Rowhammer bit-flip in Triton slows or corrupts LLM inference
LLVMhammer raised single-GPU latency 4-16% on four models and could cut BERTScore to 25%.
52
AI model
Llama-3.1-8B-Instruct
- Mentions · 7d2
- Mentions · 30d3
- All mentions3
- First seenOct 2, 2026
- Last mentioned3d ago
Timeline
3 articles · newest firstAdaptive multi-token prediction selects draft horizons online
On Llama and Qwen pairings, it was up to 4.1x faster than standard decoding and beat a fixed window in 8 of 10 settings.
46
RMCW: A Deletion-Robust Watermark Based on Reed--Muller Codes for Language Models
RMCW uses Reed-Muller code structure to make LLM watermarks survive token-deletion and rewriting attacks while retaining clean-text detectability.
32
Appears with
- LLVMhammermalware · 1
- Qwen3-4B-Inst-2507-MTPai model · 1
- Karnagemalware · 1
- Qwen2.5-7B-Instructai model · 1
- OPT-1.3Borganization · 1
- Deepseek-llm-7B-chatai model · 1
- Tritonproduct · 1
- AMDvendor · 1
- Mistral-7B-Instruct-v0.3ai model · 1
- Metaorganization · 1
- NVIDIAvendor · 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.