{"total":92,"rows":[{"id":"2108.12409","title":"Train Short, Test Long: Attention with Linear Biases Enables Input Length Extrapolation","created":"2021-08-27","cited_by_count":1103,"area":null},{"id":"2311.16867","title":"The Falcon Series of Open Language Models","created":"2023-11-28","cited_by_count":535,"area":null},{"id":"2106.07476","title":"Training Graph Neural Networks with 1000 Layers","created":"2021-06-14","cited_by_count":169,"area":null},{"id":"2312.03002","title":"The mechanistic basis of data dependence and abrupt learning in an in-context classification task","created":"2023-12-03","cited_by_count":118,"area":null},{"id":"2403.08540","title":"Language models scale reliably with over-training and on downstream tasks","created":"2024-03-13","cited_by_count":113,"area":null},{"id":"2103.13076","title":"Finetuning Pretrained Transformers into RNNs","created":"2021-03-24","cited_by_count":96,"area":null},{"id":"2102.02557","title":"Adaptive Semiparametric Language Models","created":"2021-02-04","cited_by_count":83,"area":null},{"id":"2311.09578","title":"Tied-Lora: Enhancing parameter efficiency of LoRA with weight tying","created":"2023-11-16","cited_by_count":65,"area":null},{"id":"2312.16903","title":"Spike No More: Stabilizing the Pre-training of Large Language Models","created":"2023-12-28","cited_by_count":52,"area":null},{"id":"2405.05417","title":"Fishing for Magikarp: Automatically Detecting Under-trained Tokens in Large Language Models","created":"2024-05-08","cited_by_count":46,"area":null},{"id":"2109.07513","title":"Tied & Reduced RNN-T Decoder","created":"2021-09-15","cited_by_count":36,"area":null},{"id":"2210.00584","title":"FLCert: Provably Secure Federated Learning against Poisoning Attacks","created":"2022-10-02","cited_by_count":30,"area":null},{"id":"2101.09048","title":"Selfish Sparse RNN Training","created":"2021-01-22","cited_by_count":26,"area":null},{"id":"2205.01086","title":"Wav2Seq: Pre-training Speech-to-Text Encoder-Decoder Models Using Pseudo Languages","created":"2022-05-02","cited_by_count":26,"area":null},{"id":"2206.06178","title":"Efficient recurrent architectures through activity sparsity and sparse back-propagation through time","created":"2022-06-13","cited_by_count":26,"area":null},{"id":"2110.04367","title":"Hybrid Random Features","created":"2021-10-08","cited_by_count":25,"area":null},{"id":"2201.09680","title":"Relational Memory Augmented Language Models","created":"2022-01-24","cited_by_count":24,"area":null},{"id":"2109.04740","title":"How Does Fine-tuning Affect the Geometry of Embedding Space: A Case Study on Isotropy","created":"2021-09-10","cited_by_count":23,"area":null},{"id":"2104.01371","title":"Convex Aggregation for Opinion Summarization","created":"2021-04-03","cited_by_count":21,"area":null},{"id":"2205.10835","title":"Multilingual Machine Translation with Hyper-Adapters","created":"2022-05-22","cited_by_count":15,"area":null},{"id":"2304.13539","title":"Tensor Decomposition for Model Reduction in Neural Networks: A Review","created":"2023-04-26","cited_by_count":14,"area":null},{"id":"2605.12466","title":"Solve the Loop: Attractor Models for Language and Reasoning","created":"2026-05-12","cited_by_count":13,"area":"lr"},{"id":"2109.10104","title":"InvBERT: Reconstructing Text from Contextualized Word Embeddings by inverting the BERT pipeline","created":"2021-09-21","cited_by_count":11,"area":null},{"id":"2112.14484","title":"Frequency-Aware Contrastive Learning for Neural Machine Translation","created":"2021-12-29","cited_by_count":11,"area":null},{"id":"2209.01610","title":"Generalization in Neural Networks: A Broad Survey","created":"2022-09-04","cited_by_count":11,"area":null},{"id":"2408.15332","title":"What makes math problems hard for reinforcement learning: a case study","created":"2024-08-27","cited_by_count":9,"area":"deeprl"},{"id":"2103.01819","title":"The Rediscovery Hypothesis: Language Models Need to Meet Linguistics","created":"2021-03-02","cited_by_count":7,"area":null},{"id":"2107.05124","title":"Transformers with multi-modal features and post-fusion context for e-commerce session-based recommendation","created":"2021-07-11","cited_by_count":7,"area":null},{"id":"2302.00845","title":"Coordinating Distributed Example Orders for Provably Accelerated Training","created":"2023-02-02","cited_by_count":7,"area":null},{"id":"2410.07836","title":"Masked Generative Priors Improve World Models Sequence Modelling Capabilities","created":"2024-10-10","cited_by_count":7,"area":"wm"},{"id":"2508.15390","title":"Exploiting Vocabulary Frequency Imbalance in Language Model Pre-training","created":"2025-08-21","cited_by_count":7,"area":null},{"id":"2409.07787","title":"Stable Language Model Pre-training by Reducing Embedding Variability","created":"2024-09-12","cited_by_count":6,"area":null},{"id":"2411.05036","title":"From Word Vectors to Multimodal Embeddings: Techniques, Applications, and Future Directions For Large Language Models","created":"2024-11-06","cited_by_count":6,"area":null},{"id":"2505.20211","title":"PiCa: Parameter-Efficient Fine-Tuning with Column Space Projection","created":"2025-05-26","cited_by_count":6,"area":null},{"id":"2102.07071","title":"Doping: A technique for efficient compression of LSTM models using sparse structured additive matrices","created":"2021-02-14","cited_by_count":5,"area":null},{"id":"2205.06522","title":"Joint Generation of Captions and Subtitles with Dual Decoding","created":"2022-05-13","cited_by_count":5,"area":null},{"id":"2405.16727","title":"Disentangling and Integrating Relational and Sensory Information in Transformer Architectures","created":"2024-05-26","cited_by_count":5,"area":null},{"id":"2410.05021","title":"DEPT: Decoupled Embeddings for Pre-training Language Models","created":"2024-10-07","cited_by_count":5,"area":null},{"id":"2104.09048","title":"Neural Architecture Search for Image Super-Resolution Using Densely Constructed Search Space: DeCoNAS","created":"2021-04-19","cited_by_count":4,"area":null},{"id":"2201.05020","title":"Automatic Sparse Connectivity Learning for Neural Networks","created":"2022-01-13","cited_by_count":4,"area":null},{"id":"2203.12644","title":"Linearizing Transformer with Key-Value Memory","created":"2022-03-23","cited_by_count":4,"area":null},{"id":"2205.09273","title":"Twist Decoding: Diverse Generators Guide Each Other","created":"2022-05-19","cited_by_count":4,"area":null},{"id":"2207.09927","title":"ViGAT: Bottom-up event recognition and explanation in video using factorized graph attention network","created":"2022-07-20","cited_by_count":4,"area":null},{"id":"2212.04262","title":"ConsistTL: Modeling Consistency in Transfer Learning for Low-Resource Neural Machine Translation","created":"2022-12-08","cited_by_count":4,"area":null},{"id":"2408.07802","title":"Kraken: Inherently Parallel Transformers For Efficient Multi-Device Inference","created":"2024-08-14","cited_by_count":4,"area":null},{"id":"2104.12424","title":"Attention vs non-attention for a Shapley-based explanation method","created":"2021-04-26","cited_by_count":3,"area":null},{"id":"2105.10832","title":"Spectral Pruning for Recurrent Neural Networks","created":"2021-05-23","cited_by_count":3,"area":null},{"id":"2106.05634","title":"Exploring Unsupervised Pretraining Objectives for Machine Translation","created":"2021-06-10","cited_by_count":3,"area":null},{"id":"2311.00159","title":"Longer Fixations, More Computation: Gaze-Guided Recurrent Neural Networks","created":"2023-10-31","cited_by_count":3,"area":null},{"id":"2511.15779","title":"SURFing to the Fundamental Limit of Jet Tagging","created":"2025-11-19","cited_by_count":3,"area":null},{"id":"2603.26663","title":"Weight Tying Biases Token Embeddings Towards the Output Space","created":"2026-03-27","cited_by_count":3,"area":null},{"id":"2109.10197","title":"One Source, Two Targets: Challenges and Rewards of Dual Decoding","created":"2021-09-21","cited_by_count":2,"area":null},{"id":"2210.13163","title":"Bilingual Synchronization: Restoring Translational Relationships with Editing Operations","created":"2022-10-24","cited_by_count":2,"area":null},{"id":"2211.09728","title":"Generative Adversarial Training Can Improve Neural Language Models","created":"2022-11-02","cited_by_count":2,"area":null},{"id":"2303.14301","title":"Natural Language-Based Synthetic Data Generation for Cluster Analysis","created":"2023-03-24","cited_by_count":2,"area":null},{"id":"2305.14124","title":"When Does Monolingual Data Help Multilingual Translation: The Role of Domain and Model Scale","created":"2023-05-23","cited_by_count":2,"area":null},{"id":"2505.10202","title":"VQ-Logits: Compressing the Output Bottleneck of Large Language Models via Vector Quantized Logits","created":"2025-05-15","cited_by_count":2,"area":null},{"id":"2506.11035","title":"Tversky Neural Networks: Psychologically Plausible Deep Learning with Differentiable Tversky Similarity","created":"2025-05-21","cited_by_count":2,"area":null},{"id":"2601.07218","title":"SceneNAT: Masked Generative Modeling for Language-Guided Indoor Scene Synthesis","created":"2026-01-12","cited_by_count":2,"area":null},{"id":"2601.22040","title":"Leviathan: Decoupling Input and Output Representations in Language Models","created":"2026-01-29","cited_by_count":2,"area":null},{"id":"2602.04556","title":"Rethinking Weight Tying: Pseudo-Inverse Tying for LM Stable Training and Updates","created":"2026-02-04","cited_by_count":2,"area":"lr"},{"id":"2104.09691","title":"When FastText Pays Attention: Efficient Estimation of Word Representations using Constrained Positional Weighting","created":"2021-04-19","cited_by_count":1,"area":null},{"id":"2303.14011","title":"SPEC: Summary Preference Decomposition for Low-Resource Abstractive Summarization","created":"2023-03-24","cited_by_count":1,"area":null},{"id":"2501.06126","title":"Merging Feed-Forward Sublayers for Compressed Transformers","created":"2025-01-10","cited_by_count":1,"area":null},{"id":"2509.13357","title":"Semantic Fusion with Fuzzy-Membership Features for Controllable Language Modelling","created":"2025-09-14","cited_by_count":1,"area":null},{"id":"2608.30505","title":"Tensor Methods for Language Models: From Token Representation to Training, Adaptation, Inference, Compression, and Interpretability","created":"2026-08-31","cited_by_count":1,"area":null},{"id":"2609.01936","title":"Sparse Readout Prism: Explaining Logit-Lens Scores in Features Instead of Tokens","created":"2026-09-01","cited_by_count":1,"area":null},{"id":"2103.07875","title":"Learning a Word-Level Language Model with Sentence-Level Noise Contrastive Estimation for Contextual Sentence Probability Estimation","created":"2021-03-14","cited_by_count":0,"area":null},{"id":"2103.11603","title":"Alleviate Exposure Bias in Sequence Prediction \\\\ with Recurrent Neural Networks","created":"2021-03-22","cited_by_count":0,"area":null},{"id":"2104.01572","title":"TransfoRNN: Capturing the Sequential Information in Self-Attention Representations for Language Modeling","created":"2021-04-04","cited_by_count":0,"area":null},{"id":"2108.09858","title":"Data Augmentation Using Many-To-Many RNNs for Session-Aware Recommender Systems","created":"2021-08-22","cited_by_count":0,"area":null},{"id":"2112.11776","title":"The Importance of the Current Input in Sequence Modeling","created":"2021-12-22","cited_by_count":0,"area":null},{"id":"2206.04806","title":"Syntactic Inductive Biases for Deep Learning Methods","created":"2022-06-08","cited_by_count":0,"area":null},{"id":"2305.11462","title":"Extending Memory for Language Modelling","created":"2023-05-19","cited_by_count":0,"area":null},{"id":"2305.14230","title":"Exploring Representational Disparities Between Multilingual and Bilingual Translation Models","created":"2023-05-23","cited_by_count":0,"area":null},{"id":"2307.13709","title":"Neural Bradley-Terry Rating: Quantifying Properties from Comparisons","created":"2023-07-24","cited_by_count":0,"area":null},{"id":"2312.08401","title":"Balanced and Deterministic Weight-sharing Helps Network Performance","created":"2023-12-13","cited_by_count":0,"area":null},{"id":"2312.16104","title":"Dotless Representation of Arabic Text: Analysis and Modeling","created":"2023-12-26","cited_by_count":0,"area":null},{"id":"2406.10256","title":"Explicit Word Density Estimation for Language Modelling","created":"2024-06-10","cited_by_count":0,"area":null},{"id":"2510.01929","title":"Inverse Language Modeling towards Robust and Grounded LLMs","created":"2025-10-02","cited_by_count":0,"area":null},{"id":"2603.27938","title":"Top-down string-to-dependency Neural Machine Translation","created":"2026-03-30","cited_by_count":0,"area":null},{"id":"2605.06216","title":"TIDE: Every Layer Knows the Token Beneath the Context","created":"2026-05-07","cited_by_count":0,"area":null},{"id":"2605.09403","title":"Sparsity Moves Computation: How FFN Architecture Reshapes Attention in Small Transformers","created":"2026-05-10","cited_by_count":0,"area":"moe"},{"id":"2606.27941","title":"VASAE: Naming SAE Dictionary Directions with Vocabulary-Aligned Anchoring","created":"2026-06-26","cited_by_count":0,"area":"mechinterp"},{"id":"2607.24797","title":"Reading Without a Reader: Large Language Models Collapse Reading and Writing into a Single Entangled Code","created":"2026-07-02","cited_by_count":0,"area":null},{"id":"2609.16179","title":"Z-Loss Backward Geometry in Dense Output Heads and Sparse Routers","created":"2026-07-31","cited_by_count":0,"area":"moe"},{"id":"2609.17535","title":"DANTINOX: A Unified Framework for Multi-Paradigm Language Modeling","created":"2026-07-11","cited_by_count":0,"area":"diffusion"},{"id":"2609.34007","title":"EHRAdapt: Adapting Pretrained Language Models to Electronic Health Records with Semantic Priors for Rare Clinical Events","created":"2026-09-27","cited_by_count":0,"area":null},{"id":"2609.40335","title":"Is Weight Tying Still Beneficial for Decoder-Only LLMs in Private Settings Under DP-SGD?","created":"2026-09-30","cited_by_count":0,"area":null},{"id":"2610.01395","title":"AF-Muon: An AdamW-Free Muon Optimizer for Tied-Embedding Models","created":"2026-10-01","cited_by_count":0,"area":null},{"id":"2610.02705","title":"MuonIO: Principled Norm-Aware Descent for Embedding Tables and Language Model Heads","created":"2026-10-02","cited_by_count":0,"area":null},{"id":"2610.13605","title":"zip2zip++: Towards Inference-Time Token Compression with Minimal Quality Loss","created":"2026-10-09","cited_by_count":0,"area":"inference"}],"next_cursor":null}