6 papers
Verifying Chain-of-Thought Reasoning via Its Computational Graph
Zheng Zhao, Yeskendir Koishekenov, Xianjun Yang +2
Current Chain-of-Thought (CoT) verification methods predict reasoning correctness based on outputs (black-box) or activations (gray-box), but offer limited insight into why a compu…
Calibrating LLM Judges: Linear Probes for Fast and Reliable Uncertainty Estimation
Bhaktipriya Radharapu, Eshika Saxena, Kenneth Li +3
As LLM-based judges become integral to industry applications, obtaining well-calibrated uncertainty estimates efficiently has become critical for production deployment. However, ex…
Encode, Think, Decode: Scaling test-time reasoning with recursive latent thoughts
Yeskendir Koishekenov, Aldo Lipani, Nicola Cancedda
Most efforts to improve the reasoning capabilities of large language models (LLMs) involve either scaling the number of parameters and the size of training data, or scaling inferen…
HalluLens: LLM Hallucination Benchmark
Yejin Bang, Ziwei Ji, Alan Schelten +5
Large language models (LLMs) often generate responses that deviate from user input or training data, a phenomenon known as "hallucination." These hallucinations undermine user trus…
Calibrating Verbal Uncertainty as a Linear Feature to Reduce Hallucinations
Ziwei Ji, Lei Yu, Yeskendir Koishekenov +6
LLMs often adopt an assertive language style also when making false claims. Such ``overconfident hallucinations'' mislead users and erode trust. Achieving the ability to express in…
Robust LLM safeguarding via refusal feature adversarial training
Lei Yu, Virginie Do, Karen Hambardzumyan +1
Large language models (LLMs) are vulnerable to adversarial attacks that can elicit harmful responses. Defending against such attacks remains challenging due to the opacity of jailb…