paper:arxiv-2510-08638Into the rabbit hull: From task-relevant concepts in dino to minkowski geometry
Related work— refs + corpus + external arXiv
Cited / in-corpus / arXiv badges show which signals surfaced each row. Multi-source rows weighted higher.
- A Pattern Language for Machine Learning TasksIan Fan, Tuomas Laakkonen, Neil John Ortega, Thomas Hoffmann, Vincent Wang-Mascianica Benjamin Rodatz2025≈ 72%
- ≈ 71%
- An Internal Model Principle For RobotsTamara Alshammari, Kalle G. Timperi, Mehdi Bennis, Steven M. LaValle Vadim K. Weinstein2024≈ 71%
- All You Need is Sally-Anne: ToM in AI Strongly Supported After Surpassing Tests for 3-Year-OldsJoseph Barnby, Reuth Mirsky, Stefan Sarkadi Nitay Alon2025≈ 71%
- ≈ 71%
- Measuring and modeling the perception of natural and unconstrained gaze in humans and machinesTao Gao, Nancy Kanwisher, Joshua Tenenbaum, Shimon Ullman Daniel Harari2016≈ 71%
- Remapping and navigation of an embedding space via error minimization: a fundamental organizational principle of cognition in natural and artificial systemsL\'eo Pio-Lopez, Chris Fields, Michael Levin Benedikt Hartl2026≈ 71%
- From Theory of Mind to Theory of Environment: Counterfactual Simulation of Latent Environmental DynamicsRyutaro Uchiyama2026≈ 70%
- Drivers' Manoeuvre Modelling and Prediction for Safe HRIGuido Herrmann, Ute Leonards Erwin Jose Lopez Pulgarin2021≈ 70%
- From Transformer to Biology: A Hierarchical Model for Attention in Complex Problem-SolvingYunwei Li, Tianming Yang Zhongqiao Lin2025≈ 70%
- Explaining intuitive difficulty judgments by modeling physical effort and riskBasil Saeed, Grace Bennett-Pierre, Tobias Gerstenberg, Joshua Tenenbaum, Hyowon Gweon Ilker Yildirim2019≈ 70%
- Adding Knowledge to Unsupervised Algorithms for the Recognition of IntentQianli Feng, Aleix Martinez Stuart Synakowski2020≈ 70%
- A Computable Game-Theoretic Framework for Multi-Agent Theory of MindYuxin Pan, Xiaomeng Zhu, Fangzhen Lin Fengming Zhu2025≈ 70%
- Using High-Level Patterns to Estimate How Humans Predict a Robot will BehaveLauren Bramblett, Nicola Bezzo, and Dylan P. Losey Sagar Parekh2025≈ 70%
- Proposal of Pattern Recognition as a necessary and sufficient Principle to Cognitive ScienceGilberto de Paiva2011≈ 70%
- Unsupervised Discovery of 3D Physical Objects from VideoKevin Smith, Tomer Ulman, Joshua Tenenbaum, Jiajun Wu Yilun Du2021≈ 70%
- The World Inside Neural Networksin corpus2026≈ 69%
- ≈ 69%
- ≈ 69%
- The biogenic approach to cognitionin corpus2005≈ 68%
- The Platonic Representation Hypothesisin corpus2024≈ 68%
- Manifold Steering Reveals the Shared Geometry of Neural Network Representation and Behaviorin corpus2026≈ 68%
- ≈ 68%
- ≈ 68%
- Learning without neurons in physical systemsin corpus2022≈ 67%
- ≈ 67%
- ≈ 67%
- ≈ 67%
- ≈ 67%
Similar preprints — Semantic Scholar
Cited by (4)
- Do Sparse Autoencoders Capture Concept Manifolds?
Sparse autoencoders trained on Llama3.1-8B (layer 19, 500M Pile tokens, 25 SAEs across Standard, JumpReLU, TopK, BatchTopK and Matryoshka architectures at VE > 0.85) do not capture continuous concept
- Structuring Sparsity: Block-Sparse Featurizers Capture Visual Concept Manifolds
Lifting the featurizer's atom from a direction to a block of directions recovers concept manifolds that SAEs shatter, and block sparsity is the MAP-matched prior for a representation that is a sparse
- Manifold Steering Reveals the Shared Geometry of Neural Network Representation and Behavior
Manifold steering — intervening on model activations along paths constrained to lie on a learned activation manifold M_h rather than along Euclidean linear directions — produces behavioral trajectorie
- Arithmetic in the Wild: Llama uses Base-10 Addition to Reason About Cyclic Concepts
Llama-3.1-8B solves cyclic arithmetic (e.g., "what month is six months after August?") not by performing modular addition in the period of the cyclic concept (12 for months, 7 for days of the week) as