GraphIR: Architecture-Level Search States for LLM-Guided Neural Architecture Evolution

• URL: http://arxiv.org/abs/2608.01633v1
• Date: Mon, 03 Aug 2026 03:06:58 GMT
• Title: GraphIR: Architecture-Level Search States for LLM-Guided Neural Architecture Evolution
• Abstract summary: Large language models (LLMs) enable neural computation search (NAS) directly over executable neural network programs.<n>LLMs must infer tensor dependencies, editable components, and compatibility constraints from implementation details.<n>We propose GraphIR, an architecture-aware intermediate representation that supplements executable programs with a mutation-aligned candidate state.
• Score: 34.415252210308466
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models (LLMs) enable neural architecture search (NAS) directly over executable neural network programs. However, code-level flexibility does not provide the architecture state needed for effective mutation: LLMs must infer tensor dependencies, editable components, and compatibility constraints from implementation details. To address this representation mismatch, we propose GraphIR, an architecture-aware intermediate representation that supplements executable programs with a mutation-aligned candidate state. GraphIR organizes each candidate through three complementary views: a computation skeleton describing tensor flow, a mutation surface exposing editable modules and operations, and a validity envelope capturing interface contracts, propagated shapes, and downstream dependencies. To evaluate our method, we construct NAS-Dependency, a 120-question benchmark covering six complementary dependency-reasoning dimensions. The diagnostic shows that GraphIR is particularly effective at identifying exact producer occurrences, tracing dependency propagation, and diagnosing interface and failure risks. Across six downstream benchmarks including CLRS, GraphIR achieves the best overall search performance while maintaining comparable model size and favorable end-to-end NAS efficiency when integrated into OpenEvolve. These results show that a mutation-oriented architecture state provides an effective interface between executable neural programs and LLM-guided architecture evolution.

Related papers

• VeriGraphi: A Multi-Agent Framework of Hierarchical RTL Generation for Large Hardware Designs [5.354071922171165]
  We present VeriGraphi, a framework that introduces a spec-anchored Knowledge Graph as the architectural substrate driving the RTL generation pipeline.<n>We show that VeriGraphi enables reliable hierarchical RTL generation with minimal human intervention for RISC-V.
  arXiv  Detail & Related papers  (2026-04-16T02:30:57Z)
• Can Graph Foundation Models Generalize Over Architecture? [8.734283737027633]
  Graph foundation models (GFMs) can generalize zero-shot across graphs of arbitrary scales, feature dimensions, and domains.<n>Existing GFMs rely on fixed architectural backbones, implicitly assuming that a single message-passing regime suffices across tasks.<n>We introduce a framework that adapts effective GNN architecture at inference time by discovering and mixing task-specific linear graph operators.
  arXiv  Detail & Related papers  (2026-03-24T09:22:25Z)
• CARD: Towards Conditional Design of Multi-agent Topological Structures [83.18278008173746]
  CARD (Conditional Agentic Graph Designer) is a conditional graph-generation framework that instantiates AMACP, a protocol for adaptive multi-agent communication.<n> CARD produces communication structures that are both effective and resilient to shifts in model capability or resource availability.
  arXiv  Detail & Related papers  (2026-03-01T13:02:36Z)
• Closed-Loop LLM Discovery of Non-Standard Channel Priors in Vision Models [48.83701310501069]
  Large Language Models (LLMs) offer a transformative approach to Neural Architecture Search (NAS)<n>We formulate the search as a sequence of conditional code generation tasks, where an LLM refines architectural specifications based on performance telemetry.<n>We generate a vast corpus of valid, shape-consistent architectures via Abstract Syntax Tree (AST) mutations.<n> Experimental results on CIFAR-100 validate the efficacy of this approach, demonstrating that the model yields statistically significant improvements in accuracy.
  arXiv  Detail & Related papers  (2026-01-13T13:00:30Z)
• GLOW: Graph-Language Co-Reasoning for Agentic Workflow Performance Prediction [51.83437071408662]
  We propose GLOW, a unified framework for AW performance prediction.<n>GLOW combines the graph-structure modeling capabilities of GNNs with the reasoning power of LLMs.<n>Experiments on FLORA-Bench show that GLOW outperforms state-of-the-art baselines in prediction accuracy and ranking utility.
  arXiv  Detail & Related papers  (2025-12-11T13:30:46Z)
• Dynamic Generation of Multi-LLM Agents Communication Topologies with Graph Diffusion Models [99.85131798240808]
  We introduce a novel generative framework called textitGuided Topology Diffusion (GTD)<n>Inspired by conditional discrete graph diffusion models, GTD formulates topology synthesis as an iterative construction process.<n>At each step, the generation is steered by a lightweight proxy model that predicts multi-objective rewards.<n>Experiments show that GTD can generate highly task-adaptive, sparse, and efficient communication topologies.
  arXiv  Detail & Related papers  (2025-10-09T05:28:28Z)
• GILT: An LLM-Free, Tuning-Free Graph Foundational Model for In-Context Learning [50.40400074353263]
  Graph Neural Networks (GNNs) are powerful tools for precessing relational data but often struggle to generalize to unseen graphs.<n>We introduce textbfGraph textbfIn-context textbfL textbfTransformer (GILT), a framework built on an LLM-free and tuning-free architecture.
  arXiv  Detail & Related papers  (2025-10-06T08:09:15Z)
• Learning Efficient and Generalizable Graph Retriever for Knowledge-Graph Question Answering [75.12322966980003]
  Large Language Models (LLMs) have shown strong inductive reasoning ability across various domains.<n>Most existing RAG pipelines rely on unstructured text, limiting interpretability and structured reasoning.<n>Recent studies have explored integrating knowledge graphs with LLMs for knowledge graph question answering.<n>We propose RAPL, a novel framework for efficient and effective graph retrieval in KGQA.
  arXiv  Detail & Related papers  (2025-06-11T12:03:52Z)
• Scalability Matters: Overcoming Challenges in InstructGLM with Similarity-Degree-Based Sampling [1.2805157669888096]
  We propose SDM-InstructGLM, a novel instruction-tuned Graph Language Model (InstructGLM) framework that enhances scalability and efficiency without relying on GNNs.<n>Our method introduces a similarity-degree-based biased random walk mechanism, which selectively samples and encodes graph information based on node-feature similarity and degree centrality.<n>Our results demonstrate the feasibility of LLM-only graph processing, enabling scalable and interpretable Graph Language Models (GLMs) optimized through instruction-based fine-tuning.
  arXiv  Detail & Related papers  (2025-05-02T06:08:21Z)
• LASE: Learned Adjacency Spectral Embeddings [9.227991604045416]
  We learn nodal Adjacency Spectral Embeddings (ASE) from graph inputs.<n>LASE is interpretable, parameter efficient, robust to inputs with unobserved edges.<n>LASE layers combine Graph Convolutional Network (GCN) and fully-connected Graph Attention Network (GAT) modules.
  arXiv  Detail & Related papers  (2024-12-23T17:35:19Z)
• DepGraph: Towards Any Structural Pruning [68.40343338847664]
  We study general structural pruning of arbitrary architecture like CNNs, RNNs, GNNs and Transformers. We propose a general and fully automatic method, emphDependency Graph (DepGraph), to explicitly model the dependency between layers and comprehensively group parameters for pruning. In this work, we extensively evaluate our method on several architectures and tasks, including ResNe(X)t, DenseNet, MobileNet and Vision transformer for images, GAT for graph, DGCNN for 3D point cloud, alongside LSTM for language, and demonstrate that, even with a
  arXiv  Detail & Related papers  (2023-01-30T14:02:33Z)
• Arch-Graph: Acyclic Architecture Relation Predictor for Task-Transferable Neural Architecture Search [96.31315520244605]
  Arch-Graph is a transferable NAS method that predicts task-specific optimal architectures. We show Arch-Graph's transferability and high sample efficiency across numerous tasks. It is able to find top 0.16% and 0.29% architectures on average on two search spaces under the budget of only 50 models.
  arXiv  Detail & Related papers  (2022-04-12T16:46:06Z)
• Neural combinatorial optimization beyond the TSP: Existing architectures under-represent graph structure [9.673093148930876]
  We analyze how and whether recent neural architectures can be applied to graph problems of practical importance. We show that augmenting the structural representation of problems with Distance is a promising step towards the still-ambitious goal of learning multi-purpose autonomous solvers.
  arXiv  Detail & Related papers  (2022-01-03T14:14:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.