arXiv:2602. 11852v2 Announce Type: replace Abstract: While state-of-the-art language models (LMs) surpass most humans in certain domains, their reasoning remains largely opaque, reducing trust and increasing the risk of deception and hallucination.
arXiv:2607. 00510v1 Announce Type: new Abstract: Knowing which training examples drive outputs is fundamental to auditing, correcting, and understanding language models, yet for modern LLMs this remains expensive, approximate, and largely post-hoc.
arXiv:2510. 25013v2 Announce Type: replace-cross Abstract: Mechanistic interpretability aims to reverse-engineer large language models (LLMs) into human-understandable computational circuits.
arXiv:2602. 10352v2 Announce Type: replace-cross Abstract: Self-interpretation methods prompt language models to describe their own internal states, but remain unreliable due to hyperparameter sensitivity.
arXiv:2606. 19317v1 Announce Type: cross Abstract: A longstanding goal of research on interpretable deep learning is to replace opaque neural computations with human-meaningful symbolic descriptions.
arXiv:2601. 22588v2 Announce Type: replace-cross Abstract: Large language models (LLMs) are widely used as reference-free evaluators via prompting, but this "LLM-as-a-Judge" paradigm is costly, opaque, and sensitive to prompt design.
arXiv:2511. 05963v4 Announce Type: replace Abstract: Transformers replace recurrence with a memory that grows with sequence length and self-attention that enables ad-hoc lookups over past tokens.
arXiv:2608. 07594v1 Announce Type: cross Abstract: Interpretability is often treated as a tax on capability: language models are trained as opaque systems, then explained after the fact, with methods whose reliability is difficult to establish.
arXiv:2408. 04619v2 Announce Type: replace-cross Abstract: The Transformer architecture underpins modern large language models powering state-of-the-art text generation and AI applications.
A longstanding goal of research on interpretable deep learning is to replace opaque neural computations with human-meaningful symbolic descriptions. In this paper, we propose an approach for approximating the behavior of components of deep networks with executable programs.
arXiv:2608. 04980v1 Announce Type: cross Abstract: We show that tiny transformers can profitably employ a simple form of Chain of Thought, which we call protoreasoning, allowing us to study step-by-step reasoning on ~1M-parameter models and opening up opportunities for much more detailed experimentation and analysis than is feasible for larger models.
arXiv:2407. 21082v3 Announce Type: replace-cross Abstract: This paper presents a modular approach to accelerate inference in large language models (LLMs) by adding early exit heads at intermediate transformer layers.
arXiv:2606. 19857v1 Announce Type: cross Abstract: Large language models (LLMs) are commonly prompted and interfaced with human-readable natural language, even when the intended reader is another model.