IS
Ion Stoica
cs.LGcs.AIcs.DCcs.CLstat.MLcs.DBcs.CVcs.SEcs.NIcs.PL
On Valency
published · living versionsW_u7fa5rax·v1 · currentpublished
MLSys: The New Frontier of Machine Learning Systems
with Alexander Ratner, Dan Alistarh, Gustavo Alonso, David G. Andersen +64
1 version
Preprints & journals
236 papers in the corpus · 1966–2026Combee: Scaling Prompt Learning for Self-Improving Language Model Agents2604.04247v2 · Hanchen Li, Runyuan He, Qizheng Zhang et al.2026 · 0 citationsarXiv
Crafting Interpretable Embeddings for Language Neuroscience by Asking LLMs Questions.40276238 · Benara, Vinamra, Singh, Chandan, Morris, John X et al.2026 · 2 citationsAdvances in neural information processing systems. 2024;37:124137-124162
MobileCybench: Evaluating Agent Vulnerability Discovery via Executable Probes2609.23980v1 · Andy K. Zhang, Ava Huang, Joey Ji et al.2026 · 0 citationsarXiv
Reality Is the Final Verifier: On Two Key Gaps in Agentic Software Engineering2609.12039v1 · Alexander Krentsel, Shubham Agarwal, Mert Cemri et al.2026 · 0 citationsarXiv
What Happens When the Model Eats the Stack? Rethinking the Research Agenda for Data Agents to Withstand the Bitter Lesson2609.03141v1 · Liana Patel, Siddharth Jha, Negar Arabzadeh et al.2026 · 0 citationsarXiv
FreeToken: Efficient Edge-Native MoE Serving with Bandwidth-Adaptive Execution2608.16157v1 · Shuo Yang, Xiaoze Fan, Melissa Pan et al.2026 · 0 citationsarXiv
LLM-as-a-Verifier: A General-Purpose Verification Framework2607.05391v2 · Jacky Kwok, Shulu Li, Pranav Atreya et al.2026 · 0 citationsarXiv
DualEval: Joint Model-Item Calibration for Unified LLM Evaluation2606.26429v1 · Aaron J. Li, Hao Huang, Youngmin Park et al.2026 · 0 citationsarXiv
Playful Agentic Robot Learning2606.19419v1 · Junyi Zhang, Jiaxin Ge, Hanjun Yoo et al.2026 · 0 citationsarXiv
BlendServe: Optimizing Offline Inference for Auto-regressive Large Models with Resource-aware Batching2411.16102v2 · Yilong Zhao, Shuo Yang, Kan Zhu et al.2024 · 0 citationsarXiv
Idleness is Relative: Exploiting Tool-Call Idle Windows for Offloading in Agentic Systems with MORI2606.00866v1 · Tian Xia, Hanchen Li, Zhifei Li et al.2026 · 0 citationsarXiv
BenchEvolver: Frontier Task Synthesis via Solution-Centric Evolution2606.01286v1 · Yangzhen Wu, Aaron J. Li, Wenjie Ma et al.2026 · 0 citationsarXiv
The Price Reversal Phenomenon: When Cheaper Reasoning Models Cost More2603.23971v2 · Lingjiao Chen, Chi Zhang, Yeye He et al.2026 · 0 citationsarXiv
vAttention: Verified Sparse Attention2510.05688v2 · Aditya Desai, Kumar Krishna Agrawal, Shuo Yang et al.2025 · 0 citationsProceedings of the International Conference on Learning Representations (ICLR), 2026
The Time is Here for Just-in-Time Systems: Challenges and Opportunities2605.24096v1 · Shu Liu, Alexander Krentsel, Shubham Agarwal et al.2026 · 0 citationsarXiv
Inference Time Context Sparsity: Illusion or Opportunity?2605.24168v1 · Sahil Joshi, Prithvi Dixit, Agniva Chowdhury et al.2026 · 0 citationsarXiv
Inductive Deductive Synthesis: Enabling AI to Generate Formally Verified Systems2605.23109v1 · Shubham Agarwal, Alexander Krentsel, Shu Liu et al.2026 · 0 citationsarXiv
optimize_anything: A Universal API for Optimizing any Text Parameter2605.19633v1 · Lakshya A Agrawal, Donghyun Lee, Shangyin Tan et al.2026 · 1 citationProceedings of the ACM Conference on AI and Agentic Systems (CAIS 26), May 26-29, 2026, San Jose, CA, USA
AstraFlow: Dataflow-Oriented Reinforcement Learning for Agentic LLMs2605.15565v1 · Haizhong Zheng, Yizhuo Di, Jiahui Wang et al.2026 · 0 citationsarXiv
Sparse VideoGen2: Accelerate Video Generation with Sparse Attention via Semantic-Aware Permutation2505.18875v5 · Shuo Yang, Haocheng Xi, Yilong Zhao et al.2025 · 5 citationsarXiv
RLBoost: Harvesting Preemptible Resources for Cost-Efficient Reinforcement Learning on LLMs2510.19225v3 · Yongji Wu, Xueshen Liu, Haizhong Zheng et al.2025 · 0 citationsarXiv
AI-Driven Research for Databases2604.06566v1 · Audrey Cheng, Harald Ng, Aaron Kabcenell et al.2026 · 0 citationsarXiv
Foundry: Template-Based CUDA Graph Context Materialization for Fast LLM Serving Cold Start2604.06664v1 · Xueshen Liu, Yongji Wu, Yuncheng Yao et al.2026 · 0 citationsarXiv
SonicMoE: Accelerating MoE with IO and Tile-aware Optimizations2512.14080v2 · Wentao Guo, Mayank Mishra, Xinle Cheng et al.2025 · 0 citationsarXiv
K-Search: LLM Kernel Generation via Co-Evolving Intrinsic World Model2602.19128v2 · Shiyi Cao, Ziming Mao, Joseph E. Gonzalez et al.2026 · 0 citationsarXiv
AdaEvolve: Adaptive LLM Driven Zeroth-Order Optimization2602.20133v1 · Mert Cemri, Shubham Agrawal, Akshat Gupta et al.2026 · 2 citationsarXiv
GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning2507.19457v2 · Lakshya A Agrawal, Shangyin Tan, Dilara Soylu et al.2025 · 3 citationsarXiv
TURBO: Utility-Aware Bandwidth Allocation for Cloud-Augmented Autonomous Control2503.20127v2 · Peter Schafhalter, Alexander Krentsel, Hongbo Wei et al.2025 · 15 citationsarXiv
UCCL-EP: Portable Expert-Parallel Communication2512.19849v2 · Ziming Mao, Yihan Zhang, Chihan Cui et al.2025 · 0 citationsarXiv
Let the Barbarians In: How AI Can Accelerate Systems Performance Research2512.14806v4 · Audrey Cheng, Shu Liu, Melissa Pan et al.2025 · 0 citationsarXiv
TurboDiffusion: Accelerating Video Diffusion Models by 100-200 Times2512.16093v1 · Jintao Zhang, Kaiwen Zheng, Kai Jiang et al.2025 · 0 citationsarXiv
Radial Attention: $O(n\log n)$ Sparse Attention with Energy Decay for Long Video Generation2506.19852v2 · Xingyang Li, Muyang Li, Tianle Cai et al.2025 · 0 citationsarXiv
Supporting Our AI Overlords: Redesigning Data Systems to be Agent-First2509.00997v2 · Shu Liu, Soujanya Ponnapalli, Shreya Shankar et al.2025 · 2 citationsarXiv
Accelerating Large-Scale Reasoning Model Inference with Sparse Self-Speculative Decoding2512.01278v1 · Yilong Zhao, Jiaming Tang, Kan Zhu et al.2025 · 0 citationsarXiv
LEANN: A Low-Storage Vector Index2506.08276v2 · Yichuan Wang, Zhifei Li, Shu Liu et al.2025 · 0 citationsarXiv
SkyRL-Agent: Efficient RL Training for Multi-turn LLM Agent2511.16108v1 · Shiyi Cao, Dacheng Li, Fangzhou Zhao et al.2025 · 0 citationsarXiv
EDIT-Bench: Evaluating LLM Abilities to Perform Real-World Instructed Code Edits2511.04486v2 · Wayne Chi, Valerie Chen, Ryan Shar et al.2025 · 0 citationsarXiv
SkyWalker: A Locality-Aware Cross-Region Load Balancer for LLM Inference2505.24095v2 · Tian Xia, Ziming Mao, Jamison Kerney et al.2025 · 2 citationsarXiv
Twilight: Adaptive Attention Sparsity with Hierarchical Top-$p$ Pruning2502.02770v5 · Chaofan Lin, Jiaming Tang, Shuo Yang et al.2025 · 0 citationsarXiv
VSA: Faster Video Diffusion with Trainable Sparse Attention2505.13389v5 · Peiyuan Zhang, Yongqi Chen, Haofeng Huang et al.2025 · 0 citationsarXiv
Why Do Multi-Agent LLM Systems Fail?2503.13657v3 · Mert Cemri, Melissa Z. Pan, Shuyi Yang et al.2025 · 44 citationsarXiv
GSO: Challenging Software Optimization Tasks for Evaluating SWE-Agents2505.23671v3 · Manish Shetty, Naman Jain, Jinjian Liu et al.2025 · 0 citationsarXiv
The Streaming Batch Model for Efficient and Fault-Tolerant Heterogeneous Execution2501.12407v5 · Frank Sifei Luan, Ron Yifeng Wang, Yile Gu et al.2025 · 0 citationsarXiv
Efficient Long-context Language Model Training by Core Attention Disaggregation2510.18121v1 · Yonghao Zhuang, Junda Chen, Bo Pang et al.2025 · 0 citationsarXiv
Barbarians at the Gate: How AI is Upending Systems Research2510.06189v3 · Audrey Cheng, Shu Liu, Melissa Pan et al.2025 · 1 citationarXiv
Towards Efficient and Practical GPU Multitasking in the Era of LLM2508.08448v1 · Jiarong Xing, Yifan Qiao, Simon Mo et al.2025 · 0 citationsarXiv
An Extensible Software Transport Layer for GPU Networking2504.17307v2 · Yang Zhou, Zhongjie Chen, Ziming Mao et al.2025 · 1 citationarXiv
Advancing Science- and Evidence-based AI Policy2508.02748v1 · Rishi Bommasani, Sanjeev Arora, Jennifer Chayes et al.2025 · 22 citationsarXiv
TurboSpec: Closed-loop Speculation Control System for Optimizing LLM Serving Goodput2406.14066v3 · Xiaoxuan Liu, Jongseok Park, Langxiang Hu et al.2024 · 0 citationsarXiv
OR-Bench: An Over-Refusal Benchmark for Large Language Models2405.20947v5 · Justin Cui, Wei-Lin Chiang, Ion Stoica et al.2024 · 9 citationsarXiv
Career total: 737 works. 236 are in this corpus.Showing the 50 most recent.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.