-Source LLMs from Phishing Abuse [0.7009487789080343]
  Open-source Large Language Models (LLMs) in offline and enterprise environments have introduced a largely unexamined security risk.<n>We investigate this vulnerability through GuardPhish, a large scale multi-vector phishing prompt dataset.<n>Using a deterministic five model ensemble for labeling, we achieve near perfect inter model agreement.
  arXiv  Detail & Related papers  (2026-04-19T08:02:57Z)
• ReasAlign: Reasoning Enhanced Safety Alignment against Prompt Injection Attack [52.17935054046577]
  We present ReasAlign, a model-level solution to improve safety alignment against indirect prompt injection attacks.<n>ReasAlign incorporates structured reasoning steps to analyze user queries, detect conflicting instructions, and preserve the continuity of the user's intended tasks.
  arXiv  Detail & Related papers  (2026-01-15T08:23:38Z)
• COBRA: Catastrophic Bit-flip Reliability Analysis of State-Space Models [6.546311951672279]
  We introduce RAMBO, the first framework specifically designed to target Mamba-based architectures.<n>We show that flipping merely a single critical bit can catastrophically reduce accuracy from 74.64% to 0% and increase perplexity from 18.94 to 3.75 x 106.
  arXiv  Detail & Related papers  (2025-12-14T09:50:44Z)
• Revisiting Backdoor Attacks on LLMs: A Stealthy and Practical Poisoning Framework via Harmless Inputs [54.90315421117162]
  We propose a novel poisoning method via completely harmless data.<n>Inspired by the causal reasoning in auto-regressive LLMs, we aim to establish robust associations between triggers and an affirmative response prefix.<n>We observe an interesting resistance phenomenon where the LLM initially appears to agree but subsequently refuses to answer.
  arXiv  Detail & Related papers  (2025-05-23T08:13:59Z)
• No Query, No Access [50.18709429731724]
  We introduce the textbfVictim Data-based Adrial Attack (VDBA), which operates using only victim texts.<n>To prevent access to the victim model, we create a shadow dataset with publicly available pre-trained models and clustering methods.<n>Experiments on the Emotion and SST5 datasets show that VDBA outperforms state-of-the-art methods, achieving an ASR improvement of 52.08%.
  arXiv  Detail & Related papers  (2025-05-12T06:19:59Z)
• FFCBA: Feature-based Full-target Clean-label Backdoor Attacks [10.650796825194337]
  Backdoor attacks pose a significant threat to deep neural networks.<n>Clean-label attacks are more stealthy, as they avoid modifying the labels of poisoned samples.<n>We propose the Feature-based Full-target Clean-label Backdoor Attacks (FFCBA)
  arXiv  Detail & Related papers  (2025-04-29T05:49:42Z)
• ELBA-Bench: An Efficient Learning Backdoor Attacks Benchmark for Large Language Models [55.93380086403591]
  Generative large language models are vulnerable to backdoor attacks.<n>$textitELBA-Bench$ allows attackers to inject backdoor through parameter efficient fine-tuning.<n>$textitELBA-Bench$ provides over 1300 experiments.
  arXiv  Detail & Related papers  (2025-02-22T12:55:28Z)
• Targeting Alignment: Extracting Safety Classifiers of Aligned LLMs [6.462219916993885]
  Alignment in large language models (LLMs) is used to enforce guidelines such as safety.<n>Yet, alignment fails in the face of jailbreak attacks that modify inputs to induce unsafe outputs.<n>We introduce and evaluate a new technique for jailbreak attacks.
  arXiv  Detail & Related papers  (2025-01-27T22:13:05Z)
• Versatile Weight Attack via Flipping Limited Bits [68.45224286690932]
  We study a novel attack paradigm, which modifies model parameters in the deployment stage. Considering the effectiveness and stealthiness goals, we provide a general formulation to perform the bit-flip based weight attack. We present two cases of the general formulation with different malicious purposes, i.e., single sample attack (SSA) and triggered samples attack (TSA)
  arXiv  Detail & Related papers  (2022-07-25T03:24:58Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.