Jiacheng Zhao

Shuoming Zhang, Qiuchu Yu, Yangyu Zhang et al.

LLM-based agents are increasingly used to generate GPU kernels, but they often know what optimizations to try without knowing when those optimizations are sound. We introduce KLineage, which learns this missing "when" knowledge from expert kernels: instead of relying on forward rollouts, KLineage walks expert implementations backward through validation-gated simplifications and reverses each accepted step into a reusable optimization skill. Each skill records not only the optimization intent, but also where it applies in code, what conditions made it valid, what effect it had, and what failures its assumptions avoid. A downstream LLM materializes these skills on new code surfaces under the same compile/correctness/profile gate. On five expert workloads across two NVIDIA architectures, these lineage-derived skills serve as an effective optimization curriculum, exceeding recent memory-based LLM-kernel baselines in both final kernel quality and optimization efficiency under the same fixed budget. We additionally use a separate 22-instance held-out check as a sanity test against source-case memorization.

MiniCPM Team, Wenhao An, Yingfa Chen et al. · tsinghua

The evolution of large language models (LLMs) towards applications with ultra-long contexts faces challenges posed by the high computational and memory costs of the Transformer architecture. While existing sparse and linear attention mechanisms attempt to mitigate these issues, they typically involve a trade-off between memory efficiency and model performance. This paper introduces MiniCPM-SALA, a 9B-parameter hybrid architecture that integrates the high-fidelity long-context modeling of sparse attention (InfLLM-V2) with the global efficiency of linear attention (Lightning Attention). By employing a layer selection algorithm to integrate these mechanisms in a 1:3 ratio and utilizing a hybrid positional encoding (HyPE), the model maintains efficiency and performance for long-context tasks. Furthermore, we introduce a cost-effective continual training framework that transforms pre-trained Transformer-based models into hybrid models, which reduces training costs by approximately 75% compared to training from scratch. Extensive experiments show that MiniCPM-SALA maintains general capabilities comparable to full-attention models while offering improved efficiency. On a single NVIDIA A6000D GPU, the model achieves up to 3.5x the inference speed of the full-attention model at the sequence length of 256K tokens and supports context lengths of up to 1M tokens, a scale where traditional full-attention 8B models fail because of memory constraints.

Haohui Mai, Xiaoyan Guo, Xiangyun Ding et al.

LLM-based coding agents can generate functionally correct GPU kernels, yet their performance remains far below hand-optimized libraries on critical computations such as matrix multiplication, attention, and Mixture-of-Experts (MoE). Peak GPU performance requires coordinated reasoning over tightly coupled optimizations, including tiling, shared-memory staging, software pipelining, and instruction scheduling, while existing agents rely on sparse pass/fail feedback, leaving them unable to diagnose global constraint violations. We present Argus, an agentic framework that addresses this through data-flow invariants: compile-time specifications encoding how data must be choreographed throughout kernel execution. Argus introduces a tile-based, Pythonic DSL exposing hardware instructions and compiler policies while hiding low-level representations. The DSL provides tag functions to propagate symbolic annotations through data and control flow, and tag assertions to enforce relational constraints at use sites. When violations occur, the compiler returns concrete counterexamples identifying the thread, data element, and program point, enabling dense, structured feedback for targeted fixes. Invariants are verified at compile time via abstract interpretation over a layout algebra and SMT solving, with zero runtime overhead. An in-context reinforcement learning planner learns to select optimizations and synthesize effective invariants, supported by a curated knowledge base of GPU optimization techniques. We evaluate Argus on the AMD MI300X GPU across GEMM, flash attention, and MoE kernels accounting for over 90% of GPU time in LLM inference. Generated kernels achieve 99-104% of state-of-the-art hand-optimized assembly throughput and are 2-1543x faster than existing agentic systems. Argus further generalizes to 200 KernelBench tasks, solving 100% of Level 1 and 90% of Level 2 problems.

Shuoming Zhang, Jiacheng Zhao, Qiuchu Yu et al.

This survey has provided a systematic overview of the emerging field of LLM-enabled compilation by addressing several key research questions. We first answered how LLMs are being integrated by proposing a comprehensive, multi-dimensional taxonomy that categorizes works based on their Design Philosophy (Selector, Translator, Generator), LLM Methodology, their operational Level of Code Abstraction, and the specific Task Type they address. In answering what advancements these approaches offer, we identified three primary benefits: the democratization of compiler development, the discovery of novel optimization strategies, and the broadening of the compiler's traditional scope. Finally, in addressing the field's challenges and opportunities, we highlighted the critical hurdles of ensuring correctness and achieving scalability, while identifying the development of hybrid systems as the most promising path forward. By providing these answers, this survey serves as a foundational roadmap for researchers and practitioners, charting the course for a new generation of LLM-powered, intelligent, adaptive and synergistic compilation tools.

Shuoming Zhang, Jiacheng Zhao, Ruiyuan Xu et al.

Content Warning: This paper may contain unsafe or harmful content generated by LLMs that may be offensive to readers. Large Language Models (LLMs) are extensively used as tooling platforms through structured output APIs to ensure syntax compliance so that robust integration with existing softwares like agent systems, could be achieved. However, the feature enabling functionality of grammar-guided structured output presents significant security vulnerabilities. In this work, we reveal a critical control-plane attack surface orthogonal to traditional data-plane vulnerabilities. We introduce Constrained Decoding Attack (CDA), a novel jailbreak class that weaponizes structured output constraints to bypass safety mechanisms. Unlike prior attacks focused on input prompts, CDA operates by embedding malicious intent in schema-level grammar rules (control-plane) while maintaining benign surface prompts (data-plane). We instantiate this with a proof-of-concept Chain Enum Attack, achieves 96.2% attack success rates across proprietary and open-weight LLMs on five safety benchmarks with a single query, including GPT-4o and Gemini-2.0-flash. Our findings identify a critical security blind spot in current LLM architectures and urge a paradigm shift in LLM safety to address control-plane vulnerabilities, as current mechanisms focused solely on data-plane threats leave critical systems exposed.

Rui Zhang, Yuanbo Wen, Shuyao Cheng et al.

Processor chip design technology serves as a key frontier driving breakthroughs in computer science and related fields. With the rapid advancement of information technology, conventional design paradigms face three major challenges: the physical constraints of fabrication technologies, the escalating demands for design resources, and the increasing diversity of ecosystems. Automated processor chip design has emerged as a transformative solution to address these challenges. While recent breakthroughs in Artificial Intelligence (AI), particularly Large Language Models (LLMs) techniques, have opened new possibilities for fully automated processor chip design, substantial challenges remain in establishing domain-specific LLMs for processor chip design. In this paper, we propose QiMeng, a novel system for fully automated hardware and software design of processor chips. QiMeng comprises three hierarchical layers. In the bottom-layer, we construct a domain-specific Large Processor Chip Model (LPCM) that introduces novel designs in architecture, training, and inference, to address key challenges such as knowledge representation gap, data scarcity, correctness assurance, and enormous solution space. In the middle-layer, leveraging the LPCM's knowledge representation and inference capabilities, we develop the Hardware Design Agent and the Software Design Agent to automate the design of hardware and software for processor chips. Currently, several components of QiMeng have been completed and successfully applied in various top-layer applications, demonstrating significant advantages and providing a feasible solution for efficient, fully automated hardware/software design of processor chips. Future research will focus on integrating all components and performing iterative top-down and bottom-up design processes to establish a comprehensive QiMeng system.

Shuoming Zhang, Jiacheng Zhao, Chunwei Xia et al.

Large language models (LLMs) have the potential to revolutionize how we design and implement compilers and code translation tools. However, existing LLMs struggle to handle long and complex programs. We introduce LEGO-Compiler, a novel neural compilation system that leverages LLMs to translate high-level languages into assembly code. Our approach centers on three key innovations: LEGO translation, which decomposes the input program into manageable blocks; breaking down the complex compilation process into smaller, simpler verifiable steps by organizing it as a verifiable LLM workflow by external tests; and a feedback mechanism for self-correction. Supported by formal proofs of translation composability, LEGO-Compiler demonstrates high accuracy on multiple datasets, including over 99% on ExeBench and 97.9% on industrial-grade AnsiBench. Additionally, LEGO-Compiler has also acheived near one order-of-magnitude improvement on compilable code size scalability. This work opens new avenues for applying LLMs to system-level tasks, complementing traditional compiler technologies.

Jixuan He, Yutong Guo, Jiacheng Zhao

This paper delves into the paramount significance of information dissemination within the dynamic realm of social networks. It underscores the pivotal role of information communication models in unraveling the intricacies of data propagation in the digital age. By shedding light on the profound influence of these models, it not only lays the groundwork for exploring various hierarchies and their manifestations but also serves as a catalyst for further research in this formidable field.