New research on AI and security, selected for relevance to language models, AI agents and cybersecurity. Summaries are the authors’ or publisher’s own words, linked to the original.
Peer-reviewed
Impact of frame tail effect on backoff procedure in modern dense Wi-Fi networks
Will AI scoop your science? Some researchers see a gloomy future
Will AI scoop your science? Some researchers see a gloomy future
Anticipatory intelligence in clinical microbiology
New preprints (not yet peer-reviewed)
ReSI: Recursive Safety Improvement toward Resistant and Resilient AI
Recursive self-improvement, the participation of AI systems in improving their own capabilities, is beginning to move from theoretical prospect to practice, posing both challenges and opportunities for safety alignment. Models evolve through frequent updates, and their safety alignment requires continual adaptation to each new checkpoint.
AgentFly: Scaling Agentic Reinforcement Learning with Unified Resource System
Methods to build LLM agents have evolved from prompt engineering and supervised finetuning to agentic reinforcement learning (agentic RL). However, agentic RL remains bottlenecked by its surrounding systems: agents must interact with heterogeneous environments, such as sandboxes, model services, and external APIs.
Safe at One Loop, Risky at Another: Aligning Safety Across Recurrent Depths in Looped Language Models
Looped Language Models (LoopLMs) provide a parameter efficient approach to scaling model capabilities through repeated use of shared parameters across recurrent steps. Since each recurrent depth can be read out independently, a single LoopLM exposes a broader output space across inference depths, raising an important question: whether safety is preserved throughout recurrent computation.
DuplexAgent-RSI: Recursive Harness Improvement for Full-Duplex Voice Agent Collaboration
Voice agents are converging on a collaboration pattern: a full-duplex interaction model stays on the live channel as the entry to the conversation, while search, reasoning, and coding are handled through asynchronous delegation. A duplex model supports continuous listening and speaking, but complex reasoning and tool use may exceed its capabilities.
What Transfers from Text to Vision? Capability Scaling Laws and Transfer Dynamics for VLMs
Choosing the right large language model (LLM) backbone is the most consequential decision when building a vision-language model (VLM), yet it remains fundamentally unprincipled: compute-based scaling laws fail to generalize across model families, and no framework exists for directly predicting VLM performance before training begins. We propose the Capability-Driven Multimodal Scaling Law, the first cross-family framework that predicts VLM benchmark accuracy from directly observable textual capability.
A Zero-Knowledge Signature Framework for Efficient Post-Quantum Message Authentication in Cooperative Automated Driving
Connected and Automated Vehicles (CAV) rely on authenticated Vehicle-to-Everything (V2X) communications to exchange safety-critical information among vehicles and roadside infrastructure. As the automotive industry transitions toward post-quantum cryptography (PQC), the significantly larger public keys and signatures of standardized PQC digital signature algorithms introduce substantial communication overhead, which challenges the scalability of certificate-based V2X authentication, particularly for high-frequency cooperative awareness messages (CAM).
Predicting Alignment Generalization with Value Representations
LLM developers post-train their models to exhibit prosocial values and behavioral traits, which are enumerated in an alignment target. However, while recent post-training developments have yielded models that score highly on alignment evaluations, training models on sets of narrow behaviors still influences their behavior across unseen contexts and environments in unexpected ways.
ORCAGen: Orchestrating Context-Aware Malware Deception with RAG-Guided Generative AI
Malware defenses often remove or isolate suspicious programs as quickly as possible. While effective for containment, this approach can also waste an opportunity to observe attacker behavior and deploy targeted countermeasures.