{"total":922,"rows":[{"id":"2312.06681","title":"Steering Llama 2 via Contrastive Activation Addition","created":"2023-12-09","cited_by_count":1085,"area":"mechinterp"},{"id":"2311.03658","title":"The Linear Representation Hypothesis and the Geometry of Large Language Models","created":"2023-11-07","cited_by_count":716,"area":null},{"id":"2404.14082","title":"Mechanistic Interpretability for AI Safety -- A Review","created":"2024-04-22","cited_by_count":422,"area":"mechinterp"},{"id":"2311.06668","title":"In-context Vectors: Making In Context Learning More Effective and Controllable Through Latent Space Steering","created":"2023-11-11","cited_by_count":245,"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":"2404.03592","title":"ReFT: Representation Finetuning for Language Models","created":"2024-04-04","cited_by_count":197,"area":null},{"id":"2605.29358","title":"Scaling Monosemanticity: Extracting Interpretable Features from Claude 3 Sonnet","created":"2026-05-28","cited_by_count":191,"area":"mechinterp"},{"id":"2407.21771","title":"Paying More Attention to Image: A Training-Free Method for Alleviating Hallucination in LVLMs","created":"2024-07-31","cited_by_count":179,"area":"mllm"},{"id":"2402.16835","title":"Eight Methods to Evaluate Robust Unlearning in LLMs","created":"2024-02-26","cited_by_count":163,"area":"safety"},{"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":"2507.03724","title":"MemOS: A Memory OS for AI System","created":"2025-07-04","cited_by_count":138,"area":"rag"},{"id":"2401.14446","title":"Black-Box Access is Insufficient for Rigorous AI Audits","created":"2024-01-25","cited_by_count":126,"area":null},{"id":"2405.16833","title":"Safe LoRA: the Silver Lining of Reducing Safety Risks when Fine-tuning Large Language Models","created":"2024-05-27","cited_by_count":125,"area":null},{"id":"2402.02416","title":"Aligner: Efficient Alignment by Learning to Correct","created":"2024-02-04","cited_by_count":98,"area":"alignment"},{"id":"2411.11296","title":"Steering Language Model Refusal with Sparse Autoencoders","created":"2024-11-18","cited_by_count":95,"area":"mechinterp"},{"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":"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":"2401.11206","title":"InferAligner: Inference-Time Alignment for Harmlessness through Cross-Model Guidance","created":"2024-01-20","cited_by_count":83,"area":"alignment"},{"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":"2607.15495","title":"Verbalizable Representations Form a Global Workspace in Language Models","created":"2026-07-16","cited_by_count":75,"area":null},{"id":"2404.01650","title":"Test-Time Model Adaptation with Only Forward Passes","created":"2024-04-02","cited_by_count":74,"area":null},{"id":"2505.22101","title":"MemOS: An Operating System for Memory-Augmented Generation (MAG) in Large Language Models","created":"2025-05-28","cited_by_count":73,"area":null},{"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":"2411.05403","title":"Benchmarking Distributional Alignment of Large Language Models","created":"2024-11-08","cited_by_count":61,"area":"alignment"},{"id":"2604.07729","title":"Emotion Concepts and their Function in a Large Language Model","created":"2026-04-09","cited_by_count":57,"area":null},{"id":"2402.01618","title":"Style Vectors for Steering Generative Large Language Model","created":"2024-02-02","cited_by_count":56,"area":null},{"id":"2405.09719","title":"Spectral Editing of Activations for Large Language Model Alignment","created":"2024-05-15","cited_by_count":56,"area":"alignment"},{"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":"2311.08147","title":"RECALL: A Benchmark for LLMs Robustness against External Counterfactual Knowledge","created":"2023-11-14","cited_by_count":51,"area":null},{"id":"2310.17230","title":"Codebook Features: Sparse and Discrete Interpretability for Neural Networks","created":"2023-10-26","cited_by_count":48,"area":null},{"id":"2406.07882","title":"Designing a Dashboard for Transparency and Control of Conversational AI","created":"2024-06-12","cited_by_count":46,"area":null},{"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":"2407.14937","title":"Operationalizing a Threat Model for Red-Teaming Large Language Models (LLMs)","created":"2024-07-20","cited_by_count":46,"area":"safety"},{"id":"2502.12446","title":"Multi-Attribute Steering of Language Models via Targeted Intervention","created":"2025-02-18","cited_by_count":46,"area":null},{"id":"2403.01267","title":"Dissecting Language Models: Machine Unlearning via Selective Pruning","created":"2024-03-02","cited_by_count":42,"area":"safety"},{"id":"2505.13763","title":"Language Models Are Capable of Metacognitive Monitoring and Control of Their Internal Activations","created":"2025-05-19","cited_by_count":41,"area":null},{"id":"2403.05767","title":"Extending Activation Steering to Broad Skills and Multiple Behaviours","created":"2024-03-09","cited_by_count":40,"area":"mechinterp"},{"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":"2402.15179","title":"Advancing Parameter Efficiency in Fine-tuning via Representation Editing","created":"2024-02-23","cited_by_count":37,"area":null},{"id":"2410.04962","title":"Activation Scaling for Steering and Interpreting Language Models","created":"2024-10-07","cited_by_count":35,"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":"2505.22411","title":"Mitigating Overthinking in Large Reasoning Models via Manifold Steering","created":"2025-05-28","cited_by_count":33,"area":"lr"},{"id":"2503.02080","title":"Linear Representations of Political Perspective Emerge in Large Language Models","created":"2025-03-03","cited_by_count":32,"area":"mechinterp"},{"id":"2505.17127","title":"Pixels Versus Priors: Controlling Knowledge Priors in Vision-Language Models through Visual Counterfacts","created":"2025-05-21","cited_by_count":32,"area":"mllm"},{"id":"2504.01550","title":"Representation Bending for Large Language Model Safety","created":"2025-04-02","cited_by_count":30,"area":"safety"},{"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":"2405.16229","title":"No Two Devils Alike: Unveiling Distinct Mechanisms of Fine-tuning Attacks","created":"2024-05-25","cited_by_count":29,"area":"alignment"},{"id":"2406.00244","title":"Controlling Large Language Model Agents with Entropic Activation Steering","created":"2024-06-01","cited_by_count":29,"area":"agents"},{"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":"2401.16332","title":"Tradeoffs Between Alignment and Helpfulness in Language Models with Steering Methods","created":"2024-01-29","cited_by_count":28,"area":"alignment"},{"id":"2410.06981","title":"Quantifying Feature Space Universality Across Large Language Models via Sparse Autoencoders","created":"2024-10-09","cited_by_count":27,"area":"mechinterp"},{"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":"2503.03730","title":"Towards Understanding Distilled Reasoning Models: A Representational Approach","created":"2025-03-05","cited_by_count":26,"area":"lr"},{"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},{"id":"2510.13554","title":"Attention Illuminates LLM Reasoning: The Preplan-and-Anchor Rhythm Enables Fine-Grained Policy Optimization","created":"2025-10-15","cited_by_count":25,"area":"lr"},{"id":"2507.04742","title":"Activation Steering for Chain-of-Thought Compression","created":"2025-07-07","cited_by_count":24,"area":"lr"},{"id":"2511.01689","title":"Open Character Training: Shaping the Persona of AI Assistants through Constitutional AI","created":"2025-11-03","cited_by_count":24,"area":"alignment"},{"id":"2604.05655","title":"LLM Reasoning as Trajectories: Step-Specific Representation Geometry and Correctness Signals","created":"2026-04-07","cited_by_count":24,"area":"lr"},{"id":"2502.17514","title":"SAE-V: Interpreting Multimodal Models for Enhanced Alignment","created":"2025-02-22","cited_by_count":23,"area":"mllm"},{"id":"2601.16644","title":"Sycophancy Hides Linearly in the Attention Heads","created":"2026-01-23","cited_by_count":23,"area":null},{"id":"2602.07852","title":"Emergent Misalignment is Easy, Narrow Misalignment is Hard","created":"2026-02-08","cited_by_count":23,"area":null},{"id":"2310.08043","title":"Understanding and Controlling a Maze-Solving Policy Network","created":"2023-10-12","cited_by_count":22,"area":null},{"id":"2310.16270","title":"Attention Lens: A Tool for Mechanistically Interpreting the Attention Head Information Retrieval Mechanism","created":"2023-10-25","cited_by_count":22,"area":null},{"id":"2402.10688","title":"Towards Uncovering How Large Language Model Works: An Explainability Perspective","created":"2024-02-16","cited_by_count":22,"area":"lr"},{"id":"2409.02877","title":"Configurable Foundation Models: Building LLMs from a Modular Perspective","created":"2024-09-04","cited_by_count":22,"area":null},{"id":"2602.07050","title":"Interpreting Physics in Video World Models","created":"2026-02-04","cited_by_count":22,"area":"wm"},{"id":"2505.13141","title":"Language-Specific Latent Process Hinders Cross-Lingual Performance","created":"2025-05-19","cited_by_count":21,"area":null},{"id":"2505.14827","title":"Text Generation Beyond Discrete Token Sampling","created":"2025-05-20","cited_by_count":21,"area":null},{"id":"2505.20546","title":"Paths Not Taken: Understanding and Mending the Multilingual Factual Recall Pipeline","created":"2025-05-26","cited_by_count":21,"area":null},{"id":"2506.07180","title":"Flattery in Motion: Benchmarking and Analyzing Sycophancy in Video-LLMs","created":"2025-06-08","cited_by_count":21,"area":"lr"},{"id":"2509.22067","title":"The Rogue Scalpel: Activation Steering Compromises LLM Safety","created":"2025-09-26","cited_by_count":21,"area":"safety"},{"id":"2602.06176","title":"Large Language Model Reasoning Failures","created":"2026-02-05","cited_by_count":21,"area":"lr"},{"id":"2507.02778","title":"Self-Correction Bench: Uncovering and Addressing the Self-Correction Blind Spot in Large Language Models","created":"2025-07-03","cited_by_count":20,"area":"lr"},{"id":"2511.04847","title":"Test-Time Adaptation for LLM Agents via Environment Interaction","created":"2025-11-06","cited_by_count":20,"area":"agents"},{"id":"2511.05408","title":"Steering Language Models with Weight Arithmetic","created":"2025-11-07","cited_by_count":20,"area":null},{"id":"2411.08790","title":"Can sparse autoencoders be used to decompose and interpret steering vectors?","created":"2024-11-13","cited_by_count":19,"area":"mechinterp"},{"id":"2502.06258","title":"Emergent Response Planning in LLMs","created":"2025-02-10","cited_by_count":19,"area":"lr"},{"id":"2506.00085","title":"COSMIC: Generalized Refusal Direction Identification in LLM Activations","created":"2025-05-30","cited_by_count":19,"area":null},{"id":"2402.19465","title":"Towards Tracing Trustworthiness Dynamics: Revisiting Pre-training Period of Large Language Models","created":"2024-02-29","cited_by_count":18,"area":null},{"id":"2410.01174","title":"Towards Inference-time Category-wise Safety Steering for Large Language Models","created":"2024-10-02","cited_by_count":18,"area":"alignment"},{"id":"2505.12189","title":"Mitigating Content Effects on Reasoning in Language Models through Fine-Grained Activation Steering","created":"2025-05-18","cited_by_count":18,"area":"lr"},{"id":"2505.18325","title":"Understanding and Mitigating Overrefusal in LLMs from an Unveiling Perspective of Safety Decision Boundary","created":"2025-05-23","cited_by_count":18,"area":"safety"},{"id":"2506.10922","title":"Robustly Improving LLM Fairness in Realistic Settings via Interpretability","created":"2025-06-12","cited_by_count":18,"area":null},{"id":"2502.13095","title":"Understanding and Rectifying Safety Perception Distortion in VLMs","created":"2025-02-18","cited_by_count":17,"area":"mllm"},{"id":"2510.11288","title":"Emergent Misalignment via In-Context Learning: Narrow in-context examples can produce broadly misaligned LLMs","created":"2025-10-13","cited_by_count":17,"area":null},{"id":"2510.11328","title":"Do LLMs \"Feel\"? Emotion Circuits Discovery and Control","created":"2025-10-13","cited_by_count":17,"area":null},{"id":"2405.17345","title":"Exploring and steering the moral compass of Large Language Models","created":"2024-05-27","cited_by_count":16,"area":null},{"id":"2502.00919","title":"Attention Sinks: A 'Catch, Tag, Release' Mechanism for Embeddings","created":"2025-02-02","cited_by_count":16,"area":null},{"id":"2505.18706","title":"Steering LLM Reasoning Through Bias-Only Adaptation","created":"2025-05-24","cited_by_count":16,"area":"lr"},{"id":"2604.03147","title":"Valence-Arousal Subspace in LLMs: Circular Emotion Geometry and Multi-Behavioral Control","created":"2026-04-03","cited_by_count":16,"area":null},{"id":"2605.12400","title":"OGLS-SD: On-Policy Self-Distillation with Outcome-Guided Logit Steering for LLM Reasoning","created":"2026-05-12","cited_by_count":16,"area":"lr"},{"id":"2311.04131","title":"Towards Interpretable Sequence Continuation: Analyzing Shared Circuits in Large Language Models","created":"2023-11-07","cited_by_count":15,"area":"mechinterp"},{"id":"2402.07896","title":"Suppressing Pink Elephants with Direct Principle Feedback","created":"2024-02-12","cited_by_count":15,"area":"alignment"}],"next_cursor":"100"}