Ruixiang Tang

LLMs have "hallucination neurons" for specific citation fields, and silencing them reduces fabrication.

LLMs can now translate classical machine learning code into quantum machine learning code with surprising fidelity, opening doors to automated quantum algorithm design.

Don't let your robot's brief moment of panic get lost in the noise – this new uncertainty method spotlights those critical spikes to predict failures before they happen.

Ditch the rigid safety codes: case-augmented reasoning unlocks safer, more helpful LLMs that are also more robust to attacks.