Sanjay Rawat

Sanjay Rawat

Vulnerability discovery on C/C++ source asks the analyst to choose between heavyweight static analysers, which need a working build before a single query runs, and free-form LLMs, which read source readily but invent details and lose track of cross-function dataflow on real codebases. We present NeuroLog, an end-to-end build-free pipeline that assigns each layer the role it is uniquely good at: an LLM extracts typed dataflow facts one function at a time; a Souffle rule mesh composes those facts into cross-function findings; a Z3 post-pass filters infeasible findings and emits a SAT model for each survivor. To go beyond pure static reasoning we also fold in runtime evidence: likely range invariants from a handful of corpus seeds tighten the SMT problem at near-zero cost. A second LLM agent reads each SAT model and writes a Python program that produces a candidate crashing input, validated by an AddressSanitizer harness. Combining static-narrowing-SMT (Saturn, Pinpoint) and Datalog-with-SMT (Formulog) is prior art; new here are an LLM-derived fact base, a no-build pipeline, and the SAT model as an artifact (input to crash synthesis) rather than a yes/no verdict. Across stb, cJSON, libxml2, an FFmpeg demuxer slice, and curl 8.3.0, NeuroLog re-discovers eight CVE-class issues end-to-end, including the CVSS-9.8 SOCKS5 heap overflow CVE-2023-38545, each ASan-confirmed. On libarchive HEAD we surface five memory-safety bugs (four previously unreported) across the cpio reader and the XAR/WARC/7zip writers; all filed upstream, several fixes merged, with the cpio use-after-free acknowledged in seven hours. Extraction takes ~37 s and $0.005 on stb; crash synthesis turned a static finding into a 102-byte stb_vorbis crash in two LLM iterations (no fuzzer); a likely-invariant filter from three Matroska seeds eliminates 13.2% of the FFmpeg-demuxer feasible set.

Parteek Jamwal, Minghao Shao, Boyuan Chen et al.

Large Language Models (LLMs) have demonstrated remarkable capabilities across various cybersecurity tasks, including vulnerability classification, detection, and patching. However, their potential in automated vulnerability report documentation and analysis remains underexplored. We present RAVEN (Retrieval Augmented Vulnerability Exploration Network), a framework leveraging LLM agents and Retrieval Augmented Generation (RAG) to synthesize comprehensive vulnerability analysis reports. Given vulnerable source code, RAVEN generates reports following the Google Project Zero Root Cause Analysis template. The framework uses four modules: an Explorer agent for vulnerability identification, a RAG engine retrieving relevant knowledge from curated databases including Google Project Zero reports and CWE entries, an Analyst agent for impact and exploitation assessment, and a Reporter agent for structured report generation. To ensure quality, RAVEN includes a task specific LLM Judge evaluating reports across structural integrity, ground truth alignment, code reasoning quality, and remediation quality. We evaluate RAVEN on 105 vulnerable code samples covering 15 CWE types from the NIST-SARD dataset. Results show an average quality score of 54.21%, supporting the effectiveness of our approach for automated vulnerability documentation.

Sanjay Rawat, Dumitru Ceara, Laurent Mounier et al.

In this paper, we present a hybrid approach for buffer overflow detection in C code. The approach makes use of static and dynamic analysis of the application under investigation. The static part consists in calculating taint dependency sequences (TDS) between user controlled inputs and vulnerable statements. This process is akin to program slice of interest to calculate tainted data- and control-flow path which exhibits the dependence between tainted program inputs and vulnerable statements in the code. The dynamic part consists of executing the program along TDSs to trigger the vulnerability by generating suitable inputs. We use genetic algorithm to generate inputs. We propose a fitness function that approximates the program behavior (control flow) based on the frequencies of the statements along TDSs. This runtime aspect makes the approach faster and accurate. We provide experimental results on the Verisec benchmark to validate our approach.