Hidden Threat in Synthetic Data: Covert Targeted Bias Injection through Benign Text
cs.CL, cs.AI
Submitted: 2026-08-31
Updated: 2026-08-31
Code: https://github.com/ada-flo/covert-bias-injection
Terminology
Sources
- Phi-4 Technical Report
- Program Synthesis with Large Language Models
- Constitutional AI: Harmlessness from AI Feedback
- Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
- Training Verifiers to Solve Math Word Problems
- BadLlama: cheaply removing safety fine-tuning from Llama 2-Chat 13B
- The Llama 3 Herd of Models
- Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations
- Mistral 7B
- A Rank Stabilization Scaling Factor for Fine-Tuning with LoRA
- 2 OLMo 2 Furious
- gpt-oss-120b & gpt-oss-20b Model Card
- GPT-4o System Card
- Qwen2.5 Technical Report
- Llama 2: Open Foundation and Fine-Tuned Chat Models
- SynLexLM: Scaling Legal LLMs with Synthetic Data and Curriculum Learning
- Qwen3 Technical Report
- Qwen3Guard Technical Report
- Universal and Transferable Adversarial Attacks on Aligned Language Models
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering