Alignment, interpretability, red-teaming, bias and privacy: the research on what these systems do when they misbehave.
arXiv:2608. 10964v1 Announce Type: cross Abstract: Reinforcement Fine-Tuning (RFT) has enabled medical Multimodal Large Language Models (MLLMs) to produce Chain-of-Thought (CoT) reasoning for visual question answering, yet these models suffer from $\textit{confidence miscalibration}$---a systematic gap between expressed certainty and actual diagnostic accuracy that undermines clinical trust.
arXiv:2510. 04399v3 Announce Type: replace Abstract: We develop a learning-theoretic framework for analyzing self-improving agents by decomposing self-modification into five axes.
arXiv:2602. 24287v2 Announce Type: replace-cross Abstract: In multi-turn conversations, large language models typically condition on the full conversation history: both past user prompts and assistant responses.
arXiv:2604. 01039v3 Announce Type: replace-cross Abstract: System Instructions in Large Language Models (LLMs) are commonly used to enforce safety policies, define agent behavior, and protect sensitive operational context in agentic AI applications.
arXiv:2605. 16411v3 Announce Type: replace-cross Abstract: Hallucination remains a fundamental challenge in vision-language models (VLMs), where autoregressive generation may produce linguistically plausible yet physically inconsistent or visually ungrounded responses due to likelihood maximization under joint probabilistic modeling.
arXiv:2608. 10172v1 Announce Type: new Abstract: Mechanistic interpretability explains models by identifying circuits inside them, but has no way to tell whether a circuit is a property of the model or an artifact of the method that found it.
arXiv:2608. 10891v1 Announce Type: new Abstract: Time series forecasting in privacy-sensitive domains often requires training models on released data rather than original observations.
arXiv:2608. 09997v1 Announce Type: new Abstract: Transformers have had a profound impact on the world of language processing and computer vision.
arXiv:2608. 10619v1 Announce Type: new Abstract: Message-passing neural networks (MPNNs) often struggle when task-relevant information is distributed across distant regions of a graph, since local propagation must compress remote signals through limited structural interfaces.
arXiv:2608. 11027v1 Announce Type: new Abstract: Benchmark leaderboards summarize how well a language model performs, but not how its behavior relates to that of other models or changes across generations.
arXiv:2608. 10835v1 Announce Type: cross Abstract: Large Vision-Language Models (LVLMs) achieve impressive visual reasoning and dialogue capabilities, yet frequently hallucinate content unsupported by the visual input.
arXiv:2608. 11003v1 Announce Type: cross Abstract: In this work, we study the information bottleneck under perfect privacy, with particular emphasis on the active-rate regime, where the representation-rate constraint is binding and directly limits the achievable utility.
arXiv:2605. 28626v2 Announce Type: replace Abstract: Hybrid interpretable models combine a transparent component with a black-box model by assigning some examples to the former and deferring the rest to the latter.
arXiv:2606. 29243v2 Announce Type: replace Abstract: We introduce KrishokChat, an 85,979-instance Bengali agricultural benchmark built from 284 government publications, 13 institutions, and six regional dialects.
arXiv:2505. 02257v2 Announce Type: replace-cross Abstract: In regions lacking medically certified causes of death, verbal autopsy (VA) is a widely used tool to ascertain the cause of death through interviews with caregivers.
arXiv:2603. 10485v3 Announce Type: replace-cross Abstract: In this work, we study the convergence properties of the Dual Space Preconditioned Gradient Descent, encompassing optimizers such as Normalized Gradient Descent and Gradient Clipping.
arXiv:2608. 10470v1 Announce Type: new Abstract: Fair representation learning with a continuous sensitive attribute $S$ requires a representation $Z$ that is statistically independent of $S$.
arXiv:2608. 10137v1 Announce Type: cross Abstract: Grammar Constrained Decoding (GCD) forces Language Models (LMs) to produce syntactically valid outputs by masking out non-conforming tokens at each step.
arXiv:2607. 15606v2 Announce Type: replace Abstract: Synthetic sequential tabular data are increasingly used for privacy-preserving data sharing and data-driven research, but evaluating their fidelity remains difficult because temporal structure is easily lost under conventional tabular metrics.
arXiv:2608. 10237v1 Announce Type: new Abstract: Contrastive learning and Siamese embedding models have become the foundation of modern verification systems, where decisions are governed not by discrete classification boundaries, but by relational geometry in embedding space.