Tag
The paper proposes HEST, a method that embeds LLM hidden states into the Poincaré ball with an entropy-trained probe and steers only high-entropy (hesitant) tokens along hyperbolic geodesics, improving math reasoning accuracy on Qwen2.5-Math and Llama-3.1 models where fixed-vector Euclidean steering fails.
The study investigates whether hallucination detection probes trained on monolingual text transfer to Hinglish, finding robust performance and higher hallucination rates in Hindi and Hinglish compared to English.
This paper proposes using lightweight MLP probes on activations from LLaMA-3.1-8B to detect harmful prompts with high F1 scores, offering a cost-effective alternative to large external guardrail models.