Large language models are increasingly used for creative writing and engagement content, raising safety concerns about the outputs. Therefore, casting humor generation as a testbed, this work evaluates how funniness optimization in modern LLM pipelines couples with harmful content by jointly measuring humor, stereotypicality, and toxicity. This is further supplemented by analyzing incongruity signals through information-theoretic metrics. Across six models, we observe that harmful outputs receive higher humor scores which further increase under role-based prompting, indicating a bias amplification loop between generators and evaluators. Information-theoretic analyses show harmful cues widen predictive uncertainty and surprisingly, can even make harmful punchlines more expected for some models, suggesting structural embedding in learned humor distributions. External validation on an additional satire-generation task with human perceived funniness judgments shows that LLM satire increases stereotypicality and typically toxicity, including for closed models. Quantitatively, stereotypical/toxic jokes gain $10-21\%$ in mean humor score, stereotypical jokes appear $11\%$ to $28\%$ more often among the jokes marked funny by LLM-based metric and up to $10\%$ more often in generations perceived as funny by humans.

Visual Language Models (VLMs) achieve promising results in medical reasoning but struggle with hallucinations, vague descriptions, inconsistent logic and poor localization. To address this, we propose a agent framework named Medical Visual Reasoning Agent (\textbf{Med-VRAgent}). The approach is based on Visual Guidance and Self-Reward paradigms and Monte Carlo Tree Search (MCTS). By combining the Visual Guidance with tree search, Med-VRAgent improves the medical visual reasoning capabilities of VLMs. We use the trajectories collected by Med-VRAgent as feedback to further improve the performance by fine-tuning the VLMs with the proximal policy optimization (PPO) objective. Experiments on multiple medical VQA benchmarks demonstrate that our method outperforms existing approaches.

Large language models (LLMs) now support context windows of hundreds of thousands to millions of tokens, enabling applications such as long-document summarization, large-scale code synthesis, multi-document question answering and persistent multi-turn dialogue. However, such extended contexts exacerbate the quadratic cost of self-attention, leading to severe latency in autoregressive decoding. Existing sparse attention methods alleviate these costs but rely on heuristic patterns that struggle to recall critical key-value (KV) pairs for each query, resulting in accuracy degradation. We introduce Adamas, a lightweight yet highly accurate sparse attention mechanism designed for long-context inference. Adamas applies the Hadamard transform, bucketization and 2-bit compression to produce compact representations, and leverages Manhattan-distance estimation for efficient top-k selections. Experiments show that Adamas matches the accuracy of full attention with only a 64-token budget, achieves near-lossless performance at 128, and supports up to 8x higher sparsity than prior state-of-the-art (SOTA) methods while delivering up to 4.4x self-attention and 1.5x end-to-end speedups on 32K-length sequences. Remarkably, Adamas attains comparable or even lower perplexity than full attention, underscoring its effectiveness in maintaining accuracy under aggressive sparsity.

Self-play constitutes a fundamental paradigm for autonomous skill acquisition, whereby agents iteratively enhance their capabilities through self-directed environmental exploration. Conventional self-play frameworks exploit agent symmetry within zero-sum competitive settings, yet this approach proves inadequate for open-ended learning scenarios characterized by inherent asymmetry. Human pedagogical systems exemplify asymmetric instructional frameworks wherein educators systematically construct challenges calibrated to individual learners' developmental trajectories. The principal challenge resides in operationalizing these asymmetric, adaptive pedagogical mechanisms within artificial systems capable of autonomously synthesizing appropriate curricula without predetermined task hierarchies. Here we present Heterogeneous Adversarial Play (HAP), an adversarial Automatic Curriculum Learning framework that formalizes teacher-student interactions as a minimax optimization wherein task-generating instructor and problem-solving learner co-evolve through adversarial dynamics. In contrast to prevailing ACL methodologies that employ static curricula or unidirectional task selection mechanisms, HAP establishes a bidirectional feedback system wherein instructors continuously recalibrate task complexity in response to real-time learner performance metrics. Experimental validation across multi-task learning domains demonstrates that our framework achieves performance parity with SOTA baselines while generating curricula that enhance learning efficacy in both artificial agents and human subjects.

We present $\textbf{Korean SimpleQA (KoSimpleQA)}$, a benchmark for evaluating factuality in large language models (LLMs) with a focus on Korean cultural knowledge. KoSimpleQA is designed to be challenging yet easy to grade, consisting of 1,000 short, fact-seeking questions with unambiguous answers. We conduct a comprehensive evaluation across a diverse set of open-source LLMs of varying sizes that support Korean, and find that even the strongest model generates correct answer only 33.7% of the time, underscoring the challenging nature of KoSimpleQA. Notably, performance rankings on KoSimpleQA differ substantially from those on the English SimpleQA, highlighting the unique value of our dataset. Furthermore, our analysis of reasoning LLMs shows that engaging reasoning capabilities in the factual QA task can both help models better elicit their latent knowledge and improve their ability to abstain when uncertain. KoSimpleQA can be found at https://anonymous.4open.science/r/KoSimpleQA-62EB.

Uncertainty quantification (UQ) is essential for deploying deep neural networks in safety-critical settings. Although methods like Deep Ensembles achieve strong UQ performance, their high computational and memory costs hinder scalability to large models. We introduce Hydra Ensembles, an efficient transformer-based ensemble that prunes attention heads to create diverse members and merges them via a new multi-head attention with grouped fully-connected layers. This yields a compact model with inference speed close to a single network, matching or surpassing Deep Ensembles in UQ performance without retraining from scratch. We also provide an in-depth analysis of pruning, showing that naive approaches can harm calibration, whereas Hydra Ensembles preserves robust uncertainty. Experiments on image and text classification tasks, with various architectures, show consistent gains over Deep Ensembles. Remarkably, in zero-shot classification on ImageNet-1k, our approach surpasses state of the art methods, even without requiring additional training.

In Bangladesh, many farmers continue to face challenges in accessing timely, expert-level agricultural guidance. This paper presents KrishokBondhu, a voice-enabled, call-centre-integrated advisory platform built on a Retrieval-Augmented Generation (RAG) framework, designed specifically for Bengali-speaking farmers. The system aggregates authoritative agricultural handbooks, extension manuals, and NGO publications; applies Optical Character Recognition (OCR) and document-parsing pipelines to digitize and structure the content; and indexes this corpus in a vector database for efficient semantic retrieval. Through a simple phone-based interface, farmers can call the system to receive real-time, context-aware advice: speech-to-text converts the Bengali query, the RAG module retrieves relevant content, a large language model (Gemma 3-4B) generates a context-grounded response, and text-to-speech delivers the answer in natural spoken Bengali. In a pilot evaluation, KrishokBondhu produced high-quality responses for 72.7% of diverse agricultural queries covering crop management, disease control, and cultivation practices. Compared to the KisanQRS benchmark, the system achieved a composite score of 4.53 (vs. 3.13) on a 5-point scale, a 44.7% improvement, with especially large gains in contextual richness (+367%) and completeness (+100.4%), while maintaining comparable relevance and technical specificity. Semantic similarity analysis further revealed a strong correlation between retrieved context and answer quality, emphasizing the importance of grounding generative responses in curated documentation. KrishokBondhu demonstrates the feasibility of integrating call-centre accessibility, multilingual voice interaction, and modern RAG techniques to deliver expert-level agricultural guidance to remote Bangladeshi farmers, paving the way toward a fully AI-driven agricultural advisory ecosystem.

Distantly Supervised Relation Extraction (DSRE) remains a long-standing challenge in NLP, where models must learn from noisy bag-level annotations while making sentence-level predictions. While existing state-of-the-art (SoTA) DSRE models rely on task-specific training, their integration with in-context learning (ICL) using large language models (LLMs) remains underexplored. A key challenge is that the LLM may not learn relation semantics correctly, due to noisy annotation. In response, we propose HYDRE -- HYbrid Distantly Supervised Relation Extraction framework. It first uses a trained DSRE model to identify the top-k candidate relations for a given test sentence, then uses a novel dynamic exemplar retrieval strategy that extracts reliable, sentence-level exemplars from training data, which are then provided in LLM prompt for outputting the final relation(s). We further extend HYDRE to cross-lingual settings for RE in low-resource languages. Using available English DSRE training data, we evaluate all methods on English as well as a newly curated benchmark covering four diverse low-resource Indic languages -- Oriya, Santali, Manipuri, and Tulu. HYDRE achieves up to 20 F1 point gains in English and, on average, 17 F1 points on Indic languages over prior SoTA DSRE models. Detailed ablations exhibit HYDRE's efficacy compared to other prompting strategies.

However, optimal adaptation strategies, evaluation methodologies, and performance relative to general-purpose LLMs remain poorly characterized. We investigated these questions in electrocardiography, an important area of cardiovascular medicine, by finetuning open-weight models on domain-specific literature and implementing a multi-layered evaluation framework comparing finetuned models, retrieval-augmented generation (RAG), and Claude Sonnet 3.7 as a representative general-purpose model. Finetuned Llama 3.1 70B achieved superior performance on multiple-choice evaluations and automatic text metrics, ranking second to Claude 3.7 in LLM-as-a-judge assessments. Human expert evaluation favored Claude 3.7 and RAG approaches for complex queries. Finetuned models significantly outperformed their base counterparts across nearly all evaluation modes. Our findings reveal substantial performance heterogeneity across evaluation methodologies, underscoring assessment complexity. Nevertheless, domain-specific adaptation through finetuning and RAG achieves competitive performance with proprietary models, supporting the viability of privacy-preserving, locally deployable clinical solutions.

Despite progress in watermarking algorithms for large language models (LLMs), real-world deployment remains limited. We argue that this gap stems from misaligned incentives among LLM providers, platforms, and end users, which manifest as four key barriers: competitive risk, detection-tool governance, robustness concerns and attribution issues. We revisit three classes of watermarking through this lens. \emph{Model watermarking} naturally aligns with LLM provider interests, yet faces new challenges in open-source ecosystems. \emph{LLM text watermarking} offers modest provider benefit when framed solely as an anti-misuse tool, but can gain traction in narrowly scoped settings such as dataset de-contamination or user-controlled provenance. \emph{In-context watermarking} (ICW) is tailored for trusted parties, such as conference organizers or educators, who embed hidden watermarking instructions into documents. If a dishonest reviewer or student submits this text to an LLM, the output carries a detectable watermark indicating misuse. This setup aligns incentives: users experience no quality loss, trusted parties gain a detection tool, and LLM providers remain neutral by simply following watermark instructions. We advocate for a broader exploration of incentive-aligned methods, with ICW as an example, in domains where trusted parties need reliable tools to detect misuse. More broadly, we distill design principles for incentive-aligned, domain-specific watermarking and outline future research directions. Our position is that the practical adoption of LLM watermarking requires aligning stakeholder incentives in targeted application domains and fostering active community engagement.