InfoResearchPeer-reviewedLLM-specific
How Dangerous Are Data Poisoning Attacks on LLMs? An Empirical Study
- Published
- Record updated
Summary
This empirical study tests trigger-based data poisoning on LLaMA-3 8B Instruct, using 30,220 training examples from ten SuperNaturalInstructions tasks. Under a corrected attribution procedure, a 2% removal budget identified 61 of 604 poisoned examples, giving 10.10% precision and recall, while a literal search using the known trigger found all 604 with no false positives. The poisoned adapter produced the exact target response on 94.0–100.0% of triggered non-target examples across datasets.
Topics
Related items
- CriticalGHSA-9x8x-crjq-g64h: Jupyter MCP Server: Browser DNS rebinding can hijack the Jupyter backend through management routesSimilar attack · GitHub Advisory Database
- HighGHSA-24pw-rf5p-mgp7: Jupyter MCP Server: Unauthenticated management routes can hijack the Jupyter backend in standalone HTTP modeSimilar attack · GitHub Advisory Database
- HighGHSA-m932-63jc-4jqq: k8s-mcp-server: Missing authentication on default SSE/HTTP transport allows unauthenticated Kubernetes tool executionSimilar attack · GitHub Advisory Database
- HighGHSA-3mgh-93r5-r7jf: CodeWhale: exec_shell parallel-readonly classifier auto-approves fd/rg/git grep exec flags enabling silent RCESimilar attack · GitHub Advisory Database
- HighGHSA-qx49-fqc8-xw99: Model Context Protocol Python SDK: OAuth client could send credentials to an authorization server chosen by the MCP serverSimilar attack · GitHub Advisory Database