Tag
This paper extends optimal transport-based hallucination detection to all decoder layers in NMT and abstractive summarization, finding that detection is concentrated in early layers and that the geometric signal transfers poorly to summarization due to faithfulness failures not detectable via attention concentration.
This paper proposes Global-Local Uncertainty (GLU), an unsupervised single-pass score that fuses token-level local entropy with hidden-state geometric global entropy for uncertainty quantification in LLMs, showing that the two are near-orthogonal and together capture confident-but-wrong failures.
This paper proposes unsupervised Process Reward Models (uPRM) that eliminate the need for human annotations by using LLM next-token probabilities to identify erroneous reasoning steps, achieving up to 15% accuracy improvements over LLM-as-a-Judge and performing comparably to supervised PRMs as verifiers and reward signals.
Introduces LoVer, an unsupervised verifier that uses logical rules (negation consistency, intra-group and inter-group consistency) to improve LLM reasoning without labeled data, achieving performance close to supervised verifiers on reasoning benchmarks.
A developer built an unsupervised multi-agent pipeline that lets Claude and GPT-4 autonomously prep and host a podcast, including scouting topics, planning episodes, and conversing for 10 rounds before text-to-speech output.