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
Hidden signals in sleep can predict future health
Private, offline and yours: how to search the literature with local AI systems
I told ChatGPT to ‘continue’ — it went too far
Award-winning microscopy image disqualified over AI use
ETSAM: effectively segmenting cell membranes in cryo-electron tomograms
New preprints (not yet peer-reviewed)
ReasoningShield: Safety Detection over Reasoning Traces of Large Reasoning Models
Large Reasoning Models (LRMs) leverage explicit reasoning traces, known as Chain-of-Thought (CoT) reasoning, to decompose complex problems into intermediate steps before deriving final answers. However, these reasoning traces introduce unique safety challenges: harmful content can be embedded in intermediate steps even when final answers appear benign.
Let Large Language Models Judge Each Other: Multi-Agent Peer-Reviewed Reasoning for Medical Question Answering
Objective: To enhance the accuracy, interpretability, and robustness of large language models (LLMs) in medical question answering (MedQA). Method: We designed a multi-agent peer-reviewed reasoning method in which multiple LLM agents independently generate chain-of-thought reasoning with candidate answers, then act as peer reviewers to evaluate each other's reasoning for factual correctness and logical soundness.
EviRepair: Evidence-Driven Agents for Reliable Vulnerability Repair
Recent advances in large language models (LLMs) have enabled agent-based approaches for repository-level automated vulnerability repair (AVR), showing promising repair capabilities. However, an important issue has received limited attention: how can we ensure that the evidence generated during the repair process is reliable enough to support accurate vulnerability understanding and patch generation?
Learning More by Training Less: Uncertainty-Guided Progressive Contraction for Efficient Reasoning Post-Training
Large language models have achieved remarkable advances in complex reasoning. However, scaling reasoning post-training with ever-larger corpora can substantially increase training cost without proportional performance gains.
VEPA: Verified Evolving Formal Policy for Lifelong LLM Agent Guardrails
LLM agents increasingly operate over sensitive resources, such as private documents and mutable system states, which can create serious safety and security risks. Runtime guardrails should therefore make inspectable, reproducible, and deterministic allow or refuse decisions before potentially irreversible agent actions are executed.
Do More Agents Help? Evaluating Accuracy and Cost in LLM Multi-Agent Workflows
Choosing whether to replace a single-agent controller requires evidence on both task performance and inference expenditure. We present BenchAgent, a shared evaluation framework, and compare six multi-agent workflow configurations against its single-agent reference, BenchAgent Core, across ten reasoning, coding, and tool-use benchmarks.
Collaborative Learning and Reasoning over Model Slices
Large language models (LLMs) excel across diverse tasks, benefiting from knowledge priors, in-context reasoning, and test-time computation. Despite these strengths, single-model reasoning remains vulnerable to error propagation, as later steps depend on the model's earlier outputs, leaving reasoning blind spots difficult to independently reassess.
SmartSalesBench: A Resistance-Aware Multi-Turn Benchmark for Evaluating LLM Sales Agents
We present SmartSalesBench, the first multi-turn conversational benchmark explicitly designed to evaluate large language model (LLM) sales agents against psychologically grounded, resistant buyer personas. Existing task-oriented dialogue benchmarks assume cooperative users and measure task completion; SmartSalesBench inverts this assumption, placing agents against personas programmed to resist and challenge across conversations of up to twenty turns.