Geometry-Lite decomposes LLM safety detection into layer-wise margin geometries and finds that persistent boundary positions, not layer-to-layer drift, drive most detection performance across nine models and seven benchmarks.
Truth as a trajectory: What internal representations reveal about large language model reasoning
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
verdicts
UNVERDICTED 3representative citing papers
DMET models LLM generation as controlled dynamical trajectories on a semantic manifold, with three proxy metrics that predict output quality and support adaptive decoding to lower perplexity.
Defines Entropy-Gradient Inversion as a geometric fingerprint of LRM reasoning and introduces CorR-PO to embed it in RL reward regularization, reporting improved benchmark performance.
citing papers explorer
-
Geometry-Lite: Interpretable Safety Probing via Layer-Wise Margin Geometry
Geometry-Lite decomposes LLM safety detection into layer-wise margin geometries and finds that persistent boundary positions, not layer-to-layer drift, drive most detection performance across nine models and seven benchmarks.
-
Latent Trajectory Dynamics in Large Language Models: A Manifold Evolution Framework with Empirical Validation
DMET models LLM generation as controlled dynamical trajectories on a semantic manifold, with three proxy metrics that predict output quality and support adaptive decoding to lower perplexity.
-
Entropy-Gradient Inversion: Moving Toward Internal Mechanism of Large Reasoning Models
Defines Entropy-Gradient Inversion as a geometric fingerprint of LRM reasoning and introduces CorR-PO to embed it in RL reward regularization, reporting improved benchmark performance.