arXiv AI August 19, 2026 By Alizishaan Khatri

Probing the Prefill: Detecting Code Vulnerabilities via Latent Activations

arXiv:2608. 16970v1 Announce Type: cross Abstract: LLM-based code generation is now embedded in mission-critical pipelines, but defenses against vulnerable output remain post-hoc -- static analyzers, fine-tuned classifiers, or an LLM judge that screen completed code, ignoring the generating model's own internal state.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv AI.

arXiv AI

Jun 19

Calibration Without Comprehension: Diagnosing the Limits of Fine-Tuning LLMs for Vulnerability Detection in Systems Software

arXiv:2606. 20502v1 Announce Type: cross Abstract: Whether LLMs scoring well on vulnerability benchmarks genuinely reason about security or merely pattern-match on contaminated data remains unresolved.

By Arastoo Zibaeirad, Marco Vieira

llms fine-tuning benchmarks

arXiv AI

Aug 11

Do All LLMs Know When They're Being Harmful? A Reproducibility Study of Latent-Space Safety Probes Across Model Families

arXiv:2608. 08029v1 Announce Type: cross Abstract: Khatri et al.

By Alizishaan Khatri, Dun Li Chan

llms benchmarks safety

arXiv AI

Jun 6

Willing but Unable: Separating Refusal from Capability in Code LLMs via Abliteration

arXiv:2606. 05396v1 Announce Type: cross Abstract: Producing a labeled vulnerable code at scale is a recurring obstacle for learning-based vulnerability detection: mined corpora carry substantial label noise, and existing LLM-based augmentation propagates these inaccuracies because it transforms vulnerable seeds rather than synthesising vulnerabilities from a specification.

By Cristina Carleo, Pietro Liguori, Naghmeh Ivaki, Domenico Cotroneo

llms reinforcement-learning safety data-engineering

arXiv AI

Jun 2

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

arXiv:2606. 01364v1 Announce Type: cross Abstract: 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.

By Michael J. Bommarito II

llms agents

arXiv AI

Jun 30

Closing the Activation-Cone Blind Spot: Response-Time Probing and Unified Defense

arXiv:2606. 29441v1 Announce Type: cross Abstract: Inference-time safety methods for large language models have proliferated, yet no systematic comparison exists.

By Subhadip Mitra

llms benchmarks safety

Hugging Face Trending Papers

Jun 28

Closing the Activation-Cone Blind Spot: Response-Time Probing and Unified Defense

Inference-time safety methods for large language models have proliferated, yet no systematic comparison exists. We evaluate five defense paradigms (no defense, static steering, CAST, AlphaSteer, probe-gated) across seven instruction-tuned models (7-31B) and five attack types (GCG, AutoDAN, DeepInception, prefilling, intent laundering).

llms benchmarks safety

Related stories

Calibration Without Comprehension: Diagnosing the Limits of Fine-Tuning LLMs for Vulnerability Detection in Systems Software

Do All LLMs Know When They're Being Harmful? A Reproducibility Study of Latent-Space Safety Probes Across Model Families

Willing but Unable: Separating Refusal from Capability in Code LLMs via Abliteration

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

Closing the Activation-Cone Blind Spot: Response-Time Probing and Unified Defense

Closing the Activation-Cone Blind Spot: Response-Time Probing and Unified Defense