arXiv:2607. 10578v1 Announce Type: new Abstract: Existing hypotheses represent a concept in an LLM as a single point, a linear direction, or a Gaussian cluster, yet it remains unclear how and why such structures emerge.
arXiv:2512. 07355v2 Announce Type: replace Abstract: Two traditions of interpretability have evolved side by side but seldom spoken to each other: Concept Bottleneck Models (CBMs), which prescribe what a concept should be, and Sparse Autoencoders (SAEs), which discover what concepts emerge.
arXiv:2511. 19418v3 Announce Type: replace-cross Abstract: Vision-Language Models (VLMs) excel at reasoning in linguistic space but struggle with perceptual understanding that requires dense visual perception, e.
arXiv:2607. 25531v1 Announce Type: cross Abstract: Contemporary machine learning struggles to learn continually, reuse prior knowledge, and expose a comprehensible internal structure.
arXiv:2602. 24264v2 Announce Type: replace-cross Abstract: Compositional generalization, the ability to recognize familiar parts in novel contexts, is a defining property of intelligent systems.
arXiv:2610.02117v1 Announce Type: cross Abstract: On-policy self-distillation has recently emerged as an effective approach for improving language-model reasoning by supervising students with a froze...
arXiv:2607.10744v5 Announce Type: replace Abstract: Benefiting from the powerful priors embedded in large-scale pre-training data and the emerging commonsense reasoning ability, large language models...
The paper introduces Auto-Comp, a fully automated, concept-driven pipeline that generates photorealistic compositional benchmarks for vision‑language models. Auto‑Comp creates paired Minimal and Contextual samples for each concept, enabling isolation of core binding abilities from visio‑linguistic complexity. Evaluations across 25 models reveal consistent failures in attribute and relational binding, with context helping relational tasks but hindering attribute tasks due to visual clutter.
arXiv:2602.12486v2 Announce Type: replace-cross Abstract: Humans appear to represent objects when reasoning about physics with coarse, volumetric "bodies" that smooth concavities, trading fine visual...
arXiv:2609.38625v1 Announce Type: cross Abstract: Concept Bottleneck Models (CBMs) are designed to provide interpretable intermediate representations, yet how such bottlenecks affect robustness remai...
arXiv:2609.37680v1 Announce Type: cross Abstract: One of the current premises of mechanistic interpretability research is that detailed accounts of the geometry of neural network representations can...
arXiv:2606. 09287v1 Announce Type: new Abstract: Understanding how transformer representations evolve across layers, not merely what they encode, remains an open problem in mechanistic interpretability.