15 citations · 16 across the 20 of their papers we have counts for
23 papers
Vulnerability Localization Benchmark: Measuring Agentic Security Analysis at Repository Scale
Aman Priyanshu, Supriti Vijay, Kimia Majd +8
Language-model agents increasingly operate over complete software repositories, yet cybersecurity evaluations primarily measure whether they can detect, reproduce, or repair vulner…
Antares: Foundation Models for Agentic Vulnerability Localization
Supriti Vijay, Aman Priyanshu, Didier Chapoteau +8
Vulnerability localization is a fundamental step in software security, requiring models to reason over large codebases and iteratively identify vulnerable implementations. We prese…
FAPO: Fully Automated Prompt Optimization of Multi-Step LLM Pipelines
Paul Kassianik, Baturay Saglam, Huaibo Zhao +4
Multi-step LLM pipelines fail through interactions among retrieval, reasoning, and formatting steps, so prompt-only optimization can miss bottlenecks in the chain. We present Fully…
Self-Improving In-Context Learning
Baturay Saglam, Dionysis Kalogerias
We propose to improve in-context learning (ICL) by optimizing the continuous embeddings of a fixed few-shot prompt at test time. The key observation is that the log-probabilities a…
Test-Time Safety Alignment
Baturay Saglam, Dionysis Kalogerias
Recent work has shown that a model's input word embeddings can serve as effective control variables for steering its behavior toward outputs that satisfy desired properties. Howeve…
Test-Time Detoxification without Training or Learning Anything
Baturay Saglam, Dionysis Kalogerias
Large language models can produce toxic or inappropriate text even for benign inputs, creating risks when deployed at scale. Detoxification is therefore important for safety and us…
Llama-3.1-FoundationAI-SecurityLLM-Reasoning-8B Technical Report
Zhuoran Yang, Ed Li, Jianliang He +18
We present Foundation-Sec-8B-Reasoning, the first open-source native reasoning model for cybersecurity. Built upon our previously released Foundation-Sec-8B base model (derived fro…
Risk-Averse Constrained Reinforcement Learning with Optimized Certainty Equivalents
Jane H. Lee, Baturay Saglam, Spyridon Pougkakiotis +2
Constrained optimization provides a common framework for dealing with conflicting objectives in reinforcement learning (RL). In most of these settings, the objectives (and constrai…
Think Before You Retrieve: Learning Test-Time Adaptive Search with Small Language Models
Supriti Vijay, Aman Priyanshu, Anu Vellore +2
Effective information retrieval requires reasoning over partial evidence and refining strategies as information emerges. Yet current approaches fall short: neural retrievers lack r…
Llama-3.1-FoundationAI-SecurityLLM-8B-Instruct Technical Report
Sajana Weerawardhena, Paul Kassianik, Blaine Nelson +14
Large language models (LLMs) have shown remarkable success across many domains, yet their integration into cybersecurity applications remains limited due to a lack of general-purpo…
Large Language Models Encode Semantics and Alignment in Linearly Separable Representations
Baturay Saglam, Paul Kassianik, Blaine Nelson +3
Understanding the latent space geometry of large language models (LLMs) is key to interpreting their behavior and improving alignment. Yet it remains unclear to what extent LLMs li…
Llama-3.1-FoundationAI-SecurityLLM-Base-8B Technical Report
Paul Kassianik, Baturay Saglam, Alexander Chen +15
As transformer-based large language models (LLMs) increasingly permeate society, they have revolutionized domains such as software engineering, creative writing, and digital arts.…
Learning Task Representations from In-Context Learning
Baturay Saglam, Xinyang Hu, Zhuoran Yang +2
Large language models (LLMs) have demonstrated remarkable proficiency in in-context learning (ICL), where models adapt to new tasks through example-based prompts without requiring…
Compatible Gradient Approximations for Actor-Critic Algorithms
Baturay Saglam, Dionysis Kalogerias
Deterministic policy gradient algorithms are foundational for actor-critic methods in controlling continuous systems, yet they often encounter inaccuracies due to their dependence…
Deep Reinforcement Learning Based Joint Downlink Beamforming and RIS Configuration in RIS-aided MU-MISO Systems Under Hardware Impairments and Imperfect CSI
Baturay Saglam, Doga Gurgunoglu, Suleyman S. Kozat
We introduce a novel deep reinforcement learning (DRL) approach to jointly optimize transmit beamforming and reconfigurable intelligent surface (RIS) phase shifts in a multiuser mu…
Deep Intrinsically Motivated Exploration in Continuous Control
Baturay Saglam, Suleyman S. Kozat
In continuous control, exploration is often performed through undirected strategies in which parameters of the networks or selected actions are perturbed by random noise. Although…
Actor Prioritized Experience Replay
Baturay Saglam, Furkan B. Mutlu, Dogan C. Cicek +1
A widely-studied deep reinforcement learning (RL) technique known as Prioritized Experience Replay (PER) allows agents to learn from transitions sampled with non-uniform probabilit…
Mitigating Off-Policy Bias in Actor-Critic Methods with One-Step Q-learning: A Novel Correction Approach
Baturay Saglam, Dogan C. Cicek, Furkan B. Mutlu +1
Compared to on-policy counterparts, off-policy model-free deep reinforcement learning can improve data efficiency by repeatedly using the previously gathered data. However, off-pol…
Safe and Robust Experience Sharing for Deterministic Policy Gradient Algorithms
Baturay Saglam, Dogan C. Cicek, Furkan B. Mutlu +1
Learning in high dimensional continuous tasks is challenging, mainly when the experience replay memory is very limited. We introduce a simple yet effective experience sharing mecha…
AWD3: Dynamic Reduction of the Estimation Bias
Dogan C. Cicek, Enes Duran, Baturay Saglam +3
Value-based deep Reinforcement Learning (RL) algorithms suffer from the estimation bias primarily caused by function approximation and temporal difference (TD) learning. This probl…
Off-Policy Correction for Deep Deterministic Policy Gradient Algorithms via Batch Prioritized Experience Replay
Dogan C. Cicek, Enes Duran, Baturay Saglam +2
The experience replay mechanism allows agents to use the experiences multiple times. In prior works, the sampling probability of the transitions was adjusted according to their imp…
Estimation Error Correction in Deep Reinforcement Learning for Deterministic Actor-Critic Methods
Baturay Saglam, Enes Duran, Dogan C. Cicek +2
In value-based deep reinforcement learning methods, approximation of value functions induces overestimation bias and leads to suboptimal policies. We show that in deep actor-critic…
Parameter-free Reduction of the Estimation Bias in Deep Reinforcement Learning for Deterministic Policy Gradients
Baturay Saglam, Furkan Burak Mutlu, Dogan Can Cicek +1
Approximation of the value functions in value-based deep reinforcement learning induces overestimation bias, resulting in suboptimal policies. We show that when the reinforcement s…