Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2603. 21014v2 Announce Type: replace Abstract: Mechanistic interpretability seeks to understand how Large Language Models (LLMs) represent and process information.
arXiv:2607. 18360v1 Announce Type: cross Abstract: Large language models (LLMs) now routinely draft literature reviews and assist with academic writing, which means a higher risk of fabricated references: GPTZero found 53 papers with hallucinated citations among NeurIPS 2025's accepted set.
arXiv:2607. 18485v1 Announce Type: cross Abstract: Large language model (LLM) agents are starting to take on routine work in high-performance computing (HPC), including monitoring Slurm jobs, diagnosing failed builds, inspecting simulation output, and coordinating scientific workflows.
arXiv:2603. 24917v2 Announce Type: replace-cross Abstract: Recent work shows that standard greedy-decoding extraction methods for quantifying memorization in LLMs miss how extraction risk varies across sequences.
arXiv:2607. 18828v1 Announce Type: new Abstract: Readiness stress-testing of medical AI has focused on closed-ended and multimodal benchmarks.
arXiv:2607. 18516v1 Announce Type: new Abstract: We introduce Signed Rectified Flow (Signed RF), a generalization of Rectified Flow that targets the signed measure $\pi^{sign} = (1+\alpha)\pi^+ - \alpha\pi^-$, where $\alpha>0$, $\pi^+$ is the distribution to promote, and $\pi^-$ is the distribution to suppress.
arXiv:2607. 19004v1 Announce Type: cross Abstract: We provide a simple and tight characterization of the types of inexact score oracle access that permit sampling with vanishing total variation bias, for a standard, well-behaved target family.
arXiv:2607. 18292v1 Announce Type: cross Abstract: As language models scale, answers start truer but degrade faster: scaling buys capability but erodes reliability.
arXiv:2508. 05157v2 Announce Type: replace Abstract: Federated Learning (FL) enables collaborative training across distributed clients without sharing raw data, offering strong privacy benefits.
arXiv:2607. 19061v1 Announce Type: cross Abstract: Hateful optical illusions expose a serious gap in current multimodal safety systems.
arXiv:2607. 18689v1 Announce Type: new Abstract: Reinforcement learning (RL) methods for learning-to-rank (LTR) can optimize (almost) any ranking goal, e.
arXiv:2607. 18245v1 Announce Type: new Abstract: Exact-match evaluation of agent-calling obscures qualitatively different failure modes: a model may select the right function yet hallucinate argument values, or satisfy a schema while choosing a agent for the wrong reason.
arXiv:2607. 18422v1 Announce Type: new Abstract: Overparameterized models often have continuous parameter symmetries, so different parameters define the same predictor.
arXiv:2604. 03245v2 Announce Type: replace-cross Abstract: The remarkable reasoning and code generation capabilities of large language models (LLMs) have recently motivated increasing interest in automating formal verification (FV), a process that ensures hardware correctness through mathematically precise assertions but remains highly labor-intensive, particularly through the translation of natural language into SystemVerilog Assertions (NL-to-SVA).
arXiv:2607. 19317v1 Announce Type: new Abstract: Circuit analysis can support not only model explanation but also downstream interventions such as pruning, editing, steering, and selective fine-tuning.
arXiv:2607. 18536v1 Announce Type: new Abstract: Macro placement still requires substantial manual refinement in industrial physical design flows.
arXiv:2607. 19096v1 Announce Type: new Abstract: Agent-memory workloads mix direct factual lookup, relation-chain and current-state reasoning, and broad synthesis over long histories.
arXiv:2607. 18258v1 Announce Type: new Abstract: Reinforcement learning from human feedback (RLHF) with preference-based reward models often exhibits unstable training dynamics.
arXiv:2607. 18665v1 Announce Type: new Abstract: Large language models (LLMs) increasingly support science, but they can also convert hazardous scientific knowledge into actionable misuse guidance.
arXiv:2607. 18291v1 Announce Type: cross Abstract: Time-dependent reliability analysis is crucial for ensuring the long-term safety and performance of engineering systems under uncertainties.