New tools from IBM Research can help LLM users check AI-generated content for accuracy and relevance and defend against jailbreak attacks.| IBM Research
Researchers at Hebrew University, IBM, and Yale summarize the latest in AI agent benchmarking and suggest four ways it could be improved.| IBM Research
IBM Research has co-developed a new way to embed ownership information into synthetic tabular data to discourage its misuse.| IBM Research