{"total":79,"rows":[{"id":"2405.14860","title":"Not All Language Model Features Are One-Dimensionally Linear","created":"2024-05-23","cited_by_count":212,"area":null},{"id":"2406.01506","title":"The Geometry of Categorical and Hierarchical Concepts in Large Language Models","created":"2024-06-03","cited_by_count":159,"area":null},{"id":"2405.00208","title":"A Primer on the Inner Workings of Transformer-based Language Models","created":"2024-04-30","cited_by_count":85,"area":null},{"id":"2506.03093","title":"From Flat to Hierarchical: Extracting Sparse Representations with Matching Pursuit","created":"2025-06-03","cited_by_count":49,"area":"mechinterp"},{"id":"2503.18878","title":"I Have Covered All the Bases Here: Interpreting Reasoning Features in Large Language Models via Sparse Autoencoders","created":"2025-03-24","cited_by_count":45,"area":"lr"},{"id":"2409.10502","title":"Causal Language Modeling Can Elicit Search and Reasoning Capabilities on Logic Puzzles","created":"2024-09-16","cited_by_count":38,"area":"lr"},{"id":"2604.28119","title":"Do Sparse Autoencoders Capture Concept Manifolds?","created":"2026-04-30","cited_by_count":31,"area":"mechinterp"},{"id":"2604.21691","title":"There Will Be a Scientific Theory of Deep Learning","created":"2026-04-23","cited_by_count":29,"area":null},{"id":"2504.12459","title":"On Linear Representations and Pretraining Data Frequency in Language Models","created":"2025-04-16","cited_by_count":23,"area":null},{"id":"2505.14827","title":"Text Generation Beyond Discrete Token Sampling","created":"2025-05-20","cited_by_count":21,"area":null},{"id":"2510.09782","title":"The Geometry of Reasoning: Flowing Logics in Representation Space","created":"2025-10-10","cited_by_count":21,"area":"lr"},{"id":"2406.04331","title":"PaCE: Parsimonious Concept Engineering for Large Language Models","created":"2024-06-06","cited_by_count":19,"area":null},{"id":"2510.21518","title":"Head Pursuit: Probing Attention Specialization in Multimodal Transformers","created":"2025-10-24","cited_by_count":17,"area":"mllm"},{"id":"2403.10381","title":"Monotonic Representation of Numeric Properties in Language Models","created":"2024-03-15","cited_by_count":15,"area":null},{"id":"2406.18400","title":"Do LLMs dream of elephants (when told not to)? Latent concept association and associative memory in transformers","created":"2024-06-26","cited_by_count":15,"area":null},{"id":"2410.23501","title":"All or None: Identifiable Linear Properties of Next-token Predictors in Language Modeling","created":"2024-10-30","cited_by_count":14,"area":null},{"id":"2503.08980","title":"I Predict Therefore I Am: Is Next Token Prediction Enough to Learn Human-Interpretable Concepts from Data?","created":"2025-03-12","cited_by_count":13,"area":null},{"id":"2505.18651","title":"On the Emergence of Linear Analogies in Word Embeddings","created":"2025-05-24","cited_by_count":13,"area":null},{"id":"2510.15804","title":"Emergence of Linear Truth Encodings in Language Models","created":"2025-10-17","cited_by_count":12,"area":null},{"id":"2602.15293","title":"The Information Geometry of Softmax: Probing and Steering","created":"2026-02-17","cited_by_count":11,"area":null},{"id":"2507.20333","title":"The Blessing and Curse of Dimensionality in Safety Alignment","created":"2025-07-27","cited_by_count":10,"area":"alignment"},{"id":"2601.20834","title":"Linear representations in language models can change dramatically over a conversation","created":"2026-01-28","cited_by_count":10,"area":null},{"id":"2503.04725","title":"L$^2$M: Mutual Information Scaling Law for Long-Context Language Modeling","created":"2025-03-06","cited_by_count":9,"area":null},{"id":"2502.09863","title":"Closed-Form Training Dynamics Reveal Learned Features and Linear Structure in Word2Vec-like Models","created":"2025-02-14","cited_by_count":8,"area":null},{"id":"2602.00333","title":"Efficient and accurate steering of Large Language Models through attention-guided feature learning","created":"2026-01-30","cited_by_count":8,"area":null},{"id":"2408.03459","title":"Can DPO Learn Diverse Human Values? A Theoretical Scaling Law","created":"2024-08-06","cited_by_count":7,"area":"alignment"},{"id":"2507.21141","title":"The Geometry of Harmfulness in LLMs through Subconcept Probing","created":"2025-07-23","cited_by_count":7,"area":null},{"id":"2406.17837","title":"Transformer Normalisation Layers and the Independence of Semantic Subspaces","created":"2024-06-25","cited_by_count":6,"area":null},{"id":"2411.00246","title":"ResiDual Transformer Alignment with Spectral Decomposition","created":"2024-10-31","cited_by_count":6,"area":null},{"id":"2603.28744","title":"Stop Probing, Start Coding: Why Linear Probes and Sparse Autoencoders Fail at Compositional Generalisation","created":"2026-03-30","cited_by_count":6,"area":"mechinterp"},{"id":"2406.03944","title":"Provably Neural Active Learning Succeeds via Prioritizing Perplexing Samples","created":"2024-06-06","cited_by_count":5,"area":null},{"id":"2410.01545","title":"Lines of Thought in Large Language Models","created":"2024-10-02","cited_by_count":5,"area":"lr"},{"id":"2506.03790","title":"Attention-Only Transformers via Unrolled Subspace Denoising","created":"2025-06-04","cited_by_count":5,"area":null},{"id":"2605.01167","title":"Minimizing Collateral Damage in Activation Steering","created":"2026-05-01","cited_by_count":5,"area":"mechinterp"},{"id":"2405.18400","title":"Superposed Decoding: Multiple Generations from a Single Autoregressive Inference Pass","created":"2024-05-28","cited_by_count":4,"area":null},{"id":"2412.16247","title":"Towards scientific discovery with dictionary learning: Extracting biological concepts from microscopy foundation models","created":"2024-12-20","cited_by_count":4,"area":null},{"id":"2503.11429","title":"Combining Causal Models for More Accurate Abstractions of Neural Networks","created":"2025-03-14","cited_by_count":4,"area":"mechinterp"},{"id":"2412.07947","title":"GPT-2 Through the Lens of Vector Symbolic Architectures","created":"2024-12-10","cited_by_count":3,"area":"mechinterp"},{"id":"2502.16385","title":"Toward a Flexible Framework for Linear Representation Hypothesis Using Maximum Likelihood Estimation","created":"2025-02-22","cited_by_count":3,"area":null},{"id":"2510.00136","title":"Nonparametric Identification of Latent Concepts","created":"2025-09-30","cited_by_count":3,"area":null},{"id":"2512.05534","title":"A Unified Theory of Sparse Dictionary Learning in Mechanistic Interpretability: Piecewise Biconvexity and Spurious Minima","created":"2025-12-05","cited_by_count":3,"area":"mechinterp"},{"id":"2601.19208","title":"How Do Transformers Learn to Associate Tokens: Gradient Leading Terms Bring Mechanistic Interpretability","created":"2026-01-27","cited_by_count":3,"area":"mechinterp"},{"id":"2602.09783","title":"Why Linear Interpretability Works: Invariant Subspaces as a Result of Architectural Constraints","created":"2026-02-10","cited_by_count":3,"area":"mechinterp"},{"id":"2603.01227","title":"The Lattice Representation Hypothesis of Large Language Models","created":"2026-03-01","cited_by_count":3,"area":null},{"id":"2603.06054","title":"Probing Visual Concepts in Lightweight Vision-Language Models for Automated Driving","created":"2026-03-06","cited_by_count":3,"area":"mllm"},{"id":"2502.21123","title":"Causality Is Key to Understand and Balance Multiple Goals in Trustworthy ML and Foundation Models","created":"2025-02-28","cited_by_count":2,"area":null},{"id":"2508.16989","title":"Unveiling the Latent Directions of Reflection in Large Language Models","created":"2025-08-23","cited_by_count":2,"area":"mechinterp"},{"id":"2602.04718","title":"Superposition Without Interference? Towards Isolated Interventions via Almost Orthogonal Features in Language Models","created":"2026-02-04","cited_by_count":2,"area":"mechinterp"},{"id":"2606.26749","title":"Structure Before Collapse: Transient semantic geometry in next-token prediction","created":"2026-06-25","cited_by_count":2,"area":null},{"id":"2407.08934","title":"Compositional Structures in Neural Embedding and Interaction Decompositions","created":"2024-07-12","cited_by_count":1,"area":null},{"id":"2507.04754","title":"Intervening to Learn and Compose Causally Disentangled Representations","created":"2025-07-07","cited_by_count":1,"area":null},{"id":"2511.01932","title":"Deciphering Personalization: Towards Fine-Grained Explainability in Natural Language for Personalized Image Generation Models","created":"2025-11-02","cited_by_count":1,"area":null},{"id":"2602.11448","title":"Hierarchical Concept Embedding & Pursuit for Interpretable Image Classification","created":"2026-02-11","cited_by_count":1,"area":null},{"id":"2604.24936","title":"A Unifying Framework for Unsupervised Concept Extraction","created":"2026-04-27","cited_by_count":1,"area":"mechinterp"},{"id":"2605.02640","title":"Trustworthy AI Suffers from Invariance Conflicts and Causality is The Solution","created":"2026-05-04","cited_by_count":1,"area":null},{"id":"2605.05710","title":"On the Blessing of Pre-training in Weak-to-Strong Generalization","created":"2026-05-07","cited_by_count":1,"area":null},{"id":"2605.07148","title":"Uncovering and Shaping the Latent Representation of 3D Scene Topology in Vision-Language Models","created":"2026-05-08","cited_by_count":1,"area":"mllm"},{"id":"2605.28870","title":"Representation Alignment Rests on Linear Structure","created":"2026-05-22","cited_by_count":1,"area":null},{"id":"2505.11772","title":"LAMP: Extracting Local Decision Surfaces From Large Language Models","created":"2025-05-17","cited_by_count":0,"area":null},{"id":"2505.22506","title":"Sparsification and Reconstruction from the Perspective of Representation Geometry","created":"2025-05-28","cited_by_count":0,"area":"mechinterp"},{"id":"2509.25253","title":"Knowledge distillation through geometry-aware representational alignment","created":"2025-09-27","cited_by_count":0,"area":"distill"},{"id":"2510.24672","title":"Eigenfunction Extraction for Ordered Representation Learning","created":"2025-10-28","cited_by_count":0,"area":null},{"id":"2512.10720","title":"Beyond the Black Box: Identifiable Interpretation and Control in Generative Models via Causal Minimality","created":"2025-12-11","cited_by_count":0,"area":null},{"id":"2601.16398","title":"White-Box Sensitivity Auditing with Steering Vectors","created":"2026-01-23","cited_by_count":0,"area":"mechinterp"},{"id":"2601.20420","title":"Concept Component Analysis: A Principled Approach for Concept Extraction in LLMs","created":"2026-01-28","cited_by_count":0,"area":"mechinterp"},{"id":"2602.01687","title":"Functional Subspace, where language models can use vector algebra to solve problems","created":"2026-02-02","cited_by_count":0,"area":null},{"id":"2604.20276","title":"Rethinking Intrinsic Dimension Estimation in Neural Representations","created":"2026-04-22","cited_by_count":0,"area":null},{"id":"2605.31503","title":"How can embedding models bind concepts?","created":"2026-05-29","cited_by_count":0,"area":null},{"id":"2606.08197","title":"AlignFed: Alignment-Aware Asynchronous Federated Fine-Tuning for Large Language Models in Heterogeneous Edge Environments","created":"2026-06-06","cited_by_count":0,"area":null},{"id":"2606.14347","title":"When Language Representations Interact: Separability and Cross-Lingual Effects in LLMs","created":"2026-06-12","cited_by_count":0,"area":null},{"id":"2607.10578","title":"Laguerre Geometry for Interpreting Large Language Models","created":"2026-07-12","cited_by_count":0,"area":null},{"id":"2609.04753","title":"Beneath the Surface of Chains-of-Thought: A Mechanistic Interpretation of Reasoning Operations in LLMs","created":"2026-09-04","cited_by_count":0,"area":"lr"},{"id":"2609.22695","title":"A Survey on the Linear Representation Hypothesis","created":"2026-09-19","cited_by_count":0,"area":null},{"id":"2609.24821","title":"The Answer-Basin Representation Hypothesis: We Are Not Probing or Steering Concepts","created":"2026-09-21","cited_by_count":0,"area":"mechinterp"},{"id":"2609.30996","title":"The Linear Representation Hypothesis for Vision-Language-Action Models","created":"2026-09-25","cited_by_count":0,"area":"vla"},{"id":"2609.35351","title":"Interference Beyond Geometry in Concept Extraction","created":"2026-09-28","cited_by_count":0,"area":null},{"id":"2610.00910","title":"The Geometry of Contextual Relations: Language Models Address Facts by Order of Mention","created":"2026-10-01","cited_by_count":0,"area":null},{"id":"2610.04112","title":"The Independence Prior of SAEs Fragments Visual Concepts","created":"2026-10-02","cited_by_count":0,"area":"mechinterp"},{"id":"2610.06945","title":"Beyond the Linear Representation Hypothesis: Non-Linear Activation Steering in Text-to-Image Models","created":"2026-10-03","cited_by_count":0,"area":"mechinterp"}],"next_cursor":null}