Related papers: Teaching LLMs Program Semantics via Symbolic Execu…
Several closed-source LLMs have consistently outperformed open-source alternatives in program repair tasks, primarily due to their superior reasoning capabilities and extensive pre-training. This paper introduces Repairity, a novel…
Large language models (LLMs) can perform reasoning computations both internally within their latent space and externally by generating explicit token sequences like chains of thought. Significant progress in enhancing reasoning abilities…
As large language models (LLMs) evolve from static chatbots into autonomous agents, the primary vulnerability surface shifts from final outputs to intermediate execution traces. While safety guardrails are well-benchmarked for natural…
Chain-of-thought reasoning is often treated as a monotone way to improve language-model accuracy by letting a model think longer. We identify a countervailing effect, the coupling tax: when reasoning traces and final answers share one…
Language models frequently produce plausible yet incorrect reasoning traces that are difficult to verify. We investigate fine-tuning models to use Prolog as an external symbolic reasoning tool, training Qwen2.5-3B-Instruct with Group…
Trusted Execution Environments (TEEs) provide hardware-enforced isolation that protects sensitive code and data from untrusted software. Despite their strong security guarantees, analyzing TEE applications remains challenging due to the…
Symbolic execution is a successful and very popular technique used in software verification and testing. A key limitation of symbolic execution is in dealing with code containing loops. The problem is that even a single loop can generate a…
Code-driven auditing fails when correctness depends on what the specification requires rather than how the code is written. Production blockchain networks expose this directly: byzantine consensus runs many independent clients of a shared…
Recent studies demonstrate that tool-calling capability enables large language models (LLMs) to interact with external environments for long-horizon financial tasks. While existing benchmarks have begun evaluating financial tool calling,…
Training on verifiable symbolic data is a promising way to expand the reasoning frontier of language models beyond what standard pre-training corpora provide. Yet existing procedural generators often rely on fixed puzzles or templates and…
Binary vulnerability analysis is increasingly performed by LLM-based agents in an iterative, multi-pass manner, with the model as the core decision-maker. However, how such systems organize exploration over hundreds of reasoning steps…
Self-evolving scientific agents capable of conquering the hard tail of formal mathematics require Compositional Learning Behaviours (CLBs) -- the capacity to ground and recombine novel symbolic structures in context, beyond mere…
Cyber-physical systems (CPS) development requires verifying whether system behaviors violate their requirements. This analysis often considers system behaviors expressed by execution traces and requirements expressed by signal-based…
Large Language Models (LLMs) have demonstrated significant potential in automated software security, particularly in vulnerability detection. However, existing benchmarks primarily focus on isolated, single-vulnerability samples or…
Large language models (LLMs) increasingly generate code with minimal human oversight, raising critical concerns about backdoor injection and malicious behavior. We present Cross-Trace Verification Protocol (CTVP), a novel AI control…
Instructional alignment, the match between intended cognition and enacted activity, is central to effective instruction but hard to operationalize at scale. We examine alignment in cybersecurity simulations using multimodal traces from 23…
Security Operations Centers (SOCs) face mounting operational challenges. These challenges come from increasing threat volumes, heterogeneous SIEM platforms, and time-consuming manual triage workflows. We present an end-to-end threat…
As large language models (LLMs) become increasingly capable and widely adopted, benchmarks play a central role in assessing their practical utility. For example, SWE-Bench Verified has emerged as a critical benchmark for evaluating LLMs'…
Large language models (LLMs) are increasingly applied in computer science education for tasks such as tutoring, content generation, and code assessment. However, systematic evaluations aligned with formal curricula and certification…
As large language models are deployed as autonomous agents with tool execution privileges, a critical assumption underpins their security architecture: that model errors are detectable at runtime. We present empirical evidence that this…