# Needles at Scale: LLM-Assisted Target Selection for Windows Vulnerability Research

arXiv: https://arxiv.org/abs/2606.01364 · HTML: https://mlsanity.com/paper/2606.01364/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 6 of its 13 references that are in the map.
- First posted on arXiv 31 May 2026.

## Builds on (most cited first, 6 of 6)

- [LLM Agents can Autonomously Exploit One-day Vulnerabilities](https://mlsanity.com/paper/2404.08144/) (2024)
- [Large Language Model for Vulnerability Detection and Repair: Literature Review and the Road Ahead](https://mlsanity.com/paper/2404.02525/) (2024)
- [SecVulEval: Benchmarking LLMs for Real-World C/C++ Vulnerability Detection](https://mlsanity.com/paper/2505.19828/) (2025)
- [Prompting the Priorities: A First Look at Evaluating LLMs for Vulnerability Triage and Prioritization](https://mlsanity.com/paper/2510.18508/) (2025)
- [Machine Learning for Actionable Warning Identification: A Comprehensive Survey](https://mlsanity.com/paper/2312.00324/) (2023)
- [Sift or Get Off the PoC: Applying Information Retrieval to Vulnerability Research with SiftRank](https://mlsanity.com/paper/2512.06155/) (2025)

## Abstract (arXiv)

The attack surface of a modern operating system is a haystack: thousands of signed binaries and millions of functions, almost none relevant to any given vulnerability. A human analyst or an LLM agent must pick the function worth reading before analyzing it. At whole-OS scope, this target selection, not the analysis, is the binding constraint. We present Symbolicate-Enrich-Sample, a low-cost batch pipeline that turns a corpus of production Windows binaries into a queryable, priority-ranked research queue. We (i) recover function-level symbols for stripped vendor binaries by auto-fetching the public symbol files and joining them to a recovered call graph; (ii) attach cheap, deterministic structural features to each named function and, conditioned on those features, use a low-cost language model to assign a reachability tier, a risk level, a bug-class hypothesis, and a rationale; and (iii) draw diverse, prioritized batches via a priority-weighted importance sampler. The contribution is a selection substrate: the prioritization layer a downstream detector or LLM agent runs on top of. Across a whole Windows image of 7,231,419 functions, the labels are markedly selective, and stacking deterministic filters on them leaves a ~22K-function shortlist: the candidate needles, few enough for a human or agent to work through. We characterize the pipeline's selectivity and its failure modes, describe the methodology, and report aggregate statistics; we withhold the derived dataset for legal and dual-use reasons.