The cognitive processes of the hypnotized mind and the computational operations of large language models (LLMs) share deep functional parallels. Both systems generate sophisticated, contextually appropriate behavior through automatic pattern-completion mechanisms operating with limited or unreliable executive oversight. This review examines this convergence across three principles: automaticity, in which responses emerge from associative rather than deliberative processes; suppressed monitoring, leading to errors such as confabulation in hypnosis and hallucination in LLMs; and heightened contextual dependency, where immediate cues (for example, the suggestion of a therapist or the prompt of the user) override stable knowledge. These mechanisms reveal an observer-relative meaning gap: both systems produce coherent but ungrounded outputs that require an external interpreter to supply meaning. Hypnosis and LLMs also exemplify functional agency - the capacity for complex, goal-directed, context-sensitive behavior - without subjective agency, the conscious awareness of intention and ownership that defines human action. This distinction clarifies how purposive behavior can emerge without self-reflective consciousness, governed instead by structural and contextual dynamics. Finally, both domains illuminate the phenomenon of scheming: automatic, goal-directed pattern generation that unfolds without reflective awareness. Hypnosis provides an experimental model for understanding how intention can become dissociated from conscious deliberation, offering insights into the hidden motivational dynamics of artificial systems. Recognizing these parallels suggests that the future of reliable AI lies in hybrid architectures that integrate generative fluency with mechanisms of executive monitoring, an approach inspired by the complex, self-regulating architecture of the human mind.

Increasingly, more and more people are turning to large language models (LLMs) for healthcare advice and consultation, making it important to gauge the efficacy and accuracy of the responses of LLMs to such queries. While there are pre-existing medical benchmarks literature which seeks to accomplish this very task, these benchmarks are almost universally in English, which has led to a notable gap in existing literature pertaining to multilingual LLM evaluation. Within this work, we seek to aid in addressing this gap with Cancer-Myth-Indic, an Indic language benchmark built by translating a 500-item subset of Cancer-Myth, sampled evenly across its original categories, into five under-served but widely used languages from the subcontinent (500 per language; 2,500 translated items total). Native-speaker translators followed a style guide for preserving implicit presuppositions in translation; items feature false presuppositions relating to cancer. We evaluate several popular LLMs under this presupposition stress.

Natural Language Inference (NLI) models have been used in various ways to improve the factuality of LLM outputs. This is typically done by applying an NLI model to judge whether the model output is entailed from the supposed evidence, triggering some corrective actions, such as beam reranking at inference time or RL rewards during training. While NLI models are trained to detect factual inconsistencies over complete sentences, decisions in the common autoregressive generation architecture are made for each evolving text prefix, during decoding. Addressing this setting, we generalize the entailment detection task to apply over arbitrary text prefixes, and suggest its utility for improving generation faithfulness. Providing suitable evaluation and training datasets for this task, we train MiniTruePrefixes, a novel specialized model that better detects factual inconsistencies over text prefixes, outperforming comparable baseline NLI models by 5-14 F1 points in prefix-level entailment. We further demonstrate that integrating MiniTruePrefixes into a controlled decoding framework substantially improves factual consistency in abstractive summarization. When guided by MiniTruePrefixes, LLaMA-3.2-3B-Instruct matches the faithfulness and runtime of the 8B model from the same model family, while using only half the memory.

Recently, the demand for small and efficient reasoning models to support real-world applications has driven the development of knowledge distillation techniques that balance reasoning performance and inference speed. In this paper, we further extend the DistilQwen model family, initialized from the Qwen models, by introducing four model series specifically designed to meet industrial requirements. The distilled model collection comprises: (1) slow-thinking models, optimized for reasoning tasks that require high accuracy; (2) two series of adaptive-thinking models, which dynamically adjust reasoning strategies based on input tasks to maximize efficiency across diverse scenarios; and (3) distilled reward models, which enable further reinforcement learning of reasoning models using distilled knowledge. Comprehensive evaluations across multiple benchmarks demonstrate both high inference efficiency and strong reasoning performance for these models, as well as the practical utility of distilled reward models. We further show that these models support industry practitioners by providing scalable training and inference functionalities on the Alibaba Cloud PAI (Platform for Artificial Intelligence) platform.

Understanding Rebus Puzzles (Rebus Puzzles use pictures, symbols, and letters to represent words or phrases creatively) requires a variety of skills such as image recognition, cognitive skills, commonsense reasoning, multi-step reasoning, image-based wordplay, etc., making this a challenging task for even current Vision-Language Models. In this paper, we present $\left|\,\circlearrowright\,\boxed{\text{BUS}}\,\right|$, a large and diverse benchmark of $1,333$ English Rebus Puzzles containing different artistic styles and levels of difficulty, spread across 18 categories such as food, idioms, sports, finance, entertainment, etc. We also propose $RebusDescProgICE$, a model-agnostic framework which uses a combination of an unstructured description and code-based, structured reasoning, along with better, reasoning-based in-context example selection, improving the performance of Vision-Language Models on $\left|\,\circlearrowright\,\boxed{\text{BUS}}\,\right|$ by $2.1-4.1\%$ and $20-30\%$ using closed-source and open-source models respectively compared to Chain-of-Thought Reasoning.

Mobile applications increasingly rely on sensor data to infer user context and deliver personalized experiences. Yet the mechanisms behind this personalization remain opaque to users and researchers alike. This paper presents a sandbox system that uses sensor spoofing and persona simulation to audit and visualize how mobile apps respond to inferred behaviors. Rather than treating spoofing as adversarial, we demonstrate its use as a tool for behavioral transparency and user empowerment. Our system injects multi-sensor profiles - generated from structured, lifestyle-based personas - into Android devices in real time, enabling users to observe app responses to contexts such as high activity, location shifts, or time-of-day changes. With automated screenshot capture and GPT-4 Vision-based UI summarization, our pipeline helps document subtle personalization cues. Preliminary findings show measurable app adaptations across fitness, e-commerce, and everyday service apps such as weather and navigation. We offer this toolkit as a foundation for privacy-enhancing technologies and user-facing transparency interventions.

Large language models (LLMs) with integrated search tools show strong promise in open-domain question answering (QA), yet they often struggle to produce complete answer set to complex questions such as Which actor from the film Heat won at least one Academy Award?, which requires (1) distinguishing between multiple films sharing the same title and (2) reasoning across a large set of actors to gather and integrate evidence. Existing QA benchmarks rarely evaluate both challenges jointly. To address this, we introduce DeepAmbigQAGen, an automatic data generation pipeline that constructs QA tasks grounded in text corpora and linked knowledge graph, generating natural and verifiable questions that systematically embed name ambiguity and multi-step reasoning. Based on this, we build DeepAmbigQA, a dataset of 3,600 questions requiring multi-hop reasoning and half of them explicit name ambiguity resolving. Experiments reveal that, even state-of-the-art GPT-5 show incomplete answers, achieving only 0.13 exact match on ambiguous questions and 0.21 on non-ambiguous questions. These findings highlight the need for more robust QA systems aimed at information gathering and answer completeness.

Continuous Integration (CI) is a cornerstone of modern collaborative software development, and numerous CI platforms are available. Differences in maintenance overhead, reliability, and integration depth with code-hosting platforms make migration between CI platforms a common practice. A central step in migration is translating CI configurations, which is challenging due to the intrinsic complexity of CI configurations and the need to understand semantic differences and relationships across CI platforms. With the advent of large language models (LLMs), recent advances in software engineering highlight their potential for CI configuration translation. In this paper, we present a study on LLM-based CI configuration translation, focusing on the migration from Travis CI to GitHub Actions. First, using 811 migration records, we quantify the effort involved and find that developers read an average of 38 lines of Travis configuration and write 58 lines of GitHub Actions configuration, with nearly half of the migrations requiring multiple commits. We further analyze translations produced by each of the four LLMs and identify 1,121 issues grouped into four categories: logic inconsistencies (38%), platform discrepancies (32%), environment errors (25%), and syntax errors (5%). Finally, we evaluate three enhancement strategies and show that combining guideline-based prompting with iterative refinement achieves the best performance, reaching a Build Success Rate of 75.5%-nearly a threefold improvement over GPT-4o with a basic prompt.

The rise of data-driven algorithms and, most notably, applications of deep learning has led to concerns about a lack of thorough human oversight of socially important decisions that are either delegated in their entirety to machines, or made by humans based on machine recommendations. Explainable AI (XAI) approaches attempt to mitigate these concerns by helping (typically human) users understand how and why algorithms produce specific outputs [1]. An important class of XAI methods focuses on providing explanations of black-box classifiers that attribute classification outcomes (which one may consider decisions or decision recommendations) to input characteristics (feature values) [2, 3]. Such feature attribution methods can be considered meta-reasoning functions that approximate classifier behavior with the objective of providing users a reasonably faithful intuition of behavioral fundamentals. One of the most prominent feature attribution methods is SHAP, which is based on the Shapley value in cooperative game theory that quantifies players' (feature values') contributions to coalition utility (classification outcomes) [4]. Feature attribution methods have, in general, limitations: notably, they are necessarily approximations, and as purely technical tools, they cannot fully consider crucial nuances of the socio-technical systems they are embedded in [5]; for example, the visualizations provided out-of-the-box by feature attribution software libraries may be difficult to interpret [6]. Still, Shapley value-based approaches can be considered a reasonable choice for facilitating black-box explainability, notably because (i) they are based on well-established and intuitive mathematical principles of the Shapley value and (ii) there is at least some evidence of their potential usefulness, also relative to alternative approaches [6]. However, the Shapley value cannot straight-forwardly be applied to inference from Large Language Models (LLMs), which power many of the currently emerging AI applications.

5G technology enables mobile Internet access for billions of users. Answering expert-level questions about 5G specifications requires navigating thousands of pages of cross-referenced standards that evolve across releases. Existing retrieval-augmented generation (RAG) frameworks, including telecom-specific approaches, rely on semantic similarity and cannot reliably resolve cross-references or reason about specification evolution. We present DeepSpecs, a RAG system enhanced by structural and temporal reasoning via three metadata-rich databases: SpecDB (clause-aligned specification text), ChangeDB (line-level version diffs), and TDocDB (standardization meeting documents). DeepSpecs explicitly resolves cross-references by recursively retrieving referenced clauses through metadata lookup, and traces specification evolution by mining changes and linking them to Change Requests that document design rationale. We curate two 5G QA datasets: 573 expert-annotated real-world questions from practitioner forums and educational resources, and 350 evolution-focused questions derived from approved Change Requests. Across multiple LLM backends, DeepSpecs outperforms base models and state-of-the-art telecom RAG systems; ablations confirm that explicit cross-reference resolution and evolution-aware retrieval substantially improve answer quality, underscoring the value of modeling the structural and temporal properties of 5G standards.