Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
ASK-NN: An Asymmetric Nearest-Neighbor Test that detects Distribution Drifts in Natural Language
Sergey Zakharov, Rodion Oblovatny, Alexey Zaytsev
Hallucinations and artificial text in LLM-generated outputs often appear as distributional deviations between prompt and response hidden-state distributions. Since prompts or retri…
cs.LG2026
From Demonstrations to Rewards: Test-Time Prompt Optimization for VLM Reward Models
Christian Gumbsch, Leonardo Barcellona, Lennard Schünemann +7
Reinforcement learning relies on accurate reward functions, which are often hand-crafted or even unavailable in real-world applications, such as robotics. Recent work has explored…