{"total":1085,"rows":[{"id":"2406.11717","title":"Refusal in Language Models Is Mediated by a Single Direction","created":"2024-06-17","cited_by_count":1032,"area":"safety"},{"id":"2406.15927","title":"Semantic Entropy Probes: Robust and Cheap Hallucination Detection in LLMs","created":"2024-06-22","cited_by_count":259,"area":null},{"id":"2501.17148","title":"AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders","created":"2025-01-28","cited_by_count":218,"area":"mechinterp"},{"id":"2605.29358","title":"Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet","created":"2026-05-28","cited_by_count":191,"area":"mechinterp"},{"id":"2402.16835","title":"Eight Methods to Evaluate Robust Unlearning in LLMs","created":"2024-02-26","cited_by_count":163,"area":"safety"},{"id":"2410.12877","title":"Improving Instruction-Following in Language Models through Activation Steering","created":"2024-10-15","cited_by_count":160,"area":"mechinterp"},{"id":"2408.15221","title":"LLM Defenses Are Not Robust to Multi-Turn Human Jailbreaks Yet","created":"2024-08-27","cited_by_count":151,"area":"safety"},{"id":"2407.12404","title":"Analyzing the Generalization and Reliability of Steering Vectors","created":"2024-07-17","cited_by_count":143,"area":"mechinterp"},{"id":"2507.03724","title":"MemOS: A Memory OS for AI System","created":"2025-07-04","cited_by_count":138,"area":"rag"},{"id":"2406.00045","title":"Personalized Steering of Large Language Models: Versatile Steering Vectors Through Bi-directional Preference Optimization","created":"2024-05-28","cited_by_count":118,"area":"alignment"},{"id":"2502.17420","title":"The Geometry of Refusal in Large Language Models: Concept Cones and Representational Independence","created":"2025-02-24","cited_by_count":115,"area":"alignment"},{"id":"2504.02234","title":"LLM Social Simulations Are a Promising Research Method","created":"2025-04-03","cited_by_count":99,"area":null},{"id":"2411.11296","title":"Steering Language Model Refusal with Sparse Autoencoders","created":"2024-11-18","cited_by_count":95,"area":"mechinterp"},{"id":"2411.02193","title":"Improving Steering Vectors by Targeting Sparse Autoencoder Features","created":"2024-11-04","cited_by_count":92,"area":"mechinterp"},{"id":"2403.05030","title":"Defending Against Unforeseen Failure Modes with Latent Adversarial Training","created":"2024-03-08","cited_by_count":88,"area":"safety"},{"id":"2507.11878","title":"LLMs Encode Harmfulness and Refusal Separately","created":"2025-07-16","cited_by_count":88,"area":"safety"},{"id":"2506.19823","title":"Persona Features Control Emergent Misalignment","created":"2025-06-24","cited_by_count":86,"area":null},{"id":"2502.15603","title":"Do Multilingual LLMs Think In English?","created":"2025-02-21","cited_by_count":78,"area":"lr"},{"id":"2505.20063","title":"SAEs Are Good for Steering -- If You Select the Right Features","created":"2025-05-26","cited_by_count":78,"area":"mechinterp"},{"id":"2506.18167","title":"Understanding Reasoning in Thinking Language Models via Steering Vectors","created":"2025-06-22","cited_by_count":72,"area":"lr"},{"id":"2506.11618","title":"Convergent Linear Representations of Emergent Misalignment","created":"2025-06-13","cited_by_count":71,"area":null},{"id":"2411.04986","title":"The Semantic Hub Hypothesis: Language Models Share Semantic Representations Across Languages and Modalities","created":"2024-11-07","cited_by_count":69,"area":null},{"id":"2410.23054","title":"Controlling Language and Diffusion Models by Transporting Activations","created":"2024-10-30","cited_by_count":68,"area":"diffusion"},{"id":"2409.20089","title":"Robust LLM safeguarding via refusal feature adversarial training","created":"2024-09-30","cited_by_count":64,"area":null},{"id":"2410.12299","title":"Semantics-Adaptive Activation Intervention for LLMs via Dynamic Steering Vectors","created":"2024-10-16","cited_by_count":63,"area":"mechinterp"},{"id":"2403.15498","title":"Emergent World Models and Latent Variable Estimation in Chess-Playing Language Models","created":"2024-03-21","cited_by_count":62,"area":"wm"},{"id":"2405.09719","title":"Spectral Editing of Activations for Large Language Model Alignment","created":"2024-05-15","cited_by_count":56,"area":"alignment"},{"id":"2405.09605","title":"Elements of World Knowledge (EWoK): A Cognition-Inspired Framework for Evaluating Basic World Knowledge in Language Models","created":"2024-05-15","cited_by_count":55,"area":"wm"},{"id":"2409.01658","title":"From Yes-Men to Truth-Tellers: Addressing Sycophancy in Large Language Models with Pinpoint Tuning","created":"2024-09-03","cited_by_count":54,"area":null},{"id":"2506.10805","title":"Detecting High-Stakes Interactions with Activation Probes","created":"2025-06-12","cited_by_count":53,"area":null},{"id":"2510.26243","title":"Angular Steering: Behavior Control via Rotation in Activation Space","created":"2025-10-30","cited_by_count":53,"area":null},{"id":"2512.15674","title":"Activation Oracles: Training and Evaluating LLMs as General-Purpose Activation Explainers","created":"2025-12-17","cited_by_count":52,"area":null},{"id":"2506.02867","title":"Demystifying Reasoning Dynamics with Mutual Information: Thinking Tokens are Information Peaks in LLM Reasoning","created":"2025-06-03","cited_by_count":51,"area":"lr"},{"id":"2408.11491","title":"SCANS: Mitigating the Exaggerated Safety for LLMs via Safety-Conscious Activation Steering","created":"2024-08-21","cited_by_count":50,"area":"alignment"},{"id":"2501.00070","title":"ICLR: In-Context Learning of Representations","created":"2024-12-29","cited_by_count":50,"area":null},{"id":"2504.02956","title":"Understanding Aha Moments: from External Observations to Internal Mechanisms","created":"2025-04-03","cited_by_count":50,"area":"lr"},{"id":"2508.02087","title":"When Truth Is Overridden: Uncovering the Internal Origins of Sycophancy in Large Language Models","created":"2025-08-04","cited_by_count":50,"area":null},{"id":"2506.07022","title":"AlphaSteer: Learning Refusal Steering with Principled Null-Space Constraint","created":"2025-06-08","cited_by_count":49,"area":"safety"},{"id":"2410.03415","title":"Surgical, Cheap, and Flexible: Mitigating False Refusal in Language Models via Single Vector Ablation","created":"2024-10-04","cited_by_count":48,"area":null},{"id":"2411.16721","title":"Steering Away from Harm: An Adaptive Approach to Defending Vision Language Model Against Jailbreaks","created":"2024-11-23","cited_by_count":48,"area":"mllm"},{"id":"2503.14477","title":"Calibrating Verbal Uncertainty as a Linear Feature to Reduce Hallucinations","created":"2025-03-18","cited_by_count":48,"area":null},{"id":"2505.22637","title":"Understanding (Un)Reliability of Steering Vectors in Language Models","created":"2025-05-28","cited_by_count":47,"area":"mechinterp"},{"id":"2403.08946","title":"Usable XAI: 10 Strategies Towards Exploiting Explainability in the LLM Era","created":"2024-03-13","cited_by_count":46,"area":null},{"id":"2405.17374","title":"Navigating the Safety Landscape: Measuring Risks in Finetuning Large Language Models","created":"2024-05-27","cited_by_count":46,"area":"alignment"},{"id":"2406.15518","title":"Steering Without Side Effects: Improving Post-Deployment Control of Language Models","created":"2024-06-21","cited_by_count":46,"area":"safety"},{"id":"2502.02716","title":"A Unified Understanding and Evaluation of Steering Methods","created":"2025-02-04","cited_by_count":46,"area":"mechinterp"},{"id":"2502.12446","title":"Multi-Attribute Steering of Language Models via Targeted Intervention","created":"2025-02-18","cited_by_count":46,"area":null},{"id":"2406.17759","title":"Interpreting Attention Layer Outputs with Sparse Autoencoders","created":"2024-06-25","cited_by_count":43,"area":"mechinterp"},{"id":"2507.16003","title":"Learning without training: The implicit dynamics of in-context learning","created":"2025-07-21","cited_by_count":43,"area":null},{"id":"2501.08319","title":"Enhancing Automated Interpretability with Output-Centric Feature Descriptions","created":"2025-01-14","cited_by_count":41,"area":null},{"id":"2502.07218","title":"LLM Unlearning via Neural Activation Redirection","created":"2025-02-11","cited_by_count":41,"area":"safety"},{"id":"2403.05767","title":"Extending Activation Steering to Broad Skills and Multiple Behaviours","created":"2024-03-09","cited_by_count":40,"area":"mechinterp"},{"id":"2406.12094","title":"Who's asking? User personas and the mechanics of latent misalignment","created":"2024-06-17","cited_by_count":40,"area":"safety"},{"id":"2501.02497","title":"A Survey of Test-Time Compute: From Intuitive Inference to Deliberate Reasoning","created":"2025-01-05","cited_by_count":40,"area":"lr"},{"id":"2504.14492","title":"FairSteer: Inference Time Debiasing for LLMs with Dynamic Activation Steering","created":"2025-04-20","cited_by_count":40,"area":"mechinterp"},{"id":"2403.05518","title":"Bias-Augmented Consistency Training Reduces Biased Reasoning in Chain-of-Thought","created":"2024-03-08","cited_by_count":39,"area":"lr"},{"id":"2406.17563","title":"Multi-property Steering of Large Language Models with Dynamic Activation Composition","created":"2024-06-25","cited_by_count":39,"area":"mechinterp"},{"id":"2410.16314","title":"Steering Large Language Models using Conceptors: Improving Addition-Based Activation Engineering","created":"2024-10-09","cited_by_count":39,"area":"mechinterp"},{"id":"2505.14685","title":"Language Models use Lookbacks to Track Beliefs","created":"2025-05-20","cited_by_count":39,"area":null},{"id":"2507.16795","title":"Steering Out-of-Distribution Generalization with Concept Ablation Fine-Tuning","created":"2025-07-22","cited_by_count":39,"area":null},{"id":"2504.09466","title":"AdaSteer: Your Aligned LLM is Inherently an Adaptive Jailbreak Defender","created":"2025-04-13","cited_by_count":38,"area":"alignment"},{"id":"2406.09289","title":"Understanding Jailbreak Success: A Study of Latent Space Dynamics in Large Language Models","created":"2024-06-13","cited_by_count":37,"area":"safety"},{"id":"2509.21305","title":"Sycophancy Is Not One Thing: Causal Separation of Sycophantic Behaviors in LLMs","created":"2025-09-25","cited_by_count":37,"area":null},{"id":"2408.11779","title":"Personality Alignment of Large Language Models","created":"2024-08-21","cited_by_count":36,"area":"alignment"},{"id":"2410.14273","title":"REEF: Representation Encoding Fingerprints for Large Language Models","created":"2024-10-18","cited_by_count":36,"area":null},{"id":"2410.17245","title":"Towards Reliable Evaluation of Behavior Steering Interventions in LLMs","created":"2024-10-22","cited_by_count":36,"area":"mechinterp"},{"id":"2503.21073","title":"Shared Global and Local Geometry of Language Model Embeddings","created":"2025-03-27","cited_by_count":36,"area":null},{"id":"2509.03730","title":"The Personality Illusion: Revealing Dissociation Between Self-Reports & Behavior in LLMs","created":"2025-09-03","cited_by_count":36,"area":null},{"id":"2501.09929","title":"Interpretable Steering of Large Language Models with Feature Guided Activation Additions","created":"2025-01-17","cited_by_count":35,"area":"mechinterp"},{"id":"2507.01786","title":"Probing and Steering Evaluation Awareness of Language Models","created":"2025-07-02","cited_by_count":35,"area":null},{"id":"2410.04332","title":"Gradient Routing: Masking Gradients to Localize Computation in Neural Networks","created":"2024-10-06","cited_by_count":34,"area":null},{"id":"2502.11356","title":"SAIF: A Sparse Autoencoder Framework for Interpreting and Steering Instruction Following of Language Models","created":"2025-02-17","cited_by_count":34,"area":"mechinterp"},{"id":"2503.05213","title":"Personalized Text Generation with Contrastive Activation Steering","created":"2025-03-07","cited_by_count":34,"area":"mechinterp"},{"id":"2504.04635","title":"Steering off Course: Reliability Challenges in Steering Language Models","created":"2025-04-06","cited_by_count":34,"area":null},{"id":"2410.22366","title":"One-Step is Enough: Sparse Autoencoders for Text-to-Image Diffusion Models","created":"2024-10-28","cited_by_count":33,"area":"diffusion"},{"id":"2505.20977","title":"Evaluating and Steering Modality Preferences in Multimodal Large Language Model","created":"2025-05-27","cited_by_count":33,"area":"mllm"},{"id":"2505.22411","title":"Mitigating Overthinking in Large Reasoning Models via Manifold Steering","created":"2025-05-28","cited_by_count":33,"area":"lr"},{"id":"2409.14586","title":"Backtracking Improves Generation Safety","created":"2024-09-22","cited_by_count":32,"area":"alignment"},{"id":"2411.04430","title":"Towards Unifying Interpretability and Control: Evaluation via Intervention","created":"2024-11-07","cited_by_count":32,"area":"mechinterp"},{"id":"2504.01550","title":"Representation Bending for Large Language Model Safety","created":"2025-04-02","cited_by_count":30,"area":"safety"},{"id":"2505.14617","title":"The Hawthorne Effect in Reasoning Models: Evaluating and Steering Test Awareness","created":"2025-05-20","cited_by_count":30,"area":"lr"},{"id":"2402.09236","title":"Learning Interpretable Concepts: Unifying Causal Representation Learning and Foundation Models","created":"2024-02-14","cited_by_count":29,"area":null},{"id":"2406.00244","title":"Controlling Large Language Model Agents with Entropic Activation Steering","created":"2024-06-01","cited_by_count":29,"area":"agents"},{"id":"2411.03343","title":"What Features in Prompts Jailbreak LLMs? Investigating the Mechanisms Behind Attacks","created":"2024-11-02","cited_by_count":29,"area":"safety"},{"id":"2502.18862","title":"One-shot Optimized Steering Vectors Mediate Safety-relevant Behaviors in LLMs","created":"2025-02-26","cited_by_count":29,"area":"mechinterp"},{"id":"2505.16188","title":"SAE-SSV: Supervised Steering in Sparse Representation Spaces for Reliable Control of Language Models","created":"2025-05-22","cited_by_count":29,"area":"mechinterp"},{"id":"2505.23556","title":"Understanding Refusal in Language Models with Sparse Autoencoders","created":"2025-05-29","cited_by_count":29,"area":"alignment"},{"id":"2510.13290","title":"To Steer or Not to Steer? Mechanistic Error Reduction with Abstention for Language Models","created":"2025-10-15","cited_by_count":29,"area":null},{"id":"2511.00617","title":"Belief Dynamics Reveal the Dual Nature of In-Context Learning and Activation Steering","created":"2025-11-01","cited_by_count":29,"area":"mechinterp"},{"id":"2412.00967","title":"Linear Probe Penalties Reduce LLM Sycophancy","created":"2024-12-01","cited_by_count":28,"area":"alignment"},{"id":"2410.00153","title":"Beyond Single Concept Vector: Modeling Concept Subspace in LLMs with Gaussian Distribution","created":"2024-09-30","cited_by_count":27,"area":null},{"id":"2411.12405","title":"Evaluating the Prompt Steerability of Large Language Models","created":"2024-11-19","cited_by_count":27,"area":null},{"id":"2510.07364","title":"Base Models Know How to Reason, Thinking Models Learn When","created":"2025-10-08","cited_by_count":27,"area":"lr"},{"id":"2601.14004","title":"Locate, Steer, and Improve: A Practical Survey of Actionable Mechanistic Interpretability in Large Language Models","created":"2026-01-20","cited_by_count":27,"area":"mechinterp"},{"id":"2407.12856","title":"AI-AI Bias: large language models favor communications generated by large language models","created":"2024-07-09","cited_by_count":26,"area":null},{"id":"2410.14516","title":"Do LLMs \"know\" internally when they follow instructions?","created":"2024-10-18","cited_by_count":26,"area":null},{"id":"2507.12638","title":"Reasoning-Finetuning Repurposes Latent Representations in Base Models","created":"2025-07-16","cited_by_count":26,"area":"lr"},{"id":"2602.02132","title":"There Is More to Refusal in Large Language Models than a Single Direction","created":"2026-02-02","cited_by_count":26,"area":"mechinterp"},{"id":"2603.17839","title":"How do LLMs Compute Verbal Confidence","created":"2026-03-18","cited_by_count":26,"area":null},{"id":"2605.05115","title":"Manifold Steering Reveals the Shared Geometry of Neural Network Representation and Behavior","created":"2026-05-06","cited_by_count":26,"area":null}],"next_cursor":"100"}