News
Paper reviews
Written on 25.10.2026 17:06 by Basel Fakhri
Student |
Review 1 |
Review 2 |
Review 3 |
| Sindarov, Asadbek | H2-LLM: Hardware-Dataflow Co-Exploration for Heterogeneous Hybrid-Bonding-based Low-Batch LLM Inference | FlashInfer: Efficient and Customizable Attention Engine for LLM Inference Serving | MPK: A Compiler and Runtime for Mega-Kernelizing Tensor Programs |
| Kazemi, Mehrafarin | TANGRAM: Optimized Coarse-Grained Dataflow for Scalable NN Accelerators | POD-Attention: Unlocking Full Prefill-Decode Overlap for Faster LLM Inference |
DistServe: Disaggregating Prefill and Decoding for Goodputoptimized Large Language Model Serving |
| Le, Hoang Dang Nguyen | MPK: A Compiler and Runtime for Mega-Kernelizing Tensor Programs | ARC: Warp-level Adaptive Atomic Reduction in GPUs to Accelerate Differentiable Rendering | ThunderKittens: Simple, Fast, and Adorable AI Kernels |
| Kooshky, Amirmohammad | Eyeriss: A Spatial Architecture for Energy-Efficient Dataflow for Convolutional Neural Networks |
MP-Rec: Hardware-Software Co-Design to Enable Multi-Path Recommendation | TVM: An Automated End-to-End Optimizing Compiler for Deep Learning |
| Farag, Yousef | Efficient Memory Management for Large Language Model Serving with PagedAttention | Orca: A Distributed Serving System for Transformer-Based Generative Models | ARC: Warp-level Adaptive Atomic Reduction in GPUs to Accelerate Differentiable Rendering |
| Hariharan, Tarun | POD-Attention: Unlocking Full PrefillDecode Overlap for Faster LLM Inference |
PointAcc: Efficient Point Cloud Accelerator | throttLL’eM: Predictive GPU Throttling for Energy Efficient LLM Inference Serving |
| Wang, Zirui | Zeus: Understanding and Optimizing GPU Energy Consumption of DNN Training | ELSA: Hardware-Software Co-Design for Efficient, Lightweight Self-Attention Mechanism in Neural Networks | Efficient Memory Management for Large Language Model Serving with PagedAttention |
| Patel, Nehali Sudarshan | ELSA: Hardware-Software Co-Design for Efficient, Lightweight Self-Attention Mechanism in Neural Networks | FlashInfer: Efficient and Customizable Attention Engine for LLM Inference Serving | AttAcc! Unleashing the Power of PIM for Batched Transformer-based Generative Model Inference |
| K.A., Abdul Manan | WLB-LLM: Workload-Balanced 4D Parallelism for Large Language Model Training | FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness | ZeRO: Memory Optimizations Toward Training Trillion Parameter Models |
| Subedi, Ashish | H2-LLM: Hardware-Dataflow Co-Exploration for Heterogeneous Hybrid-Bonding-based Low-Batch LLM Inference | Efficient Large-Scale Language Model Training on GPU Clusters Using Megatron-LM | DynamoLLM: Designing LLM Inference Clusters for Performance and Energy Efficiency |
| Suntoro, Marcellius William | H2-LLM: Hardware-Dataflow Co-Exploration for Heterogeneous Hybrid-Bonding-based Low-Batch LLM Inference |
Orca: A Distributed Serving System for Transformer-Based Generative Models | NanoFlow: Towards Optimal Large Language Model Serving Throughput |