Obfuscation leads jailbreak success across seven evaluated LLMs
Obfuscation averaged a 54.3% attack success rate, and self-reminder prompting cut ASR to 10% and 5% on two models.
52
AI model
Llama-3-8B
- Mentions · 7d1
- Mentions · 30d2
- All mentions4
- First seenSep 7, 2026
- Last mentioned4d ago
Timeline
4 articles · newest firstCola2 studies continuous semantic compression for hierarchical text generation
Authors report an unconverged checkpoint averaging 60.10 on eight tasks and beating Llama-3-8B on two.
48
AI models' written reasoning steps correspond to distinct internal patterns, a new study finds
KAIST and Naver AI Lab researchers show LLM reasoning steps like extraction and computation map to distinct activation patterns, strongest in middle layers.
42
Decomposition-Guided Diffusion Language Models for Inertial Confinement Fusion Prediction
ICF-DLM, the first language-model-based inertial confinement fusion predictor, cuts peak-timing error from 11.6 to 9.2 steps versus LLaMA-3-8B.
16
Appears with
- Qwen2.5-7Bai model · 2
- ICF-DLMai model · 1
- Qwen3-8Bai model · 1
- LLaMA-3-8Bproduct · 1
- Llama-3.1-8Bai model · 1
- Gemma2-9Bai model · 1
- KAISTorganization · 1
- Phi-4-14Bai model · 1
- Llama-3.2-3Bai model · 1
- Naver AI Laborganization · 1
- Mistral-7Bai model · 1
- Gemma4-31Bai model · 1
- GPT-5ai model · 1
- National Ignition Facilityorganization · 1
Entities are extracted by the model from each article. First seen , last mentioned . Watching keeps it in this browser only (no account); the watchlist page and dashboard alerts use it.