{"total":149,"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":"2403.19647","title":"Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models","created":"2024-03-28","cited_by_count":435,"area":null},{"id":"2404.14082","title":"Mechanistic Interpretability for AI Safety -- A Review","created":"2024-04-22","cited_by_count":422,"area":"mechinterp"},{"id":"2407.15549","title":"Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs","created":"2024-07-22","cited_by_count":173,"area":"safety"},{"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":"2403.17806","title":"Have Faith in Faithfulness: Going Beyond Circuit Overlap When Finding Model Mechanisms","created":"2024-03-26","cited_by_count":149,"area":null},{"id":"2407.02646","title":"A Practical Review of Mechanistic Interpretability for Transformer-Based Language Models","created":"2024-07-02","cited_by_count":118,"area":"mechinterp"},{"id":"2411.14257","title":"Do I Know This Entity? Knowledge Awareness and Hallucinations in Language Models","created":"2024-11-21","cited_by_count":115,"area":"mechinterp"},{"id":"2410.21272","title":"Arithmetic Without Algorithms: Language Models Solve Math With a Bag of Heuristics","created":"2024-10-28","cited_by_count":97,"area":"lr"},{"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":"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":"2407.10827","title":"LLM Circuit Analyses Are Consistent Across Training and Scale","created":"2024-07-15","cited_by_count":63,"area":null},{"id":"2406.16778","title":"Finding Transformer Circuits with Edge Pruning","created":"2024-06-24","cited_by_count":60,"area":null},{"id":"2504.13151","title":"MIB: A Mechanistic Interpretability Benchmark","created":"2025-04-17","cited_by_count":56,"area":"mechinterp"},{"id":"2407.10264","title":"What Makes and Breaks Safety Fine-tuning? A Mechanistic Study","created":"2024-07-14","cited_by_count":55,"area":"safety"},{"id":"2406.19501","title":"Monitoring Latent World States in Language Models with Propositional Probes","created":"2024-06-27","cited_by_count":43,"area":"wm"},{"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":"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":"2407.15018","title":"Answer, Assemble, Ace: Understanding How LMs Answer Multiple Choice Questions","created":"2024-07-21","cited_by_count":35,"area":null},{"id":"2503.02854","title":"(How) Do Language Models Track State?","created":"2025-03-04","cited_by_count":34,"area":null},{"id":"2504.02922","title":"Overcoming Sparsity Artifacts in Crosscoders to Interpret Chat-Tuning","created":"2025-04-03","cited_by_count":34,"area":null},{"id":"2505.16381","title":"PaTH Attention: Position Encoding via Accumulating Householder Transformations","created":"2025-05-22","cited_by_count":34,"area":null},{"id":"2405.21068","title":"Code Pretraining Improves Entity Tracking Abilities of Language Models","created":"2024-05-31","cited_by_count":31,"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":"2409.05448","title":"Representational Analysis of Binding in Language Models","created":"2024-09-09","cited_by_count":28,"area":null},{"id":"2510.13900","title":"Narrow Finetuning Leaves Clearly Readable Traces in Activation Differences","created":"2025-10-14","cited_by_count":28,"area":null},{"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":"2510.06182","title":"Mixing Mechanisms: How Language Models Retrieve Bound Entities In-Context","created":"2025-10-07","cited_by_count":26,"area":null},{"id":"2408.01416","title":"The Quest for the Right Mediator: Surveying Mechanistic Interpretability Through the Lens of Causal Mediation Analysis","created":"2024-08-02","cited_by_count":24,"area":"mechinterp"},{"id":"2412.01784","title":"Noise Injection Reveals Hidden Capabilities of Sandbagging Language Models","created":"2024-12-02","cited_by_count":24,"area":null},{"id":"2412.03205","title":"U-MATH: A University-Level Benchmark for Evaluating Mathematical Skills in LLMs","created":"2024-12-04","cited_by_count":24,"area":null},{"id":"2407.14662","title":"Relational Composition in Neural Networks: A Survey and Call to Action","created":"2024-07-19","cited_by_count":23,"area":null},{"id":"2502.06852","title":"EAP-GP: Mitigating Saturation Effect in Gradient-based Automated Circuit Identification","created":"2025-02-07","cited_by_count":21,"area":"mechinterp"},{"id":"2505.20896","title":"How Do Transformers Learn Variable Binding in Symbolic Programs?","created":"2025-05-27","cited_by_count":21,"area":null},{"id":"2502.11812","title":"Towards Understanding Fine-Tuning Mechanisms of LLMs via Circuit Analysis","created":"2025-02-17","cited_by_count":20,"area":null},{"id":"2410.16531","title":"Bayesian scaling laws for in-context learning","created":"2024-10-21","cited_by_count":19,"area":"safety"},{"id":"2507.10616","title":"Scalpel vs. Hammer: GRPO Amplifies Existing Capabilities, SFT Replaces Them","created":"2025-07-13","cited_by_count":15,"area":"lr"},{"id":"2510.00845","title":"Mechanistic Interpretability as Statistical Estimation: A Variance Analysis","created":"2025-10-01","cited_by_count":15,"area":"mechinterp"},{"id":"2502.04577","title":"Position-aware Automatic Circuit Discovery","created":"2025-02-07","cited_by_count":14,"area":"mechinterp"},{"id":"2503.07604","title":"Implicit Reasoning in Transformers is Reasoning through Shortcuts","created":"2025-03-10","cited_by_count":14,"area":"lr"},{"id":"2504.14379","title":"The Geometry of Self-Verification in a Task-Specific Reasoning Model","created":"2025-04-19","cited_by_count":13,"area":"lr"},{"id":"2508.21258","title":"RelP: Faithful and Efficient Circuit Discovery in Language Models via Relevance Patching","created":"2025-08-28","cited_by_count":13,"area":"mechinterp"},{"id":"2410.01434","title":"Circuit Compositions: Exploring Modular Structures in Transformer-Based Language Models","created":"2024-10-02","cited_by_count":11,"area":null},{"id":"2411.10950","title":"Understanding Multimodal LLMs: the Mechanistic Interpretability of Llava in Visual Question Answering","created":"2024-11-17","cited_by_count":11,"area":"mllm"},{"id":"2503.11302","title":"Are formal and functional linguistic mechanisms dissociated in language models?","created":"2025-03-14","cited_by_count":11,"area":null},{"id":"2510.05092","title":"Learning to Interpret Weight Differences in Language Models","created":"2025-10-06","cited_by_count":11,"area":null},{"id":"2403.03942","title":"The Heuristic Core: Understanding Subnetwork Generalization in Pretrained Language Models","created":"2024-03-06","cited_by_count":10,"area":null},{"id":"2410.22316","title":"Understanding Synthetic Context Extension via Retrieval Heads","created":"2024-10-29","cited_by_count":10,"area":"rag"},{"id":"2506.05451","title":"Interpretation Meets Safety: A Survey on Interpretation Methods and Tools for Improving LLM Safety","created":"2025-06-05","cited_by_count":10,"area":"safety"},{"id":"2507.08218","title":"Simple Mechanistic Explanations for Out-Of-Context Reasoning","created":"2025-07-10","cited_by_count":10,"area":"lr"},{"id":"2505.15105","title":"Mechanistic evaluation of Transformers and state space models","created":"2025-05-21","cited_by_count":9,"area":"ssm"},{"id":"2506.12044","title":"Why Do Some Inputs Break Low-Bit LLM Quantization?","created":"2025-05-24","cited_by_count":9,"area":null},{"id":"2410.05565","title":"Chain and Causal Attention for Efficient Entity Tracking","created":"2024-10-07","cited_by_count":8,"area":null},{"id":"2507.21084","title":"Reviving Your MNEME: Predicting The Side Effects of LLM Unlearning and Fine-Tuning via Sparse Model Diffing","created":"2025-06-19","cited_by_count":8,"area":"safety"},{"id":"2508.01916","title":"Decomposing Representation Space into Interpretable Subspaces with Unsupervised Learning","created":"2025-08-03","cited_by_count":8,"area":"mechinterp"},{"id":"2509.25758","title":"Thinking Sparks!: Emergent Attention Heads in Reasoning Models During Post Training","created":"2025-09-30","cited_by_count":8,"area":"lr"},{"id":"2411.16105","title":"Adaptive Circuit Behavior and Generalization in Mechanistic Interpretability","created":"2024-11-25","cited_by_count":7,"area":"mechinterp"},{"id":"2504.04215","title":"Towards Understanding and Improving Refusal in Compressed Models via Mechanistic Interpretability","created":"2025-04-05","cited_by_count":7,"area":"mechinterp"},{"id":"2504.07898","title":"How do Large Language Models Understand Relevance? A Mechanistic Interpretability Perspective","created":"2025-04-10","cited_by_count":7,"area":"mechinterp"},{"id":"2409.03274","title":"Recent Advances in Attack and Defense Approaches of Large Language Models","created":"2024-09-05","cited_by_count":6,"area":null},{"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":"2506.04706","title":"Line of Sight: On Linear Representations in VLLMs","created":"2025-06-05","cited_by_count":6,"area":"mechinterp"},{"id":"2605.30233","title":"Do Language Models Track Entities Across State Changes?","created":"2026-05-28","cited_by_count":6,"area":null},{"id":"2505.22200","title":"Investigating Mechanisms for In-Context Vision Language Binding","created":"2025-05-28","cited_by_count":5,"area":"mllm"},{"id":"2507.07186","title":"Planted in Pretraining, Swayed by Finetuning: A Case Study on the Origins of Cognitive Biases in LLMs","created":"2025-07-09","cited_by_count":5,"area":null},{"id":"2512.16902","title":"In-Context Algebra","created":"2025-12-18","cited_by_count":5,"area":null},{"id":"2601.05201","title":"Mechanisms of Prompt-Induced Hallucination in Vision-Language Models","created":"2026-01-08","cited_by_count":5,"area":"mllm"},{"id":"2601.09445","title":"Where Knowledge Collides: A Mechanistic Study of Intra-Memory Knowledge Conflict in Language Models","created":"2026-01-14","cited_by_count":5,"area":null},{"id":"2604.19052","title":"Cell-Based Representation of Relational Binding in Language Models","created":"2026-04-21","cited_by_count":5,"area":null},{"id":"2410.22370","title":"Survey of User Interface Design and Interaction Techniques in Generative AI Applications","created":"2024-10-28","cited_by_count":4,"area":null},{"id":"2602.15183","title":"Seeing to Generalize: How Visual Data Corrects Binding Shortcuts","created":"2026-02-16","cited_by_count":4,"area":"mllm"},{"id":"2603.22278","title":"The Dual Mechanisms of Spatial Variable Binding in Vision-Language Models","created":"2026-03-23","cited_by_count":4,"area":"mllm"},{"id":"2502.06106","title":"Fine-Tuning is Subgraph Search: A New Lens on Learning Dynamics","created":"2025-02-10","cited_by_count":3,"area":"mechinterp"},{"id":"2503.01896","title":"Neuroplasticity and Corruption in Model Mechanisms: A Case Study Of Indirect Object Identification","created":"2025-02-27","cited_by_count":3,"area":null},{"id":"2505.20076","title":"ExPLAIND: Unifying Model, Data, and Training Attribution to Study Model Behavior","created":"2025-05-26","cited_by_count":3,"area":null},{"id":"2604.24302","title":"Differentiable Faithfulness Alignment for Cross-Model Circuit Transfer","created":"2026-04-27","cited_by_count":3,"area":"mechinterp"},{"id":"2605.28860","title":"Mechanistic origins of catastrophic forgetting: why RL preserves circuits better than SFT?","created":"2026-05-21","cited_by_count":3,"area":null},{"id":"2606.08644","title":"A retrieval conditioned rebinding circuit for dynamic entity tracking in large language models","created":"2026-06-07","cited_by_count":3,"area":"mechinterp"},{"id":"2608.13567","title":"Modular Cognitive Architecture Emerges in Large Language Models","created":"2026-06-27","cited_by_count":3,"area":null},{"id":"2412.17019","title":"Reversed Attention: On The Gradient Descent Of Attention Layers In GPT","created":"2024-12-22","cited_by_count":2,"area":null},{"id":"2505.21191","title":"Unveiling Instruction-Specific Neurons & Experts: An Analytical Framework for LLM's Instruction-Following Capabilities","created":"2025-05-27","cited_by_count":2,"area":"moe"},{"id":"2507.09875","title":"Function Induction and Task Generalization: An Interpretability Study with Off-by-One Addition","created":"2025-07-14","cited_by_count":2,"area":null},{"id":"2510.07452","title":"PATCH: Mitigating PII Leakage in Language Models with Privacy-Aware Targeted Circuit PatcHing","created":"2025-10-08","cited_by_count":2,"area":null},{"id":"2510.17256","title":"Explainability of Large Language Models: Opportunities and Challenges toward Generating Trustworthy Explanations","created":"2025-10-20","cited_by_count":2,"area":"mechinterp"},{"id":"2603.20101","title":"Pitfalls in Evaluating Interpretability Agents","created":"2026-03-20","cited_by_count":2,"area":"agents"},{"id":"2605.09129","title":"Data-driven Circuit Discovery for Interpretability of Language Models","created":"2026-05-09","cited_by_count":2,"area":"mechinterp"},{"id":"2606.29522","title":"When Does Activation Steering Change What a Model Computes From?","created":"2026-06-28","cited_by_count":2,"area":"mechinterp"},{"id":"2607.18921","title":"Circuit Claims Depend on What Is Extracted and How It Is Compared","created":"2026-07-21","cited_by_count":2,"area":null},{"id":"2407.10949","title":"Representing Rule-based Chatbots with Transformers","created":"2024-07-15","cited_by_count":1,"area":null},{"id":"2502.10886","title":"MET-Bench: Multimodal Entity Tracking for Evaluating the Limitations of Vision-Language and Reasoning Models","created":"2025-02-15","cited_by_count":1,"area":"wm"},{"id":"2505.16178","title":"Understanding Fact Recall in Language Models: Why Two-Stage Training Encourages Memorization but Mixed Training Teaches Knowledge","created":"2025-05-22","cited_by_count":1,"area":null},{"id":"2505.17091","title":"Large Language Models Implicitly Learn to See and Hear Just By Reading","created":"2025-05-20","cited_by_count":1,"area":null},{"id":"2512.16914","title":"Constructive Circuit Amplification: Improving Math Reasoning in LLMs via Targeted Sub-Network Updates","created":"2025-12-18","cited_by_count":1,"area":"lr"},{"id":"2602.20904","title":"Transcoder Adapters for Reasoning-Model Diffing","created":"2026-02-24","cited_by_count":1,"area":"lr"},{"id":"2604.21139","title":"Slot Machines: How LLMs Keep Track of Multiple Entities","created":"2026-04-22","cited_by_count":1,"area":null},{"id":"2605.22476","title":"Structured-Sparse Attention for Entity Tracking with Subquadratic Sequence Complexity","created":"2026-05-21","cited_by_count":1,"area":null}],"next_cursor":"100"}