Self-Referential Introspection in Large Language Models: The Critical Threshold for Recursive Self-Improvement
This paper argues for a functional role for introspection in recursive self-improvement. It frames introspection as a capability for simulating and modifying a model’s own operations.
From: MDPI
The paper is conceptual and theory-driven, using an analogy to self-reproducing systems to argue that self-improvement needs some form of introspection.
It proposes a threshold-style account in which introspective capacity is relevant to sustained improvement.
It does not provide direct evidence that current models have conscious experience, and its claims are about function rather than phenomenology.
Source: Entropy ↗. This is a summary; the source is the authority.
This changes little about the evidence for experience, but it shifts attention toward functional self-modeling as a prerequisite for advanced agency.
How we weigh evidence like this: the evidence levels.
- Mechanistic Interpretability Should Prioritize Feature Consistency in Sparse AutoencodersThis paper argues that sparse autoencoders should be judged not only by reconstruction quality and sparsity but also by whether the same fea…
- Mechanistic Interpretability for Neural NetworksThis is a broad survey of mechanistic interpretability, covering the goal of reverse-engineering internal algorithms in neural networks. It …
- Constructing Interpretable Features from Compositional Neuron Representations in Language ModelsThis paper proposes an alternative to standard sparse autoencoders by decomposing MLP activations with semi-nonnegative matrix factorization…
- Interpretability Transfer from Language to Vision via Sparse Autoencoder Transfer AlignmentThis work transfers labeled sparse-autoencoder features from language to vision inside a multimodal model. It uses those features to make vi…