Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2511. 11927v2 Announce Type: replace-cross Abstract: Principal Component Analysis (PCA) is a standard tool for extracting a low-rank signal from noisy observations.
arXiv:2608. 09389v1 Announce Type: cross Abstract: This note aims to serve as an entry point to the literature on learning in games, a topic with significant theoretical appeal and a wide range of applications -- from machine learning and data science to economics and beyond.
arXiv:2608. 08542v1 Announce Type: new Abstract: Model merging has become the default way to give an aligned language model new skills without retraining: a practitioner folds task vectors from math, code, or domain specialists into a safety-aligned base using task arithmetic, TIES, or DARE.
arXiv:2608. 08536v1 Announce Type: new Abstract: Circuit localization is a mechanistic interpretability task whose goal is to identify a sparse subgraph of a transformer's computation graph sufficient to reproduce a particular behavior.
arXiv:2608. 09225v1 Announce Type: cross Abstract: The key-value (KV) cache is the primary throughput optimization in modern large language model (LLM) inference, enabling prefix reuse across requests.
arXiv:2608. 08202v1 Announce Type: new Abstract: Data-centric curation pipelines frequently rely on model confidence scores to flag and filter noisy or mislabeled training instances.
arXiv:2608. 08224v1 Announce Type: new Abstract: Reinforcement learning post-training unlocks complex reasoning in LLMs.
arXiv:2404. 03082v3 Announce Type: replace Abstract: Partially ordered sets (posets) are discrete mathematical structures that formalize the notion of comparison without forcing every pair of objects to be comparable.
arXiv:2608. 08008v1 Announce Type: new Abstract: Process reward models (PRMs) score intermediate reasoning steps and are widely used for search, ranking, and training, but optimization can exploit these learned proxies by increasing reward while turning correct reasoning into incorrect reasoning.
arXiv:2608. 07725v1 Announce Type: new Abstract: Average-reward reinforcement-learning regret is known up to logarithmic factors, but the numerical content of published guarantees is difficult to compare because probability mode, structural parameter, logarithmic normalization, prior information, and planning assumptions differ.
arXiv:2605. 00868v3 Announce Type: replace-cross Abstract: Ultra-wide bandgap (UWBG) Ga$_2$O$_3$ is a promising semiconductor for high-power and high-temperature electronics.
arXiv:2601. 20819v2 Announce Type: replace-cross Abstract: Machine learning predictions are increasingly used to supplement incomplete or costly-to-measure outcomes in fields such as biomedical research, environmental science, and social science.
arXiv:2604. 26494v2 Announce Type: replace-cross Abstract: Generative AI tools are widely used by youth and have introduced new privacy and safety challenges.
arXiv:2604. 23341v3 Announce Type: replace-cross Abstract: The deployment of Large Language Models (LLMs) as assistants in electric grid operations promises to streamline compliance and decision-making but exposes new vulnerabilities to prompt-based adversarial attacks.
arXiv:2604. 07223v2 Announce Type: replace-cross Abstract: As large language models (LLMs) evolve from static chatbots into autonomous agents, the primary vulnerability surface shifts from final outputs to intermediate execution traces.
arXiv:2601. 21747v4 Announce Type: replace-cross Abstract: Sepsis remains one of the most complex and heterogeneous syndromes in intensive care.
arXiv:2608. 08199v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly involved in group decision-making with other LLMs and humans.
arXiv:2601. 03100v3 Announce Type: replace-cross Abstract: Multimodal large language models (MLLMs) typically rely on a single late-layer feature from a frozen vision encoder, leaving the encoder's rich hierarchy of visual cues under-utilized.
arXiv:2608. 07881v1 Announce Type: new Abstract: Clustering mixed tabular data requires a unified metric space to bridge the inherent heterogeneity between continuous numerical measurements and discrete categorical symbols.
arXiv:2510. 04465v3 Announce Type: replace-cross Abstract: LLM agents require personal information for personalization in order to effectively act on users' behalf, but this raises privacy concerns that can discourage data sharing, limiting both the autonomy levels at which agents can operate and the effectiveness of personalization.