| XLA from the Trenches: Correctness on TPU Workloads at Scale |
2026-10-31 |
English · Chinese |
| ProLand: A New Infeasible Manifold Optimization Method for LoRA Fine-Tuning |
2026-10-30 |
English · Chinese |
| GAVAGAI: Referential Alignment for Infant-Scale Vision-Language Models via Balanced Optimal Transport |
2026-10-30 |
English · Chinese |
| Select-TTT: Query-Conditioned Selective Fast Weight Updates for In-Place Test-Time Training |
2026-10-27 |
English · Chinese |
| SWIM: Learning What Matters from Behaviors for JEPA World Models |
2026-10-27 |
English · Chinese |
| Multi-State Linear Attention with Generalized Second-Order Routing |
2026-10-27 |
English · Chinese |
| GUI-JEPA: Candidate-Conditioned JEPA Latents as a Fast Action Policy for Mobile GUI Agents |
2026-10-27 |
English · Chinese |
| Folding Attention Biases into Keys Makes Compacted KV Caches Run on Standard Kernels |
2026-10-27 |
English · Chinese |
| ResCom: Activation Residual Compensation for Large Language Model Quantization |
2026-10-25 |
English · Chinese |
| PathJev: Decision-Native Vision–Language Modelling for Histopathology |
2026-10-23 |
English · Chinese |
| EditJev: System One Routing for Image Editing Models |
2026-10-23 |
English · Chinese |
| DT-JEPA: Learning to Predict Future Representations for Decision-Making |
2026-10-23 |
English · Chinese |
| Reverse-KL Policy Distillation: Statistical Guarantees and Limits |
2026-10-22 |
English · Chinese |
| Rethinking Sequence Compression with Test-time Regression |
2026-10-22 |
English · Chinese |
| OpenAttention: One Percent Is Enough with Tail Proxies |
2026-10-22 |
English · Chinese |
| LazyFlash: Forward-Dense, Backward-Sparse Attention for Accelerated Long-Context Fine-Tuning |
2026-10-22 |
English · Chinese |
| Inside-JEPA: From Prediction to Relational Grounding in Latent World Models |
2026-10-22 |
English · Chinese |
| Right On Q: Recovering Reasoning under Quantization via Self-Generated Trajectories |
2026-10-21 |
English · Chinese |
| MoEZip: Routing-Aware KV Cache Compression for Sparse Mixture-of-Experts LLMs |
2026-10-21 |
English · Chinese |
| Building Auditable AI Text Detectors with Sparse Autoencoders |
2026-10-21 |
English · Chinese |
| ALPS: Visual State Encoders for World Models |
2026-10-21 |
English · Chinese |
| Stratified Sparse Attention: Reducing the Sparse-to-Dense Gap at Extreme Sparsity |
2026-10-20 |
English · Chinese |
| RACE: RoPE-Aligned Attention Compression Based on Query-Key Activation Energies |
2026-10-20 |
English · Chinese |
| LocaLeJEPA: Maximum Entropy Representations under Geometric Priors |
2026-10-20 |
English · Chinese |
| Kronecker Delta Rule for Long Context Linear Attention |
2026-10-19 |
English · Chinese |
| FLINT: Fixed-budget Late Interaction |
2026-10-19 |
English · Chinese |
| DeepSEA: A Multilingual Evaluation Benchmark for Southeast Asia |
2026-10-19 |
English · Chinese |
| T-MUON: Tensorizing Muon via t-SVD for Large Language Model Pre-training |
2026-10-18 |
English · Chinese |
| Muse Glimmer 30B: An On-Device Agentic Model |
2026-10-17 |
English · Chinese |
| Sink-Sparse Hybrid Attention for Efficient Inference in T2I/T2V Diffusion Transformers |
2026-10-16 |
English · Chinese |
| Persistent Planar Memory for Video World Models |
2026-10-16 |
English · Chinese |
| Jev-as-a-Judge for Brazilian Portuguese Information Retrieval: Relevance Assessment and Reranking |
2026-10-16 |
English · Chinese |
| BMuon: Budgeting Muon's Spectral Geometry for Structured Compressibility |
2026-10-16 |
English · Chinese |
| When the KV Cache Is Slower Than the GPU: Scaling Remote KV Tiers for Context-Parallel Prefill |
2026-10-15 |
English · Chinese |
| When Does Muon Actually Win? A Paired Data-, Time-, and Tuning-Budget Benchmark |
2026-10-15 |
English · Chinese |
| SoftCache: Adaptive Step Size Caching for Diffusion Transformers |
2026-10-15 |
English · Chinese |
| RoleQ: Role-Aligned Scales for Converting Language Models to Ternary |
2026-10-15 |
English · Chinese |
| Retrofitting Recurrent Vision-Language Models from Pretrained Backbones |
2026-10-15 |
English · Chinese |
| Rethinking Routing Exploration in Modern Mixture-of-Experts |
2026-10-15 |
English · Chinese |
| Prism-1: Position- and Rival-Independent Scoring for System One Models |
2026-10-15 |
English · Chinese |
| Patch where the text is new: a compression view of where byte-level models should spend compute |
2026-10-15 |
English · Chinese |
| MJev-Spatial: System One Spatial Intelligence via Native Attention Matching |
2026-10-15 |
English · Chinese |
| LCT: Losslessly Compressed Tensors with GPU friendly execution for Training |
2026-10-15 |
English · Chinese |
| LAMP: Layerwise Adaptive Mixing of Predictors for Training-Free Few-Shot Vision-Language Adaptation |
2026-10-15 |
English · Chinese |
| Gen4P: Video Generators are 4D Perceivers |
2026-10-15 |
English · Chinese |
| Fractional Power-of-Two Quantization for Efficient and Effective Multiplier-Free LLM Inference |
2026-10-15 |
English · Chinese |
| EgoP²C: Internalizing Plan-Perceive-Compose for Spatial Reasoning in Egocentric Video |
2026-10-15 |
English · Chinese |
| DropKV: Decoupling Residual-Output Perturbation for Near-Optimal KV-Cache Eviction |
2026-10-15 |
English · Chinese |
| Designing Transformers for Continuous Diffusion Language Models |
2026-10-15 |
English · Chinese |
| Compression Does Not End at Selection: Post-Selection Positional Geometry for Sparse KV Caches |
2026-10-15 |
English · Chinese |
| BitSparse: Compact Activation Storage for ReLU Feed-Forward Networks |
2026-10-15 |
English · Chinese |
| Architecting for Hardware Diversity: Meta's End-to-End Co-Design for High-Performance AI Training on AMD GPUs |
2026-10-15 |
English · Chinese |
| AdaLoop: Thinking with the Vision Encoder |
2026-10-15 |
English · Chinese |
| Time-JEPA: Grounding Video World Models with Physical Time Series |
2026-10-14 |
English · Chinese |
| PN-Mix: Pair-Norm Guided Mixed-Precision Key Cache Quantization for Efficient Long-Output LLM Inference |
2026-10-14 |
English · Chinese |
| Open-world AI: a Tale of Three Requirements |
2026-10-14 |
English · Chinese |
| One-Step Prediction Fidelity Is a Poor Proxy for Planning Performance in JEPA World Models |
2026-10-14 |
English · Chinese |
| Memory-Efficient Subspace Optimization with Residual Correction: From Momentum SGD to Muon |
2026-10-14 |
English · Chinese |
| EgoJEPA: Beyond Retargeting for Dexterous Manipulation from Egocentric Videos |
2026-10-14 |
English · Chinese |
| ESC: Escaping 2D Diffusion Ambiguity with Geometric Constraints for Consistent Novel View Synthesis |
2026-10-14 |
English · Chinese |
| Dataset-Level Regularization for Joint-Embedding Self-Supervised Learning |
2026-10-14 |
English · Chinese |
| SmolWAM: A Latent World-Action Model for Robotic Manipulation on Edge Devices |
2026-10-09 |
English · Chinese |
| ReplaySSM: Cache SSM Inputs, Not State |
2026-10-09 |
English · Chinese |
| Orthros: Phase-Aware Heterogeneous Attention for Efficient Transformers |
2026-10-09 |
English · Chinese |
| On the Design Space of Continuous Diffusion Language Models |
2026-10-09 |
English · Chinese |
| Not All Dimensions Are Equal: Outlier-Aware Attention Linearization with Grouped Quadratic Kernels |
2026-10-09 |
English · Chinese |
| Multi-Stream Attention Residuals |
2026-10-09 |
English · Chinese |
| EV-JEPA: Elastic V-JEPA for Any Device and Cross-Device Inference |
2026-10-09 |
English · Chinese |
| DualQuant: Dual-scale Calibration-free Weight-and-Activation Quantization for LLMs |
2026-10-09 |
English · Chinese |
| VFold: Symmetry-Aware Cross-Layer Value Cache Compression |
2026-10-08 |
English · Chinese |
| SparseDecoding: Decoding-Aware Pruning for Accurate and Efficient LLM Inference |
2026-10-08 |
English · Chinese |
| Smoothing the Top-k Exposure Boundary for Sparse Mixture-of-Experts |
2026-10-08 |
English · Chinese |
| Rounding in Preconditioner Space: Redesigning 4-bit AdamW Optimizer-State Quantization |
2026-10-08 |
English · Chinese |
| MetaEncoder: Exploring the Limit of Bi-Encoders for Multimodal System One Decision Making with Natural Language Interface |
2026-10-08 |
English · Chinese |
| LeWAM: End-to-End World and Action Modeling with JEPAs |
2026-10-08 |
English · Chinese |
| Few-Step Generation via Data-Space Iteration |
2026-10-08 |
English · Chinese |
| CoTS: Competitive Token Selection for Latent Reasoning in Large Language Models |
2026-10-08 |
English · Chinese |
| Video Encoders Built on Image Representations |
2026-10-05 |
English · Chinese |
| VETO: Video Efficient Token Optimization for Vision Language Models |
2026-10-01 |
English · Chinese |
| TESS: Scalable, Transferable Meta-network for Data Selection Requires a Different Loss (and Why the Obvious Choice is Problematic) |
2026-10-01 |
English · Chinese |
| TACO: Ternary Absolute-max Column-wise One-sparse Optimizer for LLM Fine-Tuning |
2026-10-01 |
English · Chinese |
| Role-aware Heuristic Episodic Attention for Conversational LLMs |
2026-10-01 |
English · Chinese |
| Omni-Embed-Mini: Binding Modalities Without Forgetting via Dense Distillation |
2026-10-01 |
English · Chinese |
| MWOP: Modality-aware Width-wise Operation Pruning for Efficient MLLMs |
2026-10-01 |
English · Chinese |
| LLM2Jev: LLMs Are Already Jev-Style Decision Models -- When and How to Fine-Tune Them |
2026-10-01 |
English · Chinese |
| IrekoGPT: Turning Structured Pruning into Post-Hoc Slimmable LLMs |
2026-10-01 |
English · Chinese |
| AF-Muon: An AdamW-Free Muon Optimizer for Tied-Embedding Models |
2026-10-01 |
English · Chinese |
| HAWK: Rethinking Multimodal Drafting for Speculative Decoding |
2026-09-30 |
English · Chinese |
| V-JEPA Policy: Building Effective World-Action Models on Predictive Visual Latents |
2026-09-29 |
English · Chinese |
| Triadic Linear Attention: Three-Dimensional Recurrent States for Long-Context Sequence Modeling |
2026-09-29 |
English · Chinese |
| Time-Anchored Diffusion Language Models: Latent-Space Caching for Fast Generation |
2026-09-29 |
English · Chinese |
| STEPQuant: When and Where Errors Matter in Delta-Rule Recurrent State Quantization |
2026-09-29 |
English · Chinese |
| SPLASH: Switching Parallel Layouts of Attention with Seamless Handoff for LLM Serving |
2026-09-29 |
English · Chinese |
| Pretraining Latent Information Feedback Transformers with Teacher Supervision |
2026-09-29 |
English · Chinese |
| On Trajectory-Aware Training for Masked Diffusion Language Models |
2026-09-29 |
English · Chinese |
| LeapQuant: Efficient Linear Attention with Accurate Recurrent State Quantization |
2026-09-29 |
English · Chinese |
| Cross-Entropy Guided Routing in Mixture-of-Experts Large Language Models |
2026-09-29 |
English · Chinese |
| Chinese-Jev: Bringing System One Model to Chinese-Language Tasks |
2026-09-29 |
English · Chinese |
| Telescopic Language Models |
2026-09-28 |
English · Chinese |
| MS-GLA: Multi-Scale Gated Linear Attention for Addressing Representational Bottlenecks via Multi-Temporal Resolution |
2026-09-28 |
English · Chinese |
| Fractional State Space Transition for Long Sequence Modeling |
2026-09-28 |
English · Chinese |
| Words Speak Louder Than Order: A Behavioral Evaluation of Gemma 4 |
2026-09-25 |
English · Chinese |
| dQwen3.5: Adapting Hybrid Language Models into Bidirectional Diffusion Language Models |
2026-09-17 |
English · Chinese |
| On-Demand Attention: Efficient Long-Context Decoding with Learned Recall |
2026-09-17 |
English · Chinese |
| ASPIRE: Asynchronous Batch Self-Speculative Decoding |
2026-09-16 |
English · Chinese |
| Persistent Recurrent Memory Between Transformer Layers Improves Language Model Generalization |
2026-09-15 |
English · Chinese |
| OPEN-1B: A Fully Auditable Training Run |
2026-09-15 |
English · Chinese |
| LoopSpec: Pipelined Self-Speculative Decoding for Looped Transformers |
2026-09-15 |
English · Chinese |
| How Calibration Content Shapes Attention-Based Reranking |
2026-09-15 |
English · Chinese |
| ECHO: Early-layer Collaborative Hierarchical Orchestration with Bonus Logits in Speculative Decoding |
2026-09-15 |
English · Chinese |
| Register Tokens for Bounded-State Reasoning in Diffusion Language Models |
2026-09-14 |
English · Chinese |
| AgentKV: Phase-Aware KV Eviction for Agentic LLMs |
2026-09-14 |
English · Chinese |
| SAS: Simple Attention Sparsification via End-to-End Optimization of Context Ranking |
2026-09-11 |
English · Chinese |
| SenseNova-U1.5: Towards Native Unified Visual Intelligence |
2026-09-10 |
English · Chinese |
| SWRouter: Similarity-Contractive Window Routing for Multi-Turn Large Language Model Conversations |
2026-09-10 |
English · Chinese |
| OmniKVQuant: KV Cache Quantization for Omni-LLMs |
2026-09-10 |
English · Chinese |
| Musec: MomentUm SpEctral Clipping for Stable Muon-type Training |
2026-09-10 |
English · Chinese |
| Data Scarcity and Model Sparsity: Mixtures-of-Experts Overfit More to Repeated Data |
2026-09-10 |
English · Chinese |
| Caption-once, Frames-on-Demand: Visual-Need Routing for Budget-Aware Agentic Long Video Understanding |
2026-09-10 |
English · Chinese |
| KVMem: Virtualizing Million-Token Agent Workspaces on a Consumer GPU |
2026-09-04 |
English · Chinese |
| BeaconKV: Key-Value Cache Compression Guided by Beacon Queries for Efficient Large Reasoning Model Inference |
2026-09-04 |
English · Chinese |
| Random Attention: Rethinking KV Cache Eviction for Efficient Reasoning |
2026-09-03 |
English · Chinese |
| Lngram v2: Latent N-Gram Memory with Interpretable Discrete Representations |
2026-09-03 |
English · Chinese |
| TwinKV: A Composable Repair Pass for KV Cache Eviction via Pairwise Key Redundancy |
2026-08-27 |
English · Chinese |
| PACE: A Unified Condense-and-Extract Paradigm for Fast VLM Inference |
2026-08-27 |
English · Chinese |
| CritICL: Inference-Time Weak-to-Strong Generalization from Small Language Model Failure Modes |
2026-08-27 |
English · Chinese |
| VBVR-Pro: A Scalable and Verifiable Suite for Native Visual Reasoning |
2026-08-26 |
English · Chinese |
| MLLMCLIP: Feature-Level Distillation of MLLM for Robust Vision-Language Representations |
2026-08-26 |
English · Chinese |
| WeMM-Embedding: WeChat Multi-Modal Embedding Technical Report |
2026-08-25 |
English · Chinese |
| RARE: Decoupling Representation Steering from Expert Routing in Mixture-of-Experts Language Models |
2026-08-21 |
English · Chinese |
| Fast A/B/n Testing: Exact Multi-Policy Comparison via Tree-Coupled Feedback Sharing |
2026-08-13 |
English · Chinese |
| Generation as Auxiliary Supervision: Enhancing Visual Understanding at Zero Inference Overhead via Decoupled Embedding Prediction |
2026-08-12 |
English · Chinese |
| Cracks in the Foundation: Seemingly Minor Architectural Choices Impact Long Context Extension |
2026-08-10 |
English · Chinese |
| DistillCache: KL-Guided Adaptive KV-Cache Eviction for Memory-Efficient LLM Inference |
2026-08-09 |
English · Chinese |
| Autonomy-of-Heads: Data-Free Sparse Attention from Frozen Query-Key Geometry |
2026-08-07 |
English · Chinese |
| MACRO: Markov Chain Routing of Transformer Layers |
2026-08-06 |
English · Chinese |
| Hierarchical Latent Prediction for Language Models |
2026-08-06 |
English · Chinese |
| Detecting Safety Training Modification in Language Models via Activation Analysis |
2026-08-06 |
English · Chinese |
| BaKron: Efficient Quantization with Kronecker-Factored Hessians |
2026-08-06 |
English · Chinese |
| Towards Physics of Multimodal Pretraining: Knowledge Flow, Modality Synergy, Early Unification, and Recipes |
2026-08-05 |
English · Chinese |
| QEvict: Recoverable Quantized KV Eviction for Attention-Drift-Robust Long-Context Decoding |
2026-08-05 |
English · Chinese |
| DBLast: Dependent Block Drafting for Stochastic Speculative Decoding |
2026-08-05 |
English · Chinese |
| TaskPress: Query-Agnostic KV Cache Compression via Task-Guided Pruning |
2026-08-04 |
English · Chinese |
| Role-Decoupled Attention Residuals |
2026-08-03 |
English · Chinese |
| Learning What to Remember: Test-Time Training via Context Distillation |
2026-08-03 |
English · Chinese |
| DART: Decoded Attention over Recurrent States for Efficient Long-Context Sequence Modeling |
2026-08-03 |
English · Chinese |
| TransMem: Transforming Hidden States into Memory for Large Language Models |
2026-07-31 |
English · Chinese |
| WIDE: Boosting Adaptive LLM Inference via Token-level Dynamic Width Pruning |
2026-07-30 |
English · Chinese |
| ReToken: One Token to Improve Vision–Language Models for Visual Retrieval |
2026-07-30 |
English · Chinese |
| Penelope: Localized Latent Recurrence for Efficient Structured Reasoning |
2026-07-28 |
English · Chinese |
| DataOrchestra: Learning to Orchestrate Per-Example Curation of Pretraining Data |
2026-07-27 |
English · Chinese |
| Windowed-MTP: Removing the Full-Context Draft-KV Tax at Million-Token Context |
2026-07-23 |
English · Chinese |
| Parameter-free Adaptive Sparse Attention via Compression-Based Content Selection |
2026-07-23 |
English · Chinese |
| Naju: A Native Discrete State-Space Model with Independent Retention and Writing for Long-Sequence Memory |
2026-07-23 |
English · Chinese |
| Hilbert Operator for Progressive Encoding (HOPE): A Mathematical Framework for Deconstructing Learned Representations in Deep Networks |
2026-07-23 |
English · Chinese |
| Anti-Periodic Positional Encoding: Möbius Boundary Conditions Make In-Context Retrieval Reliable |
2026-07-23 |
English · Chinese |
| Adaptive Depth Sparse Framework: Similarity-Driven Resource Allocation for Pre-Trained LLMs |
2026-07-23 |
English · Chinese |
| GaugeQuant: Online Learning of Quantization-Optimal Bases from LLM Symmetries |
2026-07-22 |
English · Chinese |
| PPL-Factory: Task-Aware and Budget-Aware Data Selection from Language Modeling to Reasoning |
2026-07-20 |
English · Chinese |
| C²KV: Compressed and Composable KV Cache Reuse for Efficient LLM Inference |
2026-07-20 |
English · Chinese |
| Convolution for Large Language Models |
2026-07-20 |
English · Chinese |
| OpenLanguageModel: Readable and Composable Small-Language-Model Pretraining for Education and Research |
2026-07-18 |
English · Chinese |
| Looped Latent Attention: Cross-Loop KV Compression for Looped Transformers |
2026-07-16 |
English · Chinese |
| Cross-Lingual Exploration for Parametric Knowledge |
2026-06-23 |
English · Chinese |
| MiniMax Sparse Attention |
2026-06-11 |
English · Chinese |
| Language Models Need Sleep: Learning to Self-Modify and Consolidate Memories |
2026-06-02 |
English · Chinese |
| XLGoBench: Detecting cross-lingual skill gaps with algorithmic tasks |
2026-05-29 |
English · Chinese |
| Memory Grafting: Scaling Language Model Pre-training via Offline Conditional Memory |
2026-05-20 |
English · Chinese |
| Customizing an LLM for Enterprise Software Engineering |
2026-05-15 |
English · Chinese |
| Optimized Deferral for Imbalanced Settings |
2026-04-30 |
English · Chinese |
| ProEval: Proactive Failure Discovery and Efficient Performance Estimation for Generative AI Evaluation |
2026-04-25 |
English · Chinese |
| FUSE: Ensembling Verifiers with Zero Labeled Data |
2026-04-20 |
English · Chinese |
| Switch Attention: Towards Dynamic and Fine-grained Hybrid Transformers |
2026-03-27 |
English · Chinese |
| Efficient, Property-Aligned Fan-Out Retrieval via RL-Compiled Diffusion |
2026-03-06 |
English · Chinese |
| Theoretical Perspectives on Data Quality and Synergistic Effects in Pre- and Post-Training Reasoning Models |
2026-03-01 |
English · Chinese |
| A Theoretical Framework for Modular Learning of Robust Generative Models |
2026-02-19 |
English · Chinese |
| Mobility-Embedded POIs: Learning What A Place Is and How It Is Used from Human Movement |
2026-01-29 |
English · Chinese |
| ECO: Quantized Training without Full-Precision Master Weights |
2026-01-29 |
English · Chinese |
| Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models |
2026-01-12 |
English · Chinese |
| mHC: Manifold-Constrained Hyper-Connections |
2025-12-31 |
English · Chinese |
| Detecting Post-generation Edits to Watermarked LLM Outputs via Combinatorial Watermarking |
2025-10-02 |
English · Chinese |
| GLiClass: Generalist Lightweight Model for Sequence Classification Tasks |
2025-08-11 |
English · Chinese |
| Model Directions, Not Words: Mechanistic Topic Models Using Sparse Autoencoders |
2025-07-31 |
English · Chinese |
| Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free |
2025-05-10 |
English · Chinese |
| SmolVLM: Redefining small and efficient multimodal models |
2025-04-07 |
English · Chinese |
| Muon is Scalable for LLM Training |
2025-02-24 |
English · Chinese |
| SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features |
2025-02-20 |
English · Chinese |
| MoBA: Mixture of Block Attention for Long-Context LLMs |
2025-02-18 |
English · Chinese |
| Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention |
2025-02-16 |
English · Chinese |
| Byte Latent Transformer: Patches Scale Better Than Tokens |
2024-12-13 |
English · Chinese |
| Hymba: A Hybrid-head Architecture for Small Language Models |
2024-11-20 |
English · Chinese |
| Data Mixing Laws: Optimizing Data Mixtures by Predicting Language Modeling Performance |
2024-03-25 |
English · Chinese |
| Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads |
2024-01-19 |
English · Chinese |
| Mamba: Linear-Time Sequence Modeling with Selective State Spaces |
2023-12-01 |
English · Chinese |
| AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration |
2023-06-01 |
English · Chinese |
| GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints |
2023-05-22 |
English · Chinese |
| DoReMi: Optimizing Data Mixtures Speeds Up Language Model Pretraining |
2023-05-17 |
English · Chinese |
| Visual Instruction Tuning |
2023-04-17 |
English · Chinese |
| BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models |
2023-01-30 |
English · Chinese |
| Fast Inference from Transformers via Speculative Decoding |
2022-11-30 |
English · Chinese |
| Train Short, Test Long: Attention with Linear Biases Enables Input Length Extrapolation |
2021-08-27 |
English · Chinese |
| RoFormer: Enhanced Transformer with Rotary Position Embedding |
2021-04-20 |
English · Chinese |
| Learning Transferable Visual Models From Natural Language Supervision |
2021-02-26 |
English · Chinese |
| Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity |
2021-01-11 |
English · Chinese |