mlsanityData to 14 Oct 2026

OASIF: An Efficient Obfuscation-Aware Self-Improving Framework for LLM-Based Assembly Code Instruction Following and Comprehension

arXiv 2606.29155 ↗First posted 28 Jun 2026cs.SEcited by 0

On the map: Synthetic data and self-improvement for LLMs → Code and instruction data synthesis

Abstract (arXiv)

Large Language Models (LLMs) have recently shown promise in automated binary analysis, yet they remain brittle under commercial-grade obfuscation. We present OASIF, an Obfuscation-Aware Self-evolving Instruction-Following framework for obfuscated assembly comprehension. OASIF couples a token-efficient assembly encoder with a lightweight projector to expose long obfuscated code to a pretrained code LLM under a bounded context budget and follows a three-phase training: (i) feature-space alignment, (ii) supervised instruction fine-tuning, and (iii) online self-evolving reinforcement learning with hybrid rewards, enabling continual adaptation with minimal manual verification. On VMISA-Bench, a challenging out-of-distribution suite featuring three commercial VM-based obfuscators, OASIF consistently improves open-source backbones; Qwen2.5-Coder-Instruct-14B attains Success Rate gains of +15.9, +5.8, and +16.9 percentage points (pp) on Code Virtualizer, Themida (v3.0.7), and VMProtect (v3.5), respectively, and improves the OASIF-Bench average by +9.8. OASIF further delivers stable gains across seven standard BCSD benchmarks while preserving general and domain-relevant capabilities on HumanEval, VulBench, and HumanEval-Decompile.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Code and instruction data synthesis, one of 13 directions in Synthetic data and self-improvement for LLMs. Builds on 24 of its 42 references that are in the map. First posted on arXiv 28 Jun 2026.

Builds on

24 of 42 references.

Closest ancestors · 1

Same direction, newest first

  1. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructions20 Dec

Other references · 23

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2023GPT-4 Technical Reportcited by 18,396
  3. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  4. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  5. 2021Evaluating Large Language Models Trained on Codecited by 8,320
  6. 2023Visual Instruction Tuningcited by 7,869
  7. 2023Qwen Technical Reportcited by 3,344
  8. 2024Qwen2.5-Coder Technical Reportcited by 1,203
  9. 2024DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligencecited by 1,181
  10. 2023ShareGPT4V: Improving Large Multi-Modal Models with Better Captionscited by 1,005
  11. 2023WizardCoder: Empowering Code Large Language Models with Evol-Instructcited by 733
  12. 2024Self-Rewarding Language Modelscited by 681
11 more
  1. 2024DeepSeek-Coder-V2: Breaking the Barrier of Closed-Source Models in Code Intelligencecited by 331
  2. 2025TTRL: Test-Time Reinforcement Learningcited by 275
  3. 2022jTrans: Jump-Aware Transformer for Binary Code Similaritycited by 44
  4. 2024LLM4Decompile: Decompiling Binary Code with Large Language Modelscited by 39
  5. 2023How Far Have We Gone in Vulnerability Detection Using Large Language Modelscited by 27
  6. 2023Extending Source Code Pre-Trained Language Models to Summarise Decompiled Binariescited by 26
  7. 2024CLAP: Learning Transferable Binary Code Representations with Natural Language Supervisioncited by 17
  8. 2024CodeArt: Better Code Models by Attention Regularization When Symbols Are Lackingcited by 7
  9. 2024Disassembling Obfuscated Executables with LLMcited by 7
  10. 2025ASMA-Tune: Unlocking LLMs' Assembly Code Comprehension via Structural-Semantic Instruction Tuningcited by 4
  11. 2025Deconstructing Obfuscation: A four-dimensional framework for evaluating Large Language Models assembly code deobfuscation capabilitiescited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026OctoBench: Benchmarking Scaffold-Aware Instruction Following in Repository-Grounded Agentic Coding
  2. 2025PACIFIC: a framework for generating benchmarks to check Precise Automatically Checked Instruction Following In Code
  3. 2026ACE: Self-Evolving LLM Coding Framework via Adversarial Unit Test Generation and Preference Optimization
  4. 2026Can LLMs Deobfuscate Binary Code? A Systematic Analysis of Large Language Models into Pseudocode Deobfuscation ↗
  5. 2024SelfCodeAlign: Self-Alignment for Code Generation
  6. 2026Steerable Instruction Following Coding Data Synthesis with Actor-Parametric Schema Co-Evolution
  7. 2025When Forgetting Builds Reliability: LLM Unlearning for Reliable Hardware Code Generation
  8. 2026Efficiency Hallucination: Formalizing and Measuring Behavioral Calibration in LLM-Based Code Optimization