{"total":184,"rows":[{"id":"2304.14997","title":"Towards Automated Circuit Discovery for Mechanistic Interpretability","created":"2023-04-28","cited_by_count":724,"area":"mechinterp"},{"id":"2208.01066","title":"What Can Transformers Learn In-Context? A Case Study of Simple Function Classes","created":"2022-08-01","cited_by_count":715,"area":null},{"id":"2410.05229","title":"GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models","created":"2024-10-07","cited_by_count":550,"area":"lr"},{"id":"2305.19466","title":"The Impact of Positional Encoding on Length Generalization in Transformers","created":"2023-05-31","cited_by_count":327,"area":null},{"id":"2305.15408","title":"Towards Revealing the Mystery behind Chain of Thought: A Theoretical Perspective","created":"2023-05-24","cited_by_count":294,"area":"lr"},{"id":"2210.10749","title":"Transformers Learn Shortcuts to Automata","created":"2022-10-19","cited_by_count":268,"area":null},{"id":"2402.12875","title":"Chain of Thought Empowers Transformers to Solve Inherently Serial Problems","created":"2024-02-20","cited_by_count":260,"area":"lr"},{"id":"2301.13196","title":"Looped Transformers as Programmable Computers","created":"2023-01-30","cited_by_count":237,"area":null},{"id":"2207.02098","title":"Neural Networks and the Chomsky Hierarchy","created":"2022-07-05","cited_by_count":217,"area":null},{"id":"2310.16028","title":"What Algorithms can Transformers Learn? A Study in Length Generalization","created":"2023-10-24","cited_by_count":199,"area":null},{"id":"2302.03025","title":"A Toy Model of Universality: Reverse Engineering How Networks Learn Group Operations","created":"2023-02-06","cited_by_count":144,"area":"mechinterp"},{"id":"2106.16213","title":"Saturated Transformers are Constant-Depth Threshold Circuits","created":"2021-06-30","cited_by_count":139,"area":null},{"id":"2311.00208","title":"What Formal Languages Can Transformers Express? A Survey","created":"2023-11-01","cited_by_count":124,"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":"2311.12786","title":"Mechanistically analyzing the effects of fine-tuning on procedurally defined tasks","created":"2023-11-21","cited_by_count":115,"area":null},{"id":"2202.12172","title":"Overcoming a Theoretical Limitation of Self-Attention","created":"2022-02-24","cited_by_count":113,"area":null},{"id":"2301.05062","title":"Tracr: Compiled Transformers as a Laboratory for Interpretability","created":"2023-01-12","cited_by_count":104,"area":null},{"id":"2406.04267","title":"Transformers need glasses! Information over-squashing in language tasks","created":"2024-06-06","cited_by_count":91,"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":"2402.09963","title":"Why are Sensitive Functions Hard for Transformers?","created":"2024-02-15","cited_by_count":74,"area":null},{"id":"2210.02671","title":"A Logic for Expressing Log-Precision Transformers","created":"2022-10-06","cited_by_count":67,"area":"lr"},{"id":"2301.10743","title":"Tighter Bounds on the Expressivity of Transformer Encoders","created":"2023-01-25","cited_by_count":66,"area":null},{"id":"2509.09677","title":"The Illusion of Diminishing Returns: Measuring Long Horizon Execution in LLMs","created":"2025-09-11","cited_by_count":64,"area":null},{"id":"2110.07732","title":"The Neural Data Router: Adaptive Control Flow in Transformers Improves Systematic Generalization","created":"2021-10-14","cited_by_count":62,"area":null},{"id":"2406.16778","title":"Finding Transformer Circuits with Edge Pruning","created":"2024-06-24","cited_by_count":60,"area":null},{"id":"2310.07923","title":"The Expressive Power of Transformers with Chain of Thought","created":"2023-10-11","cited_by_count":50,"area":"lr"},{"id":"2310.13897","title":"Masked Hard-Attention Transformers Recognize Exactly the Star-Free Languages","created":"2023-10-21","cited_by_count":49,"area":null},{"id":"2511.13653","title":"Weight-sparse transformers have interpretable circuits","created":"2025-11-17","cited_by_count":44,"area":"mechinterp"},{"id":"2310.19698","title":"When Do Prompting and Prefix-Tuning Work? A Theory of Capabilities and Limitations","created":"2023-10-30","cited_by_count":43,"area":null},{"id":"2306.01128","title":"Learning Transformer Programs","created":"2023-06-01","cited_by_count":41,"area":"mechinterp"},{"id":"2307.11353","title":"What can a Single Attention Layer Learn? A Study Through the Random Features Lens","created":"2023-07-21","cited_by_count":40,"area":null},{"id":"2402.04161","title":"Attention with Markov: A Framework for Principled Analysis of Transformers via Markov Chains","created":"2024-02-06","cited_by_count":40,"area":null},{"id":"2407.01178","title":"$\\text{Memory}^3$: Language Modeling with Explicit Memory","created":"2024-07-01","cited_by_count":40,"area":"rag"},{"id":"2410.02984","title":"Differentiation and Specialization of Attention Heads via the Refined Local Learning Coefficient","created":"2024-10-03","cited_by_count":40,"area":null},{"id":"2404.04393","title":"Counting Like Transformers: Compiling Temporal Counting Logic Into Softmax Transformers","created":"2024-04-05","cited_by_count":39,"area":null},{"id":"2310.03817","title":"Logical Languages Accepted by Transformer Encoders with Hard Attention","created":"2023-10-05","cited_by_count":38,"area":null},{"id":"2502.19249","title":"Between Circuits and Chomsky: Pre-pretraining on Formal Languages Imparts Linguistic Biases","created":"2025-02-26","cited_by_count":37,"area":null},{"id":"2406.06467","title":"How Far Can Transformers Reason? The Globality Barrier and Inductive Scratchpad","created":"2024-06-10","cited_by_count":36,"area":null},{"id":"2404.14994","title":"Transformers Can Represent $n$-gram Language Models","created":"2024-04-23","cited_by_count":32,"area":null},{"id":"2312.01429","title":"Transformers are uninterpretable with myopic methods: a case study with bounded Dyck grammars","created":"2023-12-03","cited_by_count":30,"area":null},{"id":"2402.03902","title":"A phase transition between positional and semantic learning in a solvable model of dot-product attention","created":"2024-02-06","cited_by_count":30,"area":null},{"id":"2406.09347","title":"Separations in the Representational Capabilities of Transformers and Recurrent Architectures","created":"2024-06-13","cited_by_count":29,"area":null},{"id":"2410.13032","title":"Hypothesis Testing the Circuit Hypothesis in LLMs","created":"2024-10-16","cited_by_count":29,"area":"mechinterp"},{"id":"2407.17686","title":"Transformers on Markov Data: Constant Depth Suffices","created":"2024-07-25","cited_by_count":27,"area":null},{"id":"2312.03656","title":"Interpretability Illusions in the Generalization of Simplified Models","created":"2023-12-06","cited_by_count":25,"area":null},{"id":"2405.20131","title":"Language Models Need Inductive Biases to Count Inductively","created":"2024-05-30","cited_by_count":25,"area":null},{"id":"2407.03310","title":"Universal Length Generalization with Turing Programs","created":"2024-07-03","cited_by_count":25,"area":null},{"id":"2407.14494","title":"InterpBench: Semi-Synthetic Transformers for Evaluating Mechanistic Interpretability Techniques","created":"2024-07-19","cited_by_count":25,"area":"mechinterp"},{"id":"2410.09087","title":"Mechanistic?","created":"2024-10-07","cited_by_count":25,"area":"mechinterp"},{"id":"2405.20671","title":"Position Coupling: Improving Length Generalization of Arithmetic Transformers Using Task Structure","created":"2024-05-31","cited_by_count":23,"area":null},{"id":"2311.12997","title":"Compositional Capabilities of Autoregressive Transformers: A Study on Synthetic, Interpretable Tasks","created":"2023-11-21","cited_by_count":22,"area":null},{"id":"2402.04875","title":"On Provable Length and Compositional Generalization","created":"2024-02-07","cited_by_count":22,"area":null},{"id":"2407.08734","title":"Transformer Circuit Faithfulness Metrics are not Robust","created":"2024-07-11","cited_by_count":22,"area":"mechinterp"},{"id":"2502.07827","title":"Implicit Language Models are RNNs: Balancing Parallelization and Expressivity","created":"2025-02-10","cited_by_count":22,"area":"lr"},{"id":"2410.19034","title":"Mixture of Parrots: Experts improve memorization more than reasoning","created":"2024-10-24","cited_by_count":21,"area":"moe"},{"id":"2504.14370","title":"Density Measures for Language Generation","created":"2025-04-19","cited_by_count":21,"area":null},{"id":"2310.16076","title":"Practical Computational Power of Linear Transformers and Their Recurrent and Self-Referential Extensions","created":"2023-10-24","cited_by_count":20,"area":null},{"id":"2308.00189","title":"Generative Models as a Complex Systems Science: How can we make sense of large language model behavior?","created":"2023-07-31","cited_by_count":19,"area":null},{"id":"2404.01601","title":"What Can Transformer Learn with Varying Depth? Case Studies on Sequence Learning Tasks","created":"2024-04-02","cited_by_count":19,"area":null},{"id":"2410.09375","title":"Looped ReLU MLPs May Be All You Need as Practical Programmable Computers","created":"2024-10-12","cited_by_count":19,"area":null},{"id":"2405.03207","title":"A Philosophical Introduction to Language Models - Part II: The Way Forward","created":"2024-05-06","cited_by_count":18,"area":null},{"id":"2408.15138","title":"How transformers learn structured data: insights from hierarchical filtering","created":"2024-08-27","cited_by_count":18,"area":null},{"id":"2307.01189","title":"Trainable Transformer in Transformer","created":"2023-07-03","cited_by_count":17,"area":null},{"id":"2410.07432","title":"Can Transformers Reason Logically? A Study in SAT Solving","created":"2024-10-09","cited_by_count":16,"area":"lr"},{"id":"2401.09074","title":"Code Simulation Challenges for Large Language Models","created":"2024-01-17","cited_by_count":15,"area":null},{"id":"2403.09053","title":"Towards a theory of model distillation","created":"2024-03-14","cited_by_count":15,"area":null},{"id":"2405.12522","title":"Sparse Autoencoders Enable Scalable and Reliable Circuit Identification in Language Models","created":"2024-05-21","cited_by_count":15,"area":"mechinterp"},{"id":"2410.03170","title":"Autoregressive Large Language Models are Computationally Universal","created":"2024-10-04","cited_by_count":15,"area":null},{"id":"2402.02212","title":"A Survey to Recent Progress Towards Understanding In-Context Learning","created":"2024-02-03","cited_by_count":14,"area":null},{"id":"2408.04532","title":"How Transformers Utilize Multi-Head Attention in In-Context Learning? A Case Study on Sparse Linear Regression","created":"2024-08-08","cited_by_count":14,"area":null},{"id":"2310.17639","title":"In-Context Learning Dynamics with Random Binary Sequences","created":"2023-10-26","cited_by_count":13,"area":null},{"id":"2402.05785","title":"Limits of Transformer Language Models on Learning to Compose Algorithms","created":"2024-02-08","cited_by_count":13,"area":null},{"id":"2407.01214","title":"Revisiting Random Walks for Learning on Graphs","created":"2024-07-01","cited_by_count":13,"area":null},{"id":"2511.05295","title":"Language Generation and Identification From Partial Enumeration: Tight Density Bounds and Topological Characterizations","created":"2025-11-07","cited_by_count":13,"area":null},{"id":"2304.01665","title":"Mastering Symbolic Operations: Augmenting Language Models with Compiled Neural Networks","created":"2023-04-04","cited_by_count":12,"area":"lr"},{"id":"2506.16055","title":"Knee-Deep in C-RASP: A Transformer Depth Hierarchy","created":"2025-06-19","cited_by_count":12,"area":null},{"id":"2210.12786","title":"When Can Transformers Ground and Compose: Insights from Compositional Generalization Benchmarks","created":"2022-10-23","cited_by_count":11,"area":null},{"id":"2408.05506","title":"Your Context Is Not an Array: Unveiling Random Access Limitations in Transformers","created":"2024-08-10","cited_by_count":11,"area":null},{"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":"2402.11628","title":"Discrete Neural Algorithmic Reasoning","created":"2024-02-18","cited_by_count":9,"area":null},{"id":"2405.05966","title":"Natural Language Processing RELIES on Linguistics","created":"2024-05-09","cited_by_count":9,"area":null},{"id":"2505.08140","title":"Lost in Transmission: When and Why LLMs Fail to Reason Globally","created":"2025-05-13","cited_by_count":9,"area":"lr"},{"id":"2511.20038","title":"Softmax Transformers are Turing-Complete","created":"2025-11-25","cited_by_count":9,"area":"lr"},{"id":"2404.02040","title":"Transformers as Transducers","created":"2024-04-02","cited_by_count":8,"area":null},{"id":"2410.01565","title":"Bayes' Power for Explaining In-Context Learning Generalizations","created":"2024-10-02","cited_by_count":8,"area":null},{"id":"2410.11474","title":"How Transformers Get Rich: Approximation and Dynamics Analysis","created":"2024-10-15","cited_by_count":8,"area":null},{"id":"2410.13779","title":"The Mystery of the Pathological Path-star Task for Language Models","created":"2024-10-17","cited_by_count":8,"area":null},{"id":"2502.16792","title":"The Role of Sparsity for Length Generalization in Transformers","created":"2025-02-24","cited_by_count":8,"area":null},{"id":"2503.16048","title":"Meta-Learning Neural Mechanisms rather than Bayesian Priors","created":"2025-03-20","cited_by_count":8,"area":null},{"id":"2508.01916","title":"Decomposing Representation Space into Interpretable Subspaces with Unsupervised Learning","created":"2025-08-03","cited_by_count":8,"area":"mechinterp"},{"id":"2510.27015","title":"Quantitative Bounds for Length Generalization in Transformers","created":"2025-10-30","cited_by_count":8,"area":null},{"id":"2406.09160","title":"Beyond the Frontier: Predicting Unseen Walls from Occupancy Grids by Learning from Floor Plans","created":"2024-06-13","cited_by_count":7,"area":null},{"id":"2410.18077","title":"ALTA: Compiler-Based Analysis of Transformers","created":"2024-10-23","cited_by_count":7,"area":null},{"id":"2502.03568","title":"Code Simulation as a Proxy for High-order Tasks in Large Language Models","created":"2025-02-05","cited_by_count":7,"area":null},{"id":"2508.05776","title":"Whither symbols in the era of advanced neural networks?","created":"2025-08-07","cited_by_count":7,"area":null},{"id":"2404.11041","title":"On the Empirical Complexity of Reasoning and Planning in LLMs","created":"2024-04-17","cited_by_count":6,"area":"lr"},{"id":"2407.11542","title":"Counting in Small Transformers: The Delicate Interplay between Attention and Feed-Forward Layers","created":"2024-07-16","cited_by_count":6,"area":null},{"id":"2509.22445","title":"Bridging Kolmogorov Complexity and Deep Learning: Asymptotically Optimal Description Length Objectives for Transformers","created":"2025-09-26","cited_by_count":6,"area":null},{"id":"2510.00368","title":"The Transformer Cookbook","created":"2025-10-01","cited_by_count":6,"area":null},{"id":"2510.14095","title":"Unlocking Out-of-Distribution Generalization in Transformers via Recursive Latent Space Reasoning","created":"2025-10-15","cited_by_count":6,"area":"lr"}],"next_cursor":"100"}