Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
March 16, 2026Open Access

Streaming Epistemic Geometry in Large Language Models: Token-Level Dynamics of Certainty, Hallucination, and Refusal Across Five Model Families

View Full Paper
Ask AI
Bookmark
Share

Authors

IAInna Alieksieienko

Discussion

Loading...

Member takes

Overview

This analysis demonstrates distinct token-level dynamics of certainty, hallucination, and refusal across language model families, indicating their reliability.

Key Points

  • The study aims to track epistemic dynamics during text generation in large language models to understand how certainty and hallucination manifest.
  • Introduced streaming epistemic geometry for token-by-token tracking in autoregressive generations.
  • Applied PCA-based subspace analysis on five independently trained model families.
  • Used a logistic classifier trained on first-token projection scores for evaluation.
  • Distinct dynamic signatures for hallucination, refusal, and certainty were identified from the first token.
  • Achieved an AUC of 0.991 for the logistic classifier on Llama-3.1-8B with successful zero-shot transfer to TruthfulQA.
  • Subspace methods flagged factual citation errors, while output entropy detected physically improbable myths.

Cite This Study

Inna Alieksieienko (2026) studied this question.

synapsesocial.com/papers/69b79e888166e15b153ac094https://doi.org/10.5281/zenodo.19018426
View Full Paper
Ask AI
Bookmark
Share