Deep-Research agents, which integrate large language models (LLMs) with search tools, have shown success in improving the effectiveness of handling complex queries that require iterative search planning and reasoning over search results. Evaluations on current benchmarks like BrowseComp relies on black-box live web search APIs, have notable limitations in (1) fairness: dynamic and opaque web APIs hinder fair comparisons and reproducibility of deep research methods; (2) transparency: lack of control over the document corpus makes it difficult to isolate retriever contributions. In other words, the current evaluations may compare a complete deep research system at a given time, but they do not foster well-controlled experiments to provide insights into the capability of underlying deep research LLMs. To address these challenges, we introduce BrowseComp-Plus, a benchmark derived from BrowseComp, employing a fixed, carefully curated corpus. Each query in BrowseComp-Plus includes human-verified supporting documents and mined challenging negatives, enabling controlled experimentation. The benchmark is shown to be effective in distinguishing the performance of deep research systems. For instance, the open-source model Search-R1, when paired with the BM25 retriever, achieves 3.86% accuracy, whereas the GPT-5 achieves 55.9%. Integrating the GPT-5 with the Qwen3-Embedding-8B retriever further enhances its accuracy to 70.1% with fewer search calls. This benchmark allows comprehensive evaluation and disentangled analysis of deep research agents and retrieval methods, fostering insights into retrieval effectiveness, citation accuracy, and context engineering in Deep-Research system.

As AI adoption expands across human society, the problem of aligning AI models to match human preferences remains a grand challenge. Currently, the AI alignment field is deeply divided between behavioral and representational approaches, resulting in narrowly aligned models that are more vulnerable to increasingly deceptive misalignment threats. In the face of this fragmentation, we propose an integrated vision for the future of the field. Drawing on related lessons from immunology and cybersecurity, we lay out a set of design principles for the development of Integrated Alignment frameworks that combine the complementary strengths of diverse alignment approaches through deep integration and adaptive coevolution. We highlight the importance of strategic diversity - deploying orthogonal alignment and misalignment detection approaches to avoid homogeneous pipelines that may be "doomed to success". We also recommend steps for greater unification of the AI alignment research field itself, through cross-collaboration, open model weights and shared community resources.

The development, integration, and maintenance of geospatial databases rely heavily on efficient and accurate matching procedures of Geospatial Entity Resolution (ER). While resolution of points-of-interest (POIs) has been widely addressed, resolution of entities with diverse geometries has been largely overlooked. This is partly due to the lack of a uniform technique for embedding heterogeneous geometries seamlessly into a neural network framework. Existing neural approaches simplify complex geometries to a single point, resulting in significant loss of spatial information. To address this limitation, we propose Omni, a geospatial ER model featuring an omni-geometry encoder. This encoder is capable of embedding point, line, polyline, polygon, and multi-polygon geometries, enabling the model to capture the complex geospatial intricacies of the places being compared. Furthermore, Omni leverages transformer-based pre-trained language models over individual textual attributes of place records in an Attribute Affinity mechanism. The model is rigorously tested on existing point-only datasets and a new diverse-geometry geospatial ER dataset. Omni produces up to 12% (F1) improvement over existing methods. Furthermore, we test the potential of Large Language Models (LLMs) to conduct geospatial ER, experimenting with prompting strategies and learning scenarios, comparing the results of pre-trained language model-based methods with LLMs. Results indicate that LLMs show competitive results.

The history of science is punctuated by serendipitous discoveries, where unexpected observations, rather than targeted hypotheses, opened new fields of inquiry. While modern autonomous laboratories excel at accelerating hypothesis testing, their optimization for efficiency risks overlooking these crucial, unplanned findings. To address this gap, we introduce SciLink, an open-source, multi-agent artificial intelligence framework designed to operationalize serendipity in materials research by creating a direct, automated link between experimental observation, novelty assessment, and theoretical simulations. The framework employs a hybrid AI strategy where specialized machine learning models perform quantitative analysis of experimental data, while large language models handle higher-level reasoning. These agents autonomously convert raw data from materials characterization techniques into falsifiable scientific claims, which are then quantitatively scored for novelty against the published literature. We demonstrate the framework's versatility across diverse research scenarios, showcasing its application to atomic-resolution and hyperspectral data, its capacity to integrate real-time human expert guidance, and its ability to close the research loop by proposing targeted follow-up experiments. By systematically analyzing all observations and contextualizing them, SciLink provides a practical framework for AI-driven materials research that not only enhances efficiency but also actively cultivates an environment ripe for serendipitous discoveries, thereby bridging the gap between automated experimentation and open-ended scientific exploration.

The global demand for aquaculture has surged over the past decade, driving the expansion of offshore fish farming systems such as net pens [1, 2]. These structures, while effective for large-scale fish production, are continuously exposed to harsh marine environments that can degrade structural integrity, compromise biosecurity, and increase the risk of fish escape or environmental contamination [3]. As a result, regular and reliable inspection of aquaculture net pens is critical to ensuring operational safety, productivity, and regulatory compliance [4]. Recent advances in underwater robotics, control systems, and computer vision have enabled significant progress in autonomous inspection [5, 6]. Remotely Operated Vehicles (ROVs), in particular, offer a practical platform for deploying sensing payloads such as cameras, sonars and performing close-range inspection in confined underwater environments [7]. However, most existing ROV-based systems operate in isolation, with limited autonomy and minimal adaptability to dynamic conditions such as power constraints, actuator degradation, and evolving mission demands [8, 9]. Moreover, mission planning and coordination typically require expert operators, limiting the scalability and responsiveness of these systems in real-world aquaculture operations [10, 11, 12]. To address these challenges, we propose AquaChat++, a novel framework that combines the reasoning capabilities of Large Language Models (LLMs) with multi-ROV coordination, battery-aware mission planning, and fault-tolerant control [13, 14]. Unlike traditional inspection pipelines that rely on fixed scripts or manual supervision, AquaChat++ enables natural language-driven task planning and dynamic allocation across multiple ROVs.

While model architecture and training objectives are well-studied, tokenization, particularly in multilingual contexts, remains a relatively neglected aspect of Large Language Model (LLM) development. Existing tokenizers often exhibit high token-to-word ratios, inefficient use of context length, and slower inference. We present a systematic study that links vocabulary size, pre-tokenization rules, and training-corpus composition to both token-to-word efficiency and model quality. To ground our analysis in a linguistically diverse context, we conduct extensive experiments on Indic scripts, which present unique challenges due to their high script diversity and orthographic complexity. Drawing on the insights from these analyses, we propose a novel algorithm for data composition that balances multilingual data for tokenizer training. Our observations on pretokenization strategies significantly improve model performance, and our data composition algorithm reduces the average token-to-word ratio by approximately 6% with respect to the conventional data randomization approach. Our tokenizer achieves more than 40% improvement on average token-to-word ratio against stateof-the-art multilingual Indic models. This improvement yields measurable gains in both model performance and inference speed. This highlights tokenization alongside architecture and training objectives as a critical lever for building efficient, scalable multilingual LLMs

Large Vision-Language Models (LVLMs), empowered by the success of Large Language Models (LLMs), have achieved impressive performance across domains. Despite the great advances in LVLMs, they still suffer from the unavailable object hallucination issue, which tends to generate objects inconsistent with the image content. The most commonly used Polling-based Object Probing Evaluation (POPE) benchmark evaluates this issue by sampling negative categories according to category-level statistics, \textit{e.g.}, category frequencies and co-occurrence. However, with the continuous advancement of LVLMs, the POPE benchmark has shown diminishing effectiveness in assessing object hallucination, as it employs a simplistic sampling strategy that overlooks image-specific information and restricts distractors to negative object categories only. In this paper, we introduce the Hallucination searching-based Object Probing Evaluation (HOPE) benchmark, aiming to generate the most misleading distractors (\textit{i.e.}, non-existent objects or incorrect image descriptions) that can trigger hallucination in LVLMs, which serves as a means to more rigorously assess their immunity to hallucination. To explore the image-specific information, the content-aware hallucination searching leverages Contrastive Language-Image Pre-Training (CLIP) to approximate the predictive behavior of LVLMs by selecting negative objects with the highest predicted likelihood as distractors. To expand the scope of hallucination assessment, the description-based hallucination searching constructs highly misleading distractors by pairing true objects with false descriptions. Experimental results show that HOPE leads to a precision drop of at least 9\% and up to 23\% across various state-of-the-art LVLMs, significantly outperforming POPE in exposing hallucination vulnerabilities. The code is available at https://github.com/xiemk/HOPE.

Biomedical named entity recognition (NER) is a high-utility natural language processing (NLP) task, and large language models (LLMs) show promise particularly in few-shot settings (i.e., limited training data). In this article, we address the performance challenges of LLMs for few-shot biomedical NER by investigating a dynamic prompting strategy involving retrieval-augmented generation (RAG). In our approach, the annotated in-context learning examples are selected based on their similarities with the input texts, and the prompt is dynamically updated for each instance during inference. We implemented and optimized static and dynamic prompt engineering techniques and evaluated them on five biomedical NER datasets. Static prompting with structured components increased average F1-scores by 12% for GPT-4, and 11% for GPT-3.5 and LLaMA 3-70B, relative to basic static prompting. Dynamic prompting further improved performance, with TF-IDF and SBERT retrieval methods yielding the best results, improving average F1-scores by 7.3% and 5.6% in 5-shot and 10-shot settings, respectively. These findings highlight the utility of contextually adaptive prompts via RAG for biomedical NER.

How do we predict others from patterns in their behavior and what are the computational constraints that limit this ability? We investigate these questions by modeling human behavior over repeated games of rock, paper, scissors from Brockbank & Vul (2024). Against algorithmic opponents that varied in strategic sophistication, people readily exploit simple transition patterns (e.g., consistently playing rock after paper) but struggle to detect more complex sequential dependencies. To understand the cognitive mechanisms underlying these abilities and their limitations, we deploy Hypothetical Minds (HM), a large language model-based agent that generates and tests hypotheses about opponent strategies, as a cognitive model of this behavior (Cross et al., 2024). We show that when applied to the same experimental conditions, HM closely mirrors human performance patterns, succeeding and failing in similar ways. To better understand the source of HM's failures and whether people might face similar cognitive bottlenecks in this context, we performed a series of ablations and augmentations targeting different components of the system. When provided with natural language descriptions of the opponents' strategies, HM successfully exploited 6/7 bot opponents with win rates >80% suggesting that accurate hypothesis generation is the primary cognitive bottleneck in this task. Further, by systematically manipulating the model's hypotheses through pedagogically-inspired interventions, we find that the model substantially updates its causal understanding of opponent behavior, revealing how model-based analyses can produce testable hypotheses about human cognition.

An ensemble of trained multimodal encoders and Vision-Language Models (VLMs) has become a standard approach for Visual Question Answering (VQA) tasks. However, such naive models often fail to produce responses with the detailed precision necessary for complex, domain-specific applications such as medical VQA. Our representation model, BIND: BLIVA In tegrated with D ense Encoding, extends prior multimodal work by refining the joint embedding space through dense, query-token-based encodings, inspired by contrastive pretraining techniques. This refined encoder powers Med-GRIM, a model designed for medical VQA tasks that leverages graph-based retrieval and prompt engineering to integrate domain-specific knowledge. Rather than relying on compute-heavy fine-tuning of vision and language models on specific datasets, Med-GRIM applies a low-compute, modular workflow with small language models (SLMs) for efficiency. Med-GRIM employs prompt-based retrieval to dynamically inject relevant knowledge, ensuring both accuracy and robustness in its responses. By assigning distinct roles to each agent within the VQA system, Med-GRIM achieves large language model performance at a fraction of the computational cost. Additionally, to support scalable research in zero-shot multi-modal medical applications, we introduce DermaGraph, a novel Graph-RAG dataset comprising diverse dermatological conditions.