Secure LLM Fine-Tuning via Safety-Aware Probing
arXiv:2505.16737v2 Announce Type: replace-cross Abstract: Large language models (LLMs) have achieved remarkable success across many applications, but their ability to generate harmful content raises s
Knowledge catalogue
arXiv:2505.16737v2 Announce Type: replace-cross Abstract: Large language models (LLMs) have achieved remarkable success across many applications, but their ability to generate harmful content raises s
arXiv:2604.21076v1 Announce Type: cross Abstract: Medication reconciliation at clinical handoffs is a high-stakes, error-prone process. Large language models are increasingly proposed to assist with t
arXiv:2510.11423v3 Announce Type: replace-cross Abstract: Community Notes, the crowd-sourced misinformation governance system on X (formerly Twitter), allows users to flag misleading posts, attach con
arXiv:2604.20658v1 Announce Type: new Abstract: Multi-agent systems built from teams of large language models (LLMs) are increasingly deployed for collaborative scientific reasoning and problem-solvin
arXiv:2604.20168v1 Announce Type: new Abstract: This paper presents the Duluth approach to SemEval-2026 Task 6 on CLARITY: Unmasking Political Question Evasions. We address Task 1 (clarity-level class
arXiv:2511.15141v2 Announce Type: replace-cross Abstract: Recently, large language models (LLMs) have been widely used as recommender systems, owing to their reasoning capability and effectiveness in
arXiv:2604.20193v1 Announce Type: new Abstract: Ensuring functional safety in human-robot interaction is challenging because AI perception is inherently probabilistic, whereas industrial standards req
arXiv:2601.12078v2 Announce Type: replace Abstract: Large language models (LLMs) excel at general-purpose tasks, yet adapting their responses to individual users remains challenging. Retrieval augment
arXiv:2604.20316v1 Announce Type: new Abstract: Function calling empowers large language models (LLMs) to interface with external tools, yet existing RL-based approaches suffer from misalignment betwe
arXiv:2604.19825v1 Announce Type: cross Abstract: State-of-the-art code generation frameworks rely on mental simulation, where LLMs internally trace execution to verify correctness. We expose a fundam
This blog post discusses higher-order optimization algorithms such as Shampoo that have been applied in neural network training for at least a decade. The article explores how these emerging optimizer
arXiv:2604.18963v1 Announce Type: cross Abstract: Knowledge distillation (KD) transfers capabilities from large language models (LLMs) to smaller students, yet it can fail unpredictably and also under
arXiv:2604.18788v1 Announce Type: new Abstract: Apple Neural Engine (ANE) is a dedicated neural processing unit (NPU) present in every Apple Silicon chip. Mixture-of-Experts (MoE) LLMs improve inferen
arXiv:2604.19331v1 Announce Type: new Abstract: Understanding how policy is debated and justified in parliament is a fundamental aspect of the democratic process. However, the volume and complexity of
arXiv:2604.19526v1 Announce Type: cross Abstract: Cross-site scripting (XSS) remains a persistent web security vulnerability, especially because obfuscation can change the surface form of a malicious
arXiv:2604.19274v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used as co-authors in collaborative writing, where users begin with rough drafts and rely on LLMs to compl
arXiv:2601.18572v2 Announce Type: replace Abstract: Personalization of LLMs by sociodemographic subgroup often improves user experience, but can also introduce or amplify biases and unfair outcomes ac
arXiv:2604.18976v1 Announce Type: new Abstract: While Large Language Models (LLMs) are widely used, they remain susceptible to jailbreak prompts that can elicit harmful or inappropriate responses. Thi
arXiv:2601.09953v2 Announce Type: replace Abstract: Standardized math assessments require expensive human pilot studies to establish the difficulty of test items. We investigate the predictive value o
arXiv:2604.19221v1 Announce Type: new Abstract: Full-duplex speech interaction, as the most natural and intuitive mode of human communication, is driving artificial intelligence toward more human-like
arXiv:2604.16622v1 Announce Type: new Abstract: Backchannels (e.g., `yeah', `mhm', and `right') are short, non-interruptive feedback signals whose lexical form and prosody jointly convey pragmatic mea
arXiv:2604.17008v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly used to generate narrative content, including children's stories, which play an important role in social a
arXiv:2601.11038v2 Announce Type: replace Abstract: We study the reasoning behavior of large language models (LLMs) under limited computation budgets. In such settings, producing useful partial soluti
arXiv:2604.17316v1 Announce Type: new Abstract: Safe clinical deployment of Large Language Models (LLMs) requires not only high accuracy but also robust uncertainty calibration to ensure models defer
arXiv:2604.18134v1 Announce Type: new Abstract: Recent advancements in self-supervised learning have led to powerful surgical vision encoders capable of spatiotemporal understanding. However, extendin
arXiv:2506.01732v2 Announce Type: replace Abstract: Large Language Models (LLMs) are pre-trained on large data from different sources and domains. These datasets often contain trillions of tokens, inc
arXiv:2507.06056v4 Announce Type: replace Abstract: Large Language Models (LLMs) are known to memorize portions of their training data, sometimes even reproduce content verbatim when prompted appropri
arXiv:2604.17191v1 Announce Type: new Abstract: Multi-agent reinforcement learning (MARL) is crucial for AI systems that operate collaboratively in distributed and adversarial settings, particularly i
arXiv:2601.04278v2 Announce Type: replace Abstract: Although machine unlearning is essential for removing private, harmful, or copyrighted content from LLMs, current benchmarks often fail to faithfull
arXiv:2602.17550v3 Announce Type: replace Abstract: Existing Reinforcement Learning with Verifiable Rewards (RLVR) algorithms, such as GRPO, rely on rigid, uniform, and symmetric trust region mechanis
arXiv:2604.16493v1 Announce Type: cross Abstract: Natural Language to SQL (NL2SQL) technology empowers non-expert users to query relational databases without requiring SQL expertise. While large langu
arXiv:2604.18085v1 Announce Type: new Abstract: Matrix-level low-rank compression is a promising way to reduce the cost of large language models, but running compression and evaluating the resulting m
arXiv:2604.17396v1 Announce Type: new Abstract: Large Language Models (LLMs) often memorize sensitive or harmful information, necessitating effective machine unlearning techniques. While existing para
arXiv:2604.17659v1 Announce Type: new Abstract: We introduce the Semantic Density Effect (SDE): the empirical finding that prompts carrying higher semantic information per token consistently produce m
arXiv:2601.04043v2 Announce Type: replace Abstract: As Multimodal Large Language Models (MLLMs) become an indispensable assistant in human life, the unsafe content generated by MLLMs poses a danger to
arXiv:2604.16258v1 Announce Type: new Abstract: Competency Questions (CQs) are a cornerstone of requirement elicitation in ontology engineering. CQs represent requirements as a set of natural language
arXiv:2601.05808v2 Announce Type: replace-cross Abstract: Large language models (LLMs) are expected to be trained to act as agents in various real-world environments, but this process relies on rich a
arXiv:2604.15588v1 Announce Type: cross Abstract: The integration of Large Language Models (LLMs) into scientific workflows presents exciting opportunities to accelerate biomedical discovery. However,
arXiv:2604.15672v1 Announce Type: cross Abstract: Speculative decoding (SD) accelerates language model inference by drafting tokens from a cheap proposal model and verifying them against an expensive
arXiv:2604.15589v1 Announce Type: cross Abstract: Existing research on large language models (LLMs) for automated code compliance has primarily focused on performance, treating the models as black box
arXiv:2509.17292v3 Announce Type: replace-cross Abstract: Cognitive distortions have been closely linked to mental health disorders, yet their automatic detection remains challenging due to contextual
arXiv:2604.15937v1 Announce Type: cross Abstract: Large Language Models (LLMs) are increasingly deployed to curate and rank human-created content, yet the nature and structure of their biases in these
arXiv:2604.15702v1 Announce Type: new Abstract: We introduce a cross-domain behavioural assay of monitoring-control coupling in LLMs, grounded in the Nelson and Narens (1990) metacognitive framework a
arXiv:2601.03746v3 Announce Type: replace Abstract: As large language models (LLMs) are more frequently used in retrieval-augmented generation pipelines, it is increasingly relevant to study their beh
arXiv:2603.07053v2 Announce Type: replace Abstract: Scientists face significant visualization challenges as time-varying datasets grow in speed and volume, often requiring specialized infrastructure a
arXiv:2604.14892v1 Announce Type: new Abstract: Evaluating medical AI systems using expert clinician panels is costly and slow, motivating the use of large language models (LLMs) as alternative adjudi
arXiv:2604.14644v1 Announce Type: new Abstract: The inability to filter out in advance all potentially problematic data from the pre-training of large language models has given rise to the need for me
arXiv:2604.14267v1 Announce Type: new Abstract: Search agents extend Large Language Models (LLMs) beyond static parametric knowledge by enabling access to up-to-date and long-tail information unavaila
arXiv:2604.14843v1 Announce Type: new Abstract: Behavioral Profile (BP) annotation is difficult to automate because it requires simultaneous coding across multiple linguistic dimensions. We treat BP a
arXiv:2604.14325v1 Announce Type: new Abstract: Large language models (LLMs) achieve strong performance and have revolutionized NLP, but their lack of explainability keeps them treated as black boxes,
arXiv:2604.14199v1 Announce Type: cross Abstract: Predicting real-world events from live market signals demands systems that fuse qualitative news with quantitative order-book dynamics under strict te
arXiv:2604.15022v1 Announce Type: cross Abstract: Cost-aware routing dynamically dispatches user queries to models of varying capability to balance performance and inference cost. However, the routing
arXiv:2604.13041v1 Announce Type: cross Abstract: Table Structure Recognition (TSR) requires the logical reasoning ability of large language models (LLMs) to handle complex table layouts, but current
arXiv:2604.13328v1 Announce Type: new Abstract: Pathology reports serve as the definitive record for breast cancer staging, yet their unstructured format impedes large-scale data curation. While Large
arXiv:2506.08125v3 Announce Type: replace-cross Abstract: Large language models (LLMs) show strong reasoning abilities but often produce unnecessarily long explanations that reduce efficiency. Althoug
arXiv:2604.05096v2 Announce Type: replace Abstract: Large language models (LLMs) acquire most of their knowledge during pretraining, which ties them to a fixed snapshot of the world and makes adaptati
arXiv:2601.07422v2 Announce Type: replace Abstract: Despite their impressive capabilities, large language models (LLMs) frequently generate hallucinations. Previous work shows that their internal stat
arXiv:2604.12898v1 Announce Type: new Abstract: Large Language Model-based Hyper Heuristic (LHH) has recently emerged as an efficient way for automatic heuristic design. However, most existing LHHs ju
arXiv:2507.21990v4 Announce Type: replace-cross Abstract: Atomized chemical knowledge, such as functional group information of molecules and reactions, plays a pivotal intermediate role in the reasoni
arXiv:2602.13851v3 Announce Type: replace-cross Abstract: Formal specifications are crucial for building verifiable and dependable software systems, yet generating accurate and verifiable specificatio