{"total":206,"rows":[{"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":"2510.26243","title":"Angular Steering: Behavior Control via Rotation in Activation Space","created":"2025-10-30","cited_by_count":53,"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":"2502.02716","title":"A Unified Understanding and Evaluation of Steering Methods","created":"2025-02-04","cited_by_count":46,"area":"mechinterp"},{"id":"2503.05021","title":"Safety is Not Only About Refusal: Reasoning-Enhanced Fine-tuning for Interpretable LLM Safety","created":"2025-03-06","cited_by_count":39,"area":"lr"},{"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":"2509.10625","title":"No Answer Needed: Predicting LLM Answer Accuracy from Question-Only Linear Probes","created":"2025-09-12","cited_by_count":38,"area":"mechinterp"},{"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":"2411.12405","title":"Evaluating the Prompt Steerability of Large Language Models","created":"2024-11-19","cited_by_count":27,"area":null},{"id":"2410.10343","title":"Locking Down the Finetuned LLMs Safety","created":"2024-10-14","cited_by_count":22,"area":null},{"id":"2502.17601","title":"Representation Engineering for Large-Language Models: Survey and Research Challenges","created":"2025-02-24","cited_by_count":21,"area":"mechinterp"},{"id":"2505.12025","title":"Spotlight Your Instructions: Instruction-following with Dynamic Attention Steering","created":"2025-05-17","cited_by_count":21,"area":null},{"id":"2602.08169","title":"Spherical Steering: Geometry-Aware Activation Rotation for Language Models","created":"2026-02-09","cited_by_count":21,"area":null},{"id":"2506.00085","title":"COSMIC: Generalized Refusal Direction Identification in LLM Activations","created":"2025-05-30","cited_by_count":19,"area":null},{"id":"2602.17560","title":"ODESteer: A Unified ODE-Based Steering Framework for LLM Alignment","created":"2026-02-19","cited_by_count":19,"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":"2603.19469","title":"A Framework for Formalizing LLM Agent Security","created":"2026-03-19","cited_by_count":16,"area":"agents"},{"id":"2501.18532","title":"Differentially Private Steering for Large Language Model Alignment","created":"2025-01-30","cited_by_count":15,"area":"alignment"},{"id":"2508.12815","title":"Learning to Steer: Input-dependent Steering for Multimodal LLMs","created":"2025-08-18","cited_by_count":15,"area":"mllm"},{"id":"2510.04309","title":"Activation Steering with a Feedback Controller","created":"2025-10-05","cited_by_count":15,"area":"alignment"},{"id":"2412.06748","title":"Refusal Tokens: A Simple Way to Calibrate Refusals in Large Language Models","created":"2024-12-09","cited_by_count":14,"area":null},{"id":"2504.17130","title":"Steering the CensorShip: Uncovering Representation Vectors for LLM \"Thought\" Control","created":"2025-04-23","cited_by_count":14,"area":"lr"},{"id":"2507.10587","title":"Anthropomimetic Uncertainty: What Verbalized Uncertainty in Language Models is Missing","created":"2025-07-11","cited_by_count":14,"area":null},{"id":"2509.22820","title":"MMPB: It's Time for Multi-Modal Personalization","created":"2025-09-26","cited_by_count":14,"area":null},{"id":"2509.25175","title":"EasySteer: A Unified Framework for High-Performance and Extensible LLM Steering","created":"2025-09-29","cited_by_count":13,"area":null},{"id":"2511.10222","title":"Speech-Audio Compositional Attacks on Multimodal LLMs and Their Mitigation with SALMONN-Guard","created":"2025-11-13","cited_by_count":13,"area":"safety"},{"id":"2509.13281","title":"RepIt: Steering Language Models with Concept-Specific Refusal Vectors","created":"2025-09-16","cited_by_count":12,"area":null},{"id":"2511.06419","title":"MONICA: Real-Time Monitoring and Calibration of Chain-of-Thought Sycophancy in Large Reasoning Models","created":"2025-11-09","cited_by_count":12,"area":"lr"},{"id":"2603.01097","title":"Understanding LoRA as Knowledge Memory: An Empirical Analysis","created":"2026-03-01","cited_by_count":12,"area":null},{"id":"2507.08799","title":"KV Cache Steering for Controlling Frozen LLMs","created":"2025-07-11","cited_by_count":11,"area":"inference"},{"id":"2603.09313","title":"Curveball Steering: The Right Direction To Steer Isn't Always Linear","created":"2026-03-10","cited_by_count":11,"area":"mechinterp"},{"id":"2502.09755","title":"Jailbreak Attack Initializations as Extractors of Compliance Directions","created":"2025-02-13","cited_by_count":10,"area":"safety"},{"id":"2510.04217","title":"MLLMEraser: Achieving Test-Time Unlearning in Multimodal Large Language Models through Activation Steering","created":"2025-10-05","cited_by_count":10,"area":"mllm"},{"id":"2505.14071","title":"Textual Steering Vectors Can Improve Visual Understanding in Multimodal Large Language Models","created":"2025-05-20","cited_by_count":9,"area":"mllm"},{"id":"2508.05775","title":"Guardians and Offenders: A Survey on Harmful Content Generation and Safety Mitigation of LLM","created":"2025-08-07","cited_by_count":9,"area":"alignment"},{"id":"2512.24143","title":"Activation Steering for Masked Diffusion Language Models","created":"2025-12-30","cited_by_count":9,"area":"diffusion"},{"id":"2602.08621","title":"Sparse Models, Sparse Safety: Unsafe Routes in Mixture-of-Experts LLMs","created":"2026-02-09","cited_by_count":9,"area":"moe"},{"id":"2604.19018","title":"Local Linearity of LLMs Enables Activation Steering via Model-Based Linear Optimal Control","created":"2026-04-21","cited_by_count":9,"area":"alignment"},{"id":"2505.20309","title":"Guiding Giants: Lightweight Controllers for Weighted Activation Steering in LLMs","created":"2025-05-22","cited_by_count":8,"area":"mechinterp"},{"id":"2508.17621","title":"Steering When Necessary: Flexible Steering Large Language Models with Backtracking","created":"2025-08-25","cited_by_count":8,"area":"mechinterp"},{"id":"2602.04896","title":"Steering Externalities: Benign Activation Steering Unintentionally Increases Jailbreak Risk for Large Language Models","created":"2026-02-03","cited_by_count":8,"area":"alignment"},{"id":"2604.09839","title":"Steered LLM Activations are Non-Surjective","created":"2026-04-10","cited_by_count":8,"area":"mechinterp"},{"id":"2502.19721","title":"Unsupervised Concept Vector Extraction for Bias Control in LLMs","created":"2025-02-27","cited_by_count":7,"area":"mechinterp"},{"id":"2503.02989","title":"Effectively Steer LLM To Follow Preference via Building Confident Directions","created":"2025-03-04","cited_by_count":7,"area":"alignment"},{"id":"2509.07983","title":"Steering Protein Language Models","created":"2025-07-01","cited_by_count":7,"area":null},{"id":"2604.08524","title":"What Drives Representation Steering? A Mechanistic Case Study on Steering Refusal","created":"2026-04-09","cited_by_count":7,"area":"alignment"},{"id":"2502.13946","title":"Why Safeguarded Ships Run Aground? Aligned Large Language Models' Safety Mechanisms Tend to Be Anchored in The Template Region","created":"2025-02-19","cited_by_count":6,"area":"alignment"},{"id":"2505.14185","title":"Safety Subspaces are Not Linearly Distinct: A Fine-Tuning Case Study","created":"2025-05-20","cited_by_count":6,"area":"alignment"},{"id":"2505.16170","title":"When Do LLMs Admit Their Mistakes? Understanding The Role Of Model Belief In Retraction","created":"2025-05-22","cited_by_count":6,"area":null},{"id":"2509.23782","title":"Bridging the Knowledge-Prediction Gap in LLMs on Multiple-Choice Questions","created":"2025-09-28","cited_by_count":6,"area":null},{"id":"2510.05498","title":"Prototype-Based Dynamic Steering for Large Language Models","created":"2025-10-07","cited_by_count":6,"area":"lr"},{"id":"2511.18284","title":"What Can We Actually Steer? A Multi-Behavior Study of Activation Control","created":"2025-11-23","cited_by_count":6,"area":"mechinterp"},{"id":"2512.16602","title":"Refusal Steering: Fine-grained Control over LLM Refusal Behaviour for Sensitive Topics","created":"2025-12-18","cited_by_count":6,"area":"safety"},{"id":"2601.09269","title":"RISER: Orchestrating Latent Reasoning Skills for Adaptive Activation Steering","created":"2026-01-14","cited_by_count":6,"area":"lr"},{"id":"2602.04935","title":"ASA: Backbone-Training-Free Representation Engineering for Tool-Calling Agents","created":"2026-02-04","cited_by_count":6,"area":"agents"},{"id":"2603.06745","title":"Enhancing Instruction Following of LLMs via Activation Steering with Dynamic Rejection","created":"2026-03-06","cited_by_count":6,"area":"mechinterp"},{"id":"2605.05892","title":"Beyond Steering Vector: Flow-based Activation Steering for Inference-Time Intervention","created":"2026-05-07","cited_by_count":6,"area":"mechinterp"},{"id":"2605.05980","title":"TACT: Mitigating Overthinking and Overacting in Coding Agents via Activation Steering","created":"2026-05-07","cited_by_count":6,"area":"lr"},{"id":"2506.21590","title":"Representation Consistency for Accurate and Coherent LLM Answer Aggregation","created":"2025-06-18","cited_by_count":5,"area":"lr"},{"id":"2509.13664","title":"Sparse Neurons Carry Strong Signals of Question Ambiguity in LLMs","created":"2025-09-17","cited_by_count":5,"area":null},{"id":"2510.08646","title":"Mitigating Over-Refusal in Aligned Large Language Models via Inference-Time Activation Energy","created":"2025-10-09","cited_by_count":5,"area":"alignment"},{"id":"2512.23260","title":"Interpretable Safety Alignment via SAE-Constructed Low-Rank Subspace Adaptation","created":"2025-12-29","cited_by_count":5,"area":"alignment"},{"id":"2601.19768","title":"GAVEL: Towards Rule-Based Safety Through Activation Monitoring","created":"2026-01-27","cited_by_count":5,"area":null},{"id":"2603.13249","title":"Steering at the Source: Style Modulation Heads for Robust Persona Control","created":"2026-02-24","cited_by_count":5,"area":"mechinterp"},{"id":"2605.01167","title":"Minimizing Collateral Damage in Activation Steering","created":"2026-05-01","cited_by_count":5,"area":"mechinterp"},{"id":"2605.01844","title":"The Cylindrical Representation Hypothesis for Language Model Steering","created":"2026-05-03","cited_by_count":5,"area":null},{"id":"2605.20745","title":"The Hidden Signal of Verifier Strictness: Controlling and Improving Step-Wise Verification via Selective Latent Steering","created":"2026-05-20","cited_by_count":5,"area":null},{"id":"2502.16174","title":"Efficient LLM Moderation with Multi-Layer Latent Prototypes","created":"2025-02-22","cited_by_count":4,"area":"safety"},{"id":"2601.19847","title":"Identifying and Transferring Reasoning-Critical Neurons: Improving LLM Inference Reliability via Activation Steering","created":"2026-01-27","cited_by_count":4,"area":"lr"},{"id":"2601.21647","title":"ILRR: Inference-Time Steering Method for Masked Diffusion Language Models","created":"2026-01-29","cited_by_count":4,"area":"diffusion"},{"id":"2604.08169","title":"Activation Steering for Aligned Open-ended Generation without Sacrificing Coherence","created":"2026-04-09","cited_by_count":4,"area":"safety"},{"id":"2507.08967","title":"Self-Improving Model Steering","created":"2025-07-11","cited_by_count":3,"area":"synth"},{"id":"2602.07013","title":"Steering to Say No: Configurable Refusal via Activation Steering in Vision Language Models","created":"2026-01-31","cited_by_count":3,"area":"mllm"},{"id":"2603.03163","title":"Conditioned Activation Transport for T2I Safety Steering","created":"2026-03-03","cited_by_count":3,"area":"mechinterp"},{"id":"2603.08486","title":"Visual Self-Fulfilling Alignment: Shaping Safety-Oriented Personas via Threat-Related Images","created":"2026-03-09","cited_by_count":3,"area":"mllm"},{"id":"2604.03867","title":"Where to Steer: Input-Dependent Layer Selection for Steering Improves LLM Alignment","created":"2026-04-04","cited_by_count":3,"area":"alignment"},{"id":"2605.05957","title":"Knowing but Not Correcting: Routine Task Requests Suppress Factual Correction in LLMs","created":"2026-05-07","cited_by_count":3,"area":null},{"id":"2605.06342","title":"Don't Lose Focus: Activation Steering via Key-Orthogonal Projections","created":"2026-05-07","cited_by_count":3,"area":"mechinterp"},{"id":"2606.11599","title":"When is Your LLM Steerable?","created":"2026-06-10","cited_by_count":3,"area":"mechinterp"},{"id":"2606.29824","title":"Neural Procedural Memory: Empowering LLM Agents with Implicit Activation Steering","created":"2026-06-29","cited_by_count":3,"area":"agents"},{"id":"2607.14943","title":"Steering Robustness into World Action Models via Mechanistic Interpretability and Optimal Control","created":"2026-07-16","cited_by_count":3,"area":"wm"},{"id":"2607.25270","title":"Where Steering Signals Come From: Activation Source Selection in Activation Steering","created":"2026-07-28","cited_by_count":3,"area":"mechinterp"},{"id":"2608.08829","title":"Deployable Per-Instance Multi-Layer Activation Steering for Large Language Models","created":"2026-08-09","cited_by_count":3,"area":"mechinterp"},{"id":"2412.07430","title":"Knowledge Graph Guided Evaluation of Abstention Techniques","created":"2024-12-10","cited_by_count":2,"area":null},{"id":"2412.18497","title":"Neuron-Level Differentiation of Memorization and Generalization in Large Language Models","created":"2024-12-24","cited_by_count":2,"area":null},{"id":"2506.05628","title":"GP-MoLFormer-Sim: Test Time Molecular Optimization through Contextual Similarity Guidance","created":"2025-06-05","cited_by_count":2,"area":null},{"id":"2507.05158","title":"Steering Information Utility in Key-Value Memory for Language Model Post-Training","created":"2025-07-07","cited_by_count":2,"area":null},{"id":"2509.12065","title":"Steering Language Models in Multi-Token Generation: A Case Study on Tense and Aspect","created":"2025-09-15","cited_by_count":2,"area":null},{"id":"2509.22739","title":"Painless Activation Steering: An Automated, Lightweight Approach for Post-Training Large Language Models","created":"2025-09-25","cited_by_count":2,"area":"mechinterp"},{"id":"2511.19131","title":"Eliciting Chain-of-Thought in Base LLMs via Gradient-Based Representation Optimization","created":"2025-11-24","cited_by_count":2,"area":"lr"},{"id":"2602.08901","title":"Mitigating Memorization Where It Happens","created":"2026-02-09","cited_by_count":2,"area":null},{"id":"2603.07837","title":"AI Steerability 360: A Toolkit for Steering Large Language Models","created":"2026-03-08","cited_by_count":2,"area":null},{"id":"2603.14278","title":"Activation Surgery: Jailbreaking White-box LLMs without Touching the Prompt","created":"2026-03-15","cited_by_count":2,"area":"safety"},{"id":"2603.27240","title":"Diagnosing and Repairing Unsafe Channels in Vision-Language Models via Causal Discovery and Dual-Modal Safety Subspace Projection","created":"2026-03-28","cited_by_count":2,"area":"mllm"},{"id":"2604.01202","title":"Therefore I am. I Think","created":"2026-04-01","cited_by_count":2,"area":"lr"},{"id":"2604.06233","title":"Blind Refusal: Language Models Refuse to Help Users Evade Unjust, Absurd, and Illegitimate Rules","created":"2026-04-03","cited_by_count":2,"area":null},{"id":"2604.15488","title":"FineSteer: A Unified Framework for Fine-Grained Inference-Time Steering in Large Language Models","created":"2026-04-16","cited_by_count":2,"area":"mechinterp"},{"id":"2605.06225","title":"Memory Inception: Latent-Space KV Cache Manipulation for Steering LLMs","created":"2026-05-07","cited_by_count":2,"area":"inference"},{"id":"2605.08348","title":"How Much Do Circuits Tell Us? Measuring the Consistency and Specificity of Language Model Circuits","created":"2026-05-08","cited_by_count":2,"area":"mechinterp"}],"next_cursor":"100"}