Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2602. 16111v2 Announce Type: replace-cross Abstract: Online media platforms track the share of impressions associated with content attributes, or prevalence, to evaluate trade-offs and set guardrails in A/B experiments.
arXiv:2603. 15684v2 Announce Type: replace-cross Abstract: Safety alignment in large language models is typically evaluated under isolated queries, yet real-world use is inherently multi-turn.
arXiv:2607. 27594v1 Announce Type: new Abstract: Post-training alignment in large reasoning models (LRMs) has significantly improved their adaptability to diverse safety compliance settings.
arXiv:2607. 27656v1 Announce Type: new Abstract: Looped Transformers create a useful train- and test-time compute axis by reusing the same Transformer block over recurrent depth, increasing effective depth at a fixed parameter count.
arXiv:2607. 27635v1 Announce Type: cross Abstract: Wearable sensors continuously capture fine-grained multivariate time-series data, providing opportunities to model behavioural patterns associated with health outcomes.
arXiv:2607. 27766v1 Announce Type: cross Abstract: On-device in-context learning (ICL) relies on pre-inference retrieval to select demonstrations for useful context before downstream model inference.
arXiv:2607. 27815v1 Announce Type: cross Abstract: Local differential privacy (LDP) protocols are vulnerable to poisoning attacks.
arXiv:2607. 26317v1 Announce Type: cross Abstract: Psychometric calibration for educational tests typically requires costly human response data.
arXiv:2607. 26107v1 Announce Type: cross Abstract: Dense vision-language understanding, including object localization, region recognition, and open-vocabulary semantic segmentation, requires associating language concepts with spatially grounded visual regions.
arXiv:2607. 27069v2 Announce Type: cross Abstract: Closed yes/no spatial benchmarks can reward a correct answer even when the image adds little support beyond no-image contexts.
arXiv:2607. 27891v1 Announce Type: cross Abstract: Temporal graph learning is commonly organized around the evolution of node states or the encoding of interaction histories.
arXiv:2607. 27290v1 Announce Type: new Abstract: Modern telecommunication, cloud, and microservice systems emit correlated alarm cascades when components fail.
arXiv:2607. 27370v1 Announce Type: new Abstract: Sybil attackers are Blockchain actors that adopt the characteristics of regular users to exploit airdrops or influence governance.
arXiv:2607. 27418v1 Announce Type: new Abstract: Long-term ship trajectory prediction is a fundamental capability for maritime safety and autonomous navigation.
arXiv:2607. 01153v3 Announce Type: replace-cross Abstract: Safety evaluations for language models increasingly depend on judgments about ambiguous natural-language behaviour: whether a model followed an instruction, refused appropriately, complied with a policy, or misreported progress in an agentic task.
arXiv:2607. 27260v1 Announce Type: new Abstract: Multimodal continual learning (MMCL) aims to learn emerging knowledge from multimodal data while preserving knowledge.
Aligning large language models to prevent them attributing consciousness to themselves inadvertently alters their representations of mindedness in other entities alongside human beliefs and values. We demonstrate that safety fine-tuning suppresses models' tendencies to attribute minds not only to themselves, but also to non-human animals and natural objects, while also driving a reduction in spiritual belief.
In this paper we present an active, constantly updated AI benchmark which measures the integrity of frontier language models against being co-opted for state-backed information operations. We draw on over 2,100 information operations from a live monitoring pipeline which tracks Russian, Chinese and Iranian state-backed information assets.
Multimodal agents for visual question answering increasingly operate as multi-step trajectories that interleave perception, retrieval, and reasoning, yet evaluation still largely reduces to final-answer accuracy. This aggregate signal cannot tell whether a correct answer was reached through grounded evidence, language priors, or accidental error cancellation.
This work presents Fairness Pruning, a lightweight structural intervention method designed for the management and future mitigation of demographic bias in large language models (LLMs). As a foundational empirical validation of this method, this work focuses on causal bias localization.