{"total":202,"rows":[{"id":"2406.16860","title":"Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs","created":"2024-06-24","cited_by_count":838,"area":"mllm"},{"id":"2402.19427","title":"Griffin: Mixing Gated Linear Recurrences with Local Attention for Efficient Language Models","created":"2024-02-29","cited_by_count":242,"area":null},{"id":"2502.10517","title":"KernelBench: Can LLMs Write Efficient GPU Kernels?","created":"2025-02-14","cited_by_count":225,"area":null},{"id":"2308.03958","title":"Simple synthetic data reduces sycophancy in large language models","created":"2023-08-07","cited_by_count":187,"area":"synth"},{"id":"2312.15796","title":"GenCast: Diffusion-based ensemble forecasting for medium-range weather","created":"2023-12-25","cited_by_count":177,"area":null},{"id":"2304.02948","title":"FengWu: Pushing the Skillful Global Medium-range Weather Forecast beyond 10 Days Lead","created":"2023-04-06","cited_by_count":175,"area":null},{"id":"2305.09636","title":"SoundStorm: Efficient Parallel Audio Generation","created":"2023-05-16","cited_by_count":114,"area":null},{"id":"2312.15234","title":"Towards Efficient Generative Large Language Model Serving: A Survey from Algorithms to Systems","created":"2023-12-23","cited_by_count":102,"area":"inference"},{"id":"2305.15255","title":"Spoken Question Answering and Speech Continuation Using Spectrogram-Powered LLM","created":"2023-05-24","cited_by_count":94,"area":null},{"id":"2403.08635","title":"Human Alignment of Large Language Models through Online Preference Optimisation","created":"2024-03-13","cited_by_count":91,"area":"alignment"},{"id":"2305.08298","title":"Symbol tuning improves in-context learning in language models","created":"2023-05-15","cited_by_count":75,"area":null},{"id":"2405.14655","title":"Multi-turn Reinforcement Learning from Preference Human Feedback","created":"2024-05-23","cited_by_count":67,"area":"alignment"},{"id":"2405.19107","title":"Offline Regularised Reinforcement Learning for Large Language Models Alignment","created":"2024-05-29","cited_by_count":48,"area":"alignment"},{"id":"2312.02179","title":"Training Chain-of-Thought via Latent-Variable Inference","created":"2023-11-28","cited_by_count":47,"area":"lr"},{"id":"2502.14051","title":"RocketKV: Accelerating Long-Context LLM Inference via Two-Stage KV Cache Compression","created":"2025-02-19","cited_by_count":41,"area":"inference"},{"id":"2507.07101","title":"Small Batch Size Training for Language Models: When Vanilla SGD Works, and Why Gradient Accumulation Is Wasteful","created":"2025-07-09","cited_by_count":41,"area":null},{"id":"2505.09343","title":"Insights into DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI Architectures","created":"2025-05-14","cited_by_count":39,"area":"moe"},{"id":"2404.07236","title":"Lightweight Deep Learning for Resource-Constrained Environments: A Survey","created":"2024-04-08","cited_by_count":38,"area":null},{"id":"2309.10952","title":"LMDX: Language Model-based Document Information Extraction and Localization","created":"2023-09-19","cited_by_count":37,"area":null},{"id":"2311.05908","title":"FlashFFTConv: Efficient Convolutions for Long Sequences with Tensor Cores","created":"2023-11-10","cited_by_count":33,"area":null},{"id":"2405.10313","title":"How Far Are We From AGI: Are LLMs All We Need?","created":"2024-05-16","cited_by_count":31,"area":null},{"id":"2406.12930","title":"Tender: Accelerating Large Language Models via Tensor Decomposition and Runtime Requantization","created":"2024-06-16","cited_by_count":28,"area":null},{"id":"2410.21680","title":"Revisiting Reliability in Large-Scale Machine Learning Research Clusters","created":"2024-10-29","cited_by_count":28,"area":null},{"id":"2405.02429","title":"CALRec: Contrastive Alignment of Generative LLMs for Sequential Recommendation","created":"2024-05-03","cited_by_count":24,"area":null},{"id":"2410.07265","title":"A Survey: Collaborative Hardware and Software Design in the Era of Large Language Models","created":"2024-10-08","cited_by_count":23,"area":null},{"id":"2511.15915","title":"AccelOpt: A Self-Improving LLM Agentic System for AI Accelerator Kernel Optimization","created":"2025-11-19","cited_by_count":23,"area":"agents"},{"id":"2404.09173","title":"TransformerFAM: Feedback attention is working memory","created":"2024-04-14","cited_by_count":22,"area":null},{"id":"2502.07578","title":"PIM Is All You Need: A CXL-Enabled GPU-Free System for Large Language Model Inference","created":"2025-02-11","cited_by_count":22,"area":"inference"},{"id":"2503.00295","title":"Robust Multi-Objective Preference Alignment with Online DPO","created":"2025-03-01","cited_by_count":22,"area":"alignment"},{"id":"2306.15552","title":"A Survey on Deep Learning Hardware Accelerators for Heterogeneous HPC Platforms","created":"2023-06-27","cited_by_count":21,"area":null},{"id":"2312.14972","title":"Scaling Down to Scale Up: A Cost-Benefit Analysis of Replacing OpenAI's LLM with Open Source SLMs in Production","created":"2023-12-20","cited_by_count":21,"area":null},{"id":"2404.01367","title":"Bigger is not Always Better: Scaling Properties of Latent Diffusion Models","created":"2024-04-01","cited_by_count":21,"area":"diffusion"},{"id":"2405.03103","title":"Learning from Students: Applying t-Distributions to Explore Accurate and Efficient Formats for LLMs","created":"2024-05-06","cited_by_count":18,"area":null},{"id":"2405.07518","title":"SambaNova SN40L: Scaling the AI Memory Wall with Dataflow and Composition of Experts","created":"2024-05-13","cited_by_count":18,"area":null},{"id":"2405.12807","title":"FAdam: Adam is a natural gradient optimizer using diagonal empirical Fisher information","created":"2024-05-21","cited_by_count":18,"area":null},{"id":"2501.11211","title":"Ditto: Accelerating Diffusion Model via Temporal Value Similarity","created":"2025-01-20","cited_by_count":18,"area":"diffusion"},{"id":"2306.08068","title":"DORSal: Diffusion for Object-centric Representations of Scenes et al","created":"2023-06-13","cited_by_count":16,"area":"diffusion"},{"id":"2404.04815","title":"Allo: A Programming Model for Composable Accelerator Design","created":"2024-04-07","cited_by_count":16,"area":null},{"id":"2410.12032","title":"MLPerf Power: Benchmarking the Energy Efficiency of Machine Learning Systems from Microwatts to Megawatts for Sustainable AI","created":"2024-10-15","cited_by_count":16,"area":null},{"id":"2502.18755","title":"M-ANT: Efficient Low-bit Group Quantization for LLMs via Mathematically Adaptive Numerical Type","created":"2025-02-26","cited_by_count":16,"area":null},{"id":"2503.12347","title":"Synthesizing Privacy-Preserving Text Data via Finetuning without Finetuning Billion-Scale LLMs","created":"2025-03-16","cited_by_count":16,"area":null},{"id":"2407.13853","title":"Forecasting GPU Performance for Deep Learning Training and Inference","created":"2024-07-18","cited_by_count":15,"area":null},{"id":"2411.17651","title":"APEX: An Extensible and Dynamism-Aware Simulator for Automated Parallel Execution in LLM Serving","created":"2024-11-26","cited_by_count":15,"area":"inference"},{"id":"2412.11702","title":"Flex-PE: Flexible and SIMD Multi-Precision Processing Element for AI Workloads","created":"2024-12-16","cited_by_count":15,"area":null},{"id":"2305.12102","title":"Unified Embedding: Battle-Tested Feature Representations for Web-Scale ML Systems","created":"2023-05-20","cited_by_count":14,"area":null},{"id":"2310.02784","title":"MAD Max Beyond Single-Node: Enabling Large Machine Learning Model Acceleration on Distributed Systems","created":"2023-10-04","cited_by_count":14,"area":null},{"id":"2402.10930","title":"ConSmax: Hardware-Friendly Alternative Softmax with Learnable Parameters","created":"2024-01-31","cited_by_count":14,"area":null},{"id":"2405.14009","title":"ReCycle: Resilient Training of Large DNNs using Pipeline Adaptation","created":"2024-05-22","cited_by_count":14,"area":null},{"id":"2405.17454","title":"Generative AI for the Optimization of Next-Generation Wireless Networks: Basics, State-of-the-Art, and Open Challenges","created":"2024-05-22","cited_by_count":14,"area":null},{"id":"2503.18599","title":"Oaken: Fast and Efficient LLM Serving with Online-Offline Hybrid KV Cache Quantization","created":"2025-03-24","cited_by_count":14,"area":"inference"},{"id":"2307.12169","title":"Rail-only: A Low-Cost High-Performance Network for Training LLMs with Trillion Parameters","created":"2023-07-22","cited_by_count":13,"area":"moe"},{"id":"2507.00390","title":"MoNE: Replacing Redundant Experts with Lightweight Novices for Structured Pruning of MoE","created":"2025-07-01","cited_by_count":13,"area":"moe"},{"id":"2305.17547","title":"Translatotron 3: Speech to Speech Translation with Monolingual Data","created":"2023-05-27","cited_by_count":12,"area":null},{"id":"2401.08660","title":"Gemini Pro Defeated by GPT-4V: Evidence from Education","created":"2023-12-27","cited_by_count":12,"area":null},{"id":"2408.04413","title":"Deeploy: Enabling Energy-Efficient Deployment of Small Language Models On Heterogeneous Microcontrollers","created":"2024-08-08","cited_by_count":11,"area":null},{"id":"2510.14557","title":"MX+: Pushing the Limits of Microscaling Formats for Efficient Large Language Model Serving","created":"2025-10-16","cited_by_count":11,"area":"inference"},{"id":"2401.11202","title":"PartIR: Composing SPMD Partitioning Strategies for Machine Learning","created":"2024-01-20","cited_by_count":10,"area":null},{"id":"2405.18616","title":"Wavelet-Based Image Tokenizer for Vision Transformers","created":"2024-05-28","cited_by_count":10,"area":null},{"id":"2502.01671","title":"Life-Cycle Emissions of AI Hardware: A Cradle-To-Grave Approach and Generational Trends","created":"2025-02-01","cited_by_count":10,"area":null},{"id":"2309.16318","title":"DeepPCR: Parallelizing Sequential Operations in Neural Networks","created":"2023-09-28","cited_by_count":9,"area":"diffusion"},{"id":"2405.00645","title":"HGQ: High Granularity Quantization for Real-time Neural Networks on FPGAs","created":"2024-05-01","cited_by_count":9,"area":null},{"id":"2407.09111","title":"Inference Optimization of Foundation Models on AI Accelerators","created":"2024-07-12","cited_by_count":9,"area":null},{"id":"2408.03560","title":"In2Core: Leveraging Influence Functions for Coreset Selection in Instruction Finetuning of Large Language Models","created":"2024-08-07","cited_by_count":9,"area":null},{"id":"2501.10658","title":"LUT-DLA: Lookup Table as Efficient Extreme Low-Bit Deep Learning Accelerator","created":"2025-01-18","cited_by_count":9,"area":null},{"id":"2602.23036","title":"LLMServingSim 2.0: A Unified Simulator for Heterogeneous and Disaggregated LLM Serving Infrastructure","created":"2026-02-26","cited_by_count":9,"area":"inference"},{"id":"2312.05417","title":"ESPN: Memory-Efficient Multi-Vector Information Retrieval","created":"2023-12-09","cited_by_count":8,"area":null},{"id":"2406.08332","title":"UDON: Universal Dynamic Online distillatioN for generic image representations","created":"2024-06-12","cited_by_count":8,"area":"distill"},{"id":"2408.14158","title":"Fire-Flyer AI-HPC: A Cost-Effective Software-Hardware Co-Design for Deep Learning","created":"2024-08-26","cited_by_count":8,"area":null},{"id":"2412.01377","title":"Adapting Large Language Models to Log Analysis with Interpretable Domain Knowledge","created":"2024-12-02","cited_by_count":8,"area":null},{"id":"2603.17435","title":"ZipServ: Fast and Memory-Efficient LLM Inference with Hardware-Aware Lossless Compression","created":"2026-03-18","cited_by_count":8,"area":"inference"},{"id":"2310.00898","title":"Enabling Language Models to Implicitly Learn Self-Improvement","created":"2023-10-02","cited_by_count":7,"area":"alignment"},{"id":"2405.13938","title":"eXmY: A Data Type and Technique for Arbitrary Bit Precision Quantization","created":"2024-05-22","cited_by_count":7,"area":null},{"id":"2406.10903","title":"New Solutions on LLM Acceleration, Optimization, and Application","created":"2024-06-16","cited_by_count":7,"area":null},{"id":"2406.12774","title":"Towards Exact Gradient-based Training on Analog In-memory Computing","created":"2024-06-18","cited_by_count":7,"area":null},{"id":"2501.03905","title":"MixNet: A Runtime Reconfigurable Optical-Electrical Fabric for Distributed Mixture-of-Experts Training","created":"2025-01-07","cited_by_count":7,"area":"moe"},{"id":"2501.04052","title":"RaZeR: Pushing the Limits of NVFP4 Quantization with Redundant Zero Remapping","created":"2025-01-06","cited_by_count":7,"area":null},{"id":"2502.03885","title":"InfiniteHBD: Building Datacenter-Scale High-Bandwidth Domain for LLM with Optical Circuit Switching Transceivers","created":"2025-02-06","cited_by_count":7,"area":null},{"id":"2504.09307","title":"Lumos: Efficient Performance Modeling and Estimation for Large-scale LLM Training","created":"2025-04-12","cited_by_count":7,"area":null},{"id":"2505.01386","title":"CATransformers: Carbon Aware Transformers Through Joint Model-Hardware Optimization","created":"2025-05-02","cited_by_count":7,"area":null},{"id":"2505.22913","title":"Mustafar: Promoting Unstructured Sparsity for KV Cache Pruning in LLM Inference","created":"2025-05-28","cited_by_count":7,"area":"inference"},{"id":"2605.26097","title":"Forgetting in Language Models: Capacity, Optimization, and Self-Generated Replay","created":"2026-05-25","cited_by_count":7,"area":"synth"},{"id":"2306.08888","title":"ArchGym: An Open-Source Gymnasium for Machine Learning Assisted Architecture Design","created":"2023-06-15","cited_by_count":6,"area":null},{"id":"2310.11096","title":"Sparse-DySta: Sparsity-Aware Dynamic and Static Scheduling for Sparse Multi-DNN Workloads","created":"2023-10-17","cited_by_count":6,"area":null},{"id":"2401.08895","title":"cedar: Optimized and Unified Machine Learning Input Data Pipelines","created":"2024-01-17","cited_by_count":6,"area":null},{"id":"2402.07393","title":"TeMPO: Efficient Time-Multiplexed Dynamic Photonic Tensor Core for Edge AI with Compact Slow-Light Electro-Optic Modulator","created":"2024-02-12","cited_by_count":6,"area":null},{"id":"2501.16634","title":"Towards Resource-Efficient Compound AI Systems","created":"2025-01-28","cited_by_count":6,"area":null},{"id":"2510.08544","title":"SPAD: Specialized Prefill and Decode Hardware for Disaggregated LLM Inference","created":"2025-10-09","cited_by_count":6,"area":null},{"id":"2602.03545","title":"Persona Generators: Generating Diverse Synthetic Personas for Arbitrary Contexts","created":"2026-02-03","cited_by_count":6,"area":null},{"id":"2305.19505","title":"M3ICRO: Machine Learning-Enabled Compact Photonic Tensor Core based on PRogrammable Multi-Operand Multimode Interference","created":"2023-05-31","cited_by_count":5,"area":null},{"id":"2311.06720","title":"Cappy: Outperforming and Boosting Large Multi-Task LMs with a Small Scorer","created":"2023-11-12","cited_by_count":5,"area":null},{"id":"2311.08357","title":"Sparsity-Preserving Differentially Private Training of Large Embedding Models","created":"2023-11-14","cited_by_count":5,"area":null},{"id":"2411.18065","title":"FlexiBit: Fully Flexible Precision Bit-parallel Accelerator Architecture for Arbitrary Mixed Precision AI","created":"2024-11-27","cited_by_count":5,"area":null},{"id":"2502.06309","title":"Analog In-memory Training on General Non-ideal Resistive Elements: The Impact of Response Functions","created":"2025-02-10","cited_by_count":5,"area":null},{"id":"2506.06644","title":"Spark Transformer: Reactivating Sparsity in FFN and Attention","created":"2025-06-07","cited_by_count":5,"area":null},{"id":"2509.06794","title":"Dato: A Task-Based Programming Model for Dataflow Accelerators","created":"2025-09-08","cited_by_count":5,"area":null},{"id":"2510.16805","title":"Mixed-Precision Quantization for Language Models: Techniques and Prospects","created":"2025-10-19","cited_by_count":5,"area":null},{"id":"2304.05301","title":"TACOS: Topology-Aware Collective Algorithm Synthesizer for Distributed Machine Learning","created":"2023-04-11","cited_by_count":4,"area":null},{"id":"2402.00028","title":"Neural Rendering and Its Hardware Acceleration: A Review","created":"2024-01-06","cited_by_count":4,"area":null},{"id":"2410.15977","title":"Enabling Energy-Efficient Deployment of Large Language Models on Memristor Crossbar: A Synergy of Large and Small","created":"2024-10-21","cited_by_count":4,"area":null},{"id":"2411.16712","title":"SafeLight: Enhancing Security in Optical Convolutional Neural Network Accelerators","created":"2024-11-22","cited_by_count":4,"area":null}],"next_cursor":"100"}