{"total":92,"rows":[{"id":"2304.01373","title":"Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling","created":"2023-04-03","cited_by_count":1786,"area":null},{"id":"2108.13264","title":"Deep Reinforcement Learning at the Edge of the Statistical Precipice","created":"2021-08-30","cited_by_count":807,"area":null},{"id":"2202.00622","title":"Datamodels: Predicting Predictions from Training Data","created":"2022-02-01","cited_by_count":189,"area":null},{"id":"2305.13169","title":"A Pretrainer's Guide to Training Data: Measuring the Effects of Data Age, Domain Coverage, Quality, & Toxicity","created":"2023-05-22","cited_by_count":182,"area":null},{"id":"2405.14782","title":"Lessons from the Trenches on Reproducible Evaluation of Language Models","created":"2024-05-23","cited_by_count":179,"area":null},{"id":"2109.04404","title":"All Bark and No Bite: Rogue Dimensions in Transformer Language Models Obscure Representational Quality","created":"2021-09-09","cited_by_count":156,"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":"2210.17451","title":"AdaMix: Mixture-of-Adaptations for Parameter-efficient Model Tuning","created":"2022-10-31","cited_by_count":137,"area":null},{"id":"2209.02535","title":"Analyzing Transformers in Embedding Space","created":"2022-09-06","cited_by_count":132,"area":null},{"id":"2305.06329","title":"Similarity of Neural Network Models: A Survey of Functional and Representational Measures","created":"2023-05-10","cited_by_count":131,"area":null},{"id":"2309.07311","title":"Sudden Drops in the Loss: Syntax Acquisition, Phase Transitions, and Simplicity Bias in MLMs","created":"2023-09-13","cited_by_count":124,"area":null},{"id":"2212.14034","title":"Cramming: Training a Language Model on a Single GPU in One Day","created":"2022-12-28","cited_by_count":81,"area":null},{"id":"2203.04212","title":"Measuring the Mixing of Contextual Information in the Transformer","created":"2022-03-08","cited_by_count":65,"area":null},{"id":"2401.12143","title":"Anisotropy Is Inherent to Self-Attention in Transformers","created":"2024-01-22","cited_by_count":63,"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":"2203.13112","title":"minicons: Enabling Flexible Behavioral and Representational Analyses of Transformer Language Models","created":"2022-03-24","cited_by_count":58,"area":null},{"id":"2207.14251","title":"Measuring Causal Effects of Data Statistics on Language Model's `Factual' Predictions","created":"2022-07-28","cited_by_count":57,"area":null},{"id":"2205.12411","title":"Linear Connectivity Reveals Generalization Strategies","created":"2022-05-24","cited_by_count":56,"area":null},{"id":"2205.11380","title":"Outliers Dimensions that Disrupt Transformers Are Driven by Frequency","created":"2022-05-23","cited_by_count":55,"area":null},{"id":"2207.06991","title":"Language Modelling with Pixels","created":"2022-07-14","cited_by_count":55,"area":null},{"id":"2109.07152","title":"Incorporating Residual and Normalization Layers into Analysis of Masked Language Models","created":"2021-09-15","cited_by_count":51,"area":null},{"id":"2210.04337","title":"Quantifying Social Biases Using Templates is Unreliable","created":"2022-10-09","cited_by_count":39,"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":"2308.15419","title":"Characterizing Learning Curves During Language Model Pre-Training: Learning, Forgetting, and Stability","created":"2023-08-29","cited_by_count":35,"area":null},{"id":"2503.09543","title":"PolyPythias: Stability and Outliers across Fifty Language Model Pre-Training Runs","created":"2025-03-12","cited_by_count":35,"area":null},{"id":"2302.00456","title":"Analyzing Feed-Forward Blocks in Transformers through the Lens of Attention Maps","created":"2023-02-01","cited_by_count":33,"area":null},{"id":"2110.07596","title":"Retrieval-guided Counterfactual Generation for QA","created":"2021-10-14","cited_by_count":29,"area":null},{"id":"2410.11840","title":"A Hitchhiker's Guide to Scaling Law Estimation","created":"2024-10-15","cited_by_count":27,"area":null},{"id":"2307.04138","title":"On The Impact of Machine Learning Randomness on Group Fairness","created":"2023-07-09","cited_by_count":26,"area":null},{"id":"2305.12313","title":"When are ensembles really effective?","created":"2023-05-21","cited_by_count":21,"area":null},{"id":"2408.00531","title":"ReSi: A Comprehensive Benchmark for Representational Similarity Measures","created":"2024-08-01","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":"2501.14959","title":"Systemizing Multiplicity: The Curious Case of Arbitrariness in Machine Learning","created":"2025-01-24","cited_by_count":17,"area":null},{"id":"2308.09543","title":"Latent State Models of Training Dynamics","created":"2023-08-18","cited_by_count":16,"area":null},{"id":"2403.00986","title":"Merging Text Transformer Models from Different Initializations","created":"2024-03-01","cited_by_count":16,"area":null},{"id":"2210.01963","title":"COMPS: Conceptual Minimal Pair Sentences for testing Robust Property Knowledge and its Inheritance in Pre-trained Language Models","created":"2022-10-05","cited_by_count":15,"area":null},{"id":"2406.00053","title":"Dual Process Learning: Controlling Use of In-Context vs. In-Weights Strategies with Weight Forgetting","created":"2024-05-28","cited_by_count":15,"area":null},{"id":"2210.05043","title":"Multi-CLS BERT: An Efficient Alternative to Traditional Ensembling","created":"2022-10-10","cited_by_count":14,"area":null},{"id":"2310.11428","title":"Butterfly Effects of SGD Noise: Error Amplification in Behavior Cloning and Autoregression","created":"2023-10-17","cited_by_count":14,"area":null},{"id":"2312.01082","title":"A Survey on Stability of Learning with Limited Labelled Data and its Sensitivity to the Effects of Randomness","created":"2023-12-02","cited_by_count":14,"area":null},{"id":"2311.04900","title":"How Abstract Is Linguistic Generalization in Large Language Models? Experiments with Argument Structure","created":"2023-11-08","cited_by_count":13,"area":null},{"id":"2405.13828","title":"Babysit A Language Model From Scratch: Interactive Language Learning by Trials and Demonstrations","created":"2024-05-22","cited_by_count":12,"area":null},{"id":"2204.04344","title":"Towards Better Chinese-centric Neural Machine Translation for Low-resource Languages","created":"2022-04-09","cited_by_count":11,"area":null},{"id":"2204.06251","title":"Experimental Standards for Deep Learning in Natural Language Processing Research","created":"2022-04-13","cited_by_count":10,"area":null},{"id":"2412.04619","title":"Sometimes I am a Tree: Data Drives Unstable Hierarchical Generalization","created":"2024-12-05","cited_by_count":10,"area":null},{"id":"2204.03243","title":"Pretraining Text Encoders with Adversarial Mixture of Training Signal Generators","created":"2022-04-07","cited_by_count":9,"area":null},{"id":"2205.02289","title":"A Dataset for N-ary Relation Extraction of Drug Combinations","created":"2022-05-04","cited_by_count":9,"area":null},{"id":"2506.13234","title":"The Butterfly Effect: Neural Network Training Trajectories Are Highly Sensitive to Initial Conditions","created":"2025-06-16","cited_by_count":9,"area":null},{"id":"2205.09744","title":"Overcoming Language Disparity in Online Content Classification with Multimodal Learning","created":"2022-05-19","cited_by_count":8,"area":null},{"id":"2302.09865","title":"Can discrete information extraction prompts generalize across language models?","created":"2023-02-20","cited_by_count":8,"area":null},{"id":"2402.07899","title":"A systematic investigation of learnability from single child linguistic input","created":"2024-02-12","cited_by_count":8,"area":null},{"id":"2205.11833","title":"Diverse Lottery Tickets Boost Ensemble from a Single Pretrained Model","created":"2022-05-24","cited_by_count":7,"area":null},{"id":"2205.12410","title":"AdaMix: Mixture-of-Adaptations for Parameter-efficient Model Tuning","created":"2022-05-24","cited_by_count":7,"area":null},{"id":"2302.07778","title":"Measuring the Instability of Fine-Tuning","created":"2023-02-15","cited_by_count":7,"area":null},{"id":"2305.13067","title":"Distilling Robustness into Natural Language Inference Models with Domain-Targeted Augmentation","created":"2023-05-22","cited_by_count":7,"area":"distill"},{"id":"2502.17356","title":"Random Scaling of Emergent Capabilities","created":"2025-02-24","cited_by_count":7,"area":null},{"id":"2202.07652","title":"Predicting on the Edge: Identifying Where a Larger Model Does Better","created":"2022-02-15","cited_by_count":6,"area":null},{"id":"2212.09686","title":"A Natural Bias for Language Generation Models","created":"2022-12-19","cited_by_count":6,"area":null},{"id":"2305.14555","title":"All Roads Lead to Rome? Exploring the Invariance of Transformers' Representations","created":"2023-05-23","cited_by_count":6,"area":null},{"id":"2510.13796","title":"The Mechanistic Emergence of Symbol Grounding in Language Models","created":"2025-10-15","cited_by_count":6,"area":null},{"id":"2111.09612","title":"How Emotionally Stable is ALBERT? Testing Robustness with Stochastic Weight Averaging on a Sentiment Analysis Task","created":"2021-11-18","cited_by_count":5,"area":null},{"id":"2302.04054","title":"Towards Inferential Reproducibility of Machine Learning Research","created":"2023-02-08","cited_by_count":4,"area":null},{"id":"2303.16992","title":"ContraSim -- Analyzing Neural Representations Based on Contrastive Learning","created":"2023-03-29","cited_by_count":4,"area":null},{"id":"2407.12161","title":"Interpretability in Action: Exploratory Analysis of VPT, a Minecraft Agent","created":"2024-07-16","cited_by_count":4,"area":"agents"},{"id":"2407.13070","title":"The Cost of Arbitrariness for Individuals: Examining the Legal and Technical Challenges of Model Multiplicity","created":"2024-05-28","cited_by_count":4,"area":null},{"id":"2410.01483","title":"Foldable SuperNets: Scalable Merging of Transformers with Different Initializations and Tasks","created":"2024-10-02","cited_by_count":4,"area":null},{"id":"2505.09810","title":"Lossless Compression for LLM Tensor Incremental Snapshots","created":"2025-05-14","cited_by_count":4,"area":null},{"id":"2207.00560","title":"Is neural language acquisition similar to natural? A chronological probing study","created":"2022-07-01","cited_by_count":3,"area":null},{"id":"2211.13527","title":"Beyond Mahalanobis-Based Scores for Textual OOD Detection","created":"2022-11-24","cited_by_count":3,"area":null},{"id":"2305.12709","title":"Cross-lingual Transfer Can Worsen Bias in Sentiment Analysis","created":"2023-05-22","cited_by_count":3,"area":null},{"id":"2306.05307","title":"Are fairness metric scores enough to assess discrimination biases in machine learning?","created":"2023-06-08","cited_by_count":3,"area":null},{"id":"2406.02329","title":"On Affine Homotopy between Language Encoders","created":"2024-06-04","cited_by_count":3,"area":null},{"id":"2606.06533","title":"Position: Don't Just \"Fix it in Post\": A Science of AI Must Study Training Dynamics","created":"2026-06-03","cited_by_count":3,"area":null},{"id":"2406.12471","title":"Fighting Randomness with Randomness: Mitigating Optimisation Instability of Fine-Tuning using Delayed Ensemble and Noisy Interpolation","created":"2024-06-18","cited_by_count":2,"area":null},{"id":"2411.08687","title":"Diagnosing Neural Convergence with Topological Alignment Spectra","created":"2024-11-13","cited_by_count":2,"area":null},{"id":"2509.19332","title":"Quantifying Compositionality of Classic and State-of-the-Art Embeddings","created":"2025-09-14","cited_by_count":2,"area":null},{"id":"2602.16740","title":"Quantifying LLM Attention-Head Stability: Implications for Circuit Universality","created":"2026-02-17","cited_by_count":2,"area":"mechinterp"},{"id":"2603.26539","title":"How Open Must Language Models be to Enable Reliable Scientific Inference?","created":"2026-03-27","cited_by_count":2,"area":null},{"id":"2108.13032","title":"Shatter: An Efficient Transformer Encoder with Single-Headed Self-Attention and Relative Sequence Partitioning","created":"2021-08-30","cited_by_count":1,"area":null},{"id":"2211.14639","title":"Gender Biases Unexpectedly Fluctuate in the Pre-training Stage of Masked Language Models","created":"2022-11-26","cited_by_count":1,"area":null},{"id":"2302.03750","title":"Linking convolutional kernel size to generalization bias in face analysis CNNs","created":"2023-02-07","cited_by_count":1,"area":null},{"id":"2312.03762","title":"Colour versus Shape Goal Misgeneralization in Reinforcement Learning: A Case Study","created":"2023-12-05","cited_by_count":1,"area":null},{"id":"2402.12817","title":"On Sensitivity of Learning with Limited Labelled Data to the Effects of Randomness: Impact of Interactions and Systematic Choices","created":"2024-02-20","cited_by_count":1,"area":null},{"id":"2402.15925","title":"MultiContrievers: Analysis of Dense Retrieval Representations","created":"2024-02-24","cited_by_count":1,"area":null},{"id":"2602.07135","title":"Landscaper: Understanding Loss Landscapes Through Multi-Dimensional Topological Analysis","created":"2026-02-06","cited_by_count":1,"area":null},{"id":"2608.29266","title":"Measurement Validity in LLM Cultural Alignment","created":"2026-08-29","cited_by_count":1,"area":null},{"id":"2311.04166","title":"Perturbed examples reveal invariances shared by language models","created":"2023-11-07","cited_by_count":0,"area":null},{"id":"2410.12511","title":"Advancing Fairness in Natural Language Processing: From Traditional Methods to Explainability","created":"2024-10-16","cited_by_count":0,"area":null},{"id":"2608.02705","title":"Can Training Logs Make Model Comparisons More Precise?","created":"2026-08-03","cited_by_count":0,"area":null},{"id":"2610.14184","title":"ULPack: Reference-Conditioned Lossless Compression of Checkpoint Trajectories","created":"2026-10-09","cited_by_count":0,"area":null},{"id":"2610.16825","title":"Mean Accuracy Is the Wrong Estimand for a Release Decision","created":"2026-10-12","cited_by_count":0,"area":null},{"id":"2610.17093","title":"SFT Leaves a Reproducible Signature in Activation Geometry: Calibrating Post-Training Interpretability Against Run-to-Run Noise","created":"2026-10-12","cited_by_count":0,"area":null}],"next_cursor":null}