The philosophy of mathematical practice (PMP) looks to evidence from working mathematics to help settle philosophical questions. One prominent program under the PMP banner is the study of explanation in mathematics, which aims to understand what sorts of proofs mathematicians consider explanatory and what role the pursuit of explanation plays in mathematical practice. In an effort to address worries about cherry-picked examples and file-drawer problems in PMP, a handful of authors have recently turned to corpus analysis methods as a promising alternative to small-scale case studies. This paper reports the results from such a corpus study facilitated by Google's Gemini 2.5 Pro, a model whose reasoning capabilities, advances in hallucination control and large context window allow for the accurate analysis of hundreds of pages of text per query. Based on a sample of 5000 mathematics papers from arXiv.org, the experiments yielded a dataset of hundreds of useful annotated examples. Its aim was to gain insight on questions like the following: How often do mathematicians make claims about explanation in the relevant sense? Do mathematicians' explanatory practices vary in any noticeable way by subject matter? Which philosophical theories of explanation are most consistent with a large body of non-cherry-picked examples? How might philosophers make further use of AI tools to gain insights from large datasets of this kind? As the first PMP study making extensive use of LLM methods, it also seeks to begin a conversation about these methods as research tools in practice-oriented philosophy and to evaluate the strengths and weaknesses of current models for such work.

Edgar Allan Poe noted, "Truth often lurks in the shadow of error," highlighting the deep complexity intrinsic to the interplay between truth and falsehood, notably under conditions of cognitive and informational asymmetry. This dynamic is strikingly evident in large language models (LLMs). Despite their impressive linguistic generation capabilities, LLMs sometimes produce information that appears factually accurate but is, in reality, fabricated, an issue often referred to as 'hallucinations'. The prevalence of these hallucinations can mislead users, affecting their judgments and decisions. In sectors such as finance, law, and healthcare, such misinformation risks causing substantial economic losses, legal disputes, and health risks, with wide-ranging consequences.In our research, we have methodically categorized, analyzed the causes, detection methods, and solutions related to LLM hallucinations. Our efforts have particularly focused on understanding the roots of hallucinations and evaluating the efficacy of current strategies in revealing the underlying logic, thereby paving the way for the development of innovative and potent approaches. By examining why certain measures are effective against hallucinations, our study aims to foster a comprehensive approach to tackling this issue within the domain of LLMs.

Medical diagnosis prediction plays a critical role in disease detection and personalized healthcare. While machine learning (ML) models have been widely adopted for this task, their reliance on supervised training limits their ability to generalize to unseen cases, particularly given the high cost of acquiring large, labeled datasets. Large language models (LLMs) have shown promise in leveraging language abilities and biomedical knowledge for diagnosis prediction. However, they often suffer from hallucinations, lack structured medical reasoning, and produce useless outputs. T o address these challenges, we propose KERAP, a knowledge graph (KG)-enhanced reasoning approach that improves LLM-based diagnosis prediction through a multi-agent architecture. Our framework consists of a linkage agent for attribute mapping, a retrieval agent for structured knowledge extraction, and a prediction agent that iteratively refines diagnosis predictions. Experimental results demonstrate that KERAP enhances diagnostic reliability efficiently, offering a scalable and interpretable solution for zero-shot medical diagnosis prediction.

Continual fine-tuning of Large Language Models (LLMs) is hampered by the trade-off between efficiency and expressiveness. Low-Rank Adaptation (LoRA) offers efficiency but constrains the model's ability to learn new tasks and transfer knowledge due to its low-rank nature and reliance on explicit parameter constraints. We propose GORP (Gradient LOw Rank Projection) for Continual Learning, a novel training strategy that overcomes these limitations by synergistically combining full and low-rank parameters and jointly updating within a unified low-rank gradient subspace. GORP expands the optimization space while preserving efficiency and mitigating catastrophic forgetting. Extensive experiments on continual learning benchmarks demonstrate GORP's superior performance compared to existing state-of-the-art approaches. Code is available at https://github.com/Wcxwcxw/GORP.

Recent research has demonstrated the effectiveness of Artificial Intelligence (AI), and more specifically, Large Language Models (LLMs), in supporting network configuration synthesis and automating network diagnosis tasks, among others. In this preliminary work, we restrict our focus to the application of AI agents to network troubleshooting and elaborate on the need for a standardized, reproducible, and open benchmarking platform, where to build and evaluate AI agents with low operational effort.

Recently, mobile manipulation has attracted increasing attention for enabling language-conditioned robotic control in household tasks. However, existing methods still face challenges in coordinating mobile base and manipulator, primarily due to two limitations. On the one hand, they fail to explicitly model the influence of the mobile base on manipulator control, which easily leads to error accumulation under high degrees of freedom. On the other hand, they treat the entire mobile manipulation process with the same visual observation modality (e.g., either all 2D or all 3D), overlooking the distinct multimodal perception requirements at different stages during mobile manipulation. To address this, we propose the Adaptive Coordination Diffusion Transformer (AC-DiT), which enhances mobile base and manipulator coordination for end-to-end mobile manipulation. First, since the motion of the mobile base directly influences the manipulator's actions, we introduce a mobility-to-body conditioning mechanism that guides the model to first extract base motion representations, which are then used as context prior for predicting whole-body actions. This enables whole-body control that accounts for the potential impact of the mobile base's motion. Second, to meet the perception requirements at different stages of mobile manipulation, we design a perception-aware multimodal conditioning strategy that dynamically adjusts the fusion weights between various 2D visual images and 3D point clouds, yielding visual features tailored to the current perceptual needs. This allows the model to, for example, adaptively rely more on 2D inputs when semantic information is crucial for action prediction, while placing greater emphasis on 3D geometric information when precise spatial understanding is required. We validate AC-DiT through extensive experiments on both simulated and real-world mobile manipulation tasks.

Grading handwritten, open-ended responses remains a major bottleneck in large university STEM courses. We introduce Pensieve (https://www.pensieve.co), an AI-assisted grading platform that leverages large language models (LLMs) to transcribe and evaluate student work, providing instructors with rubric-aligned scores, transcriptions, and confidence ratings. Unlike prior tools that focus narrowly on specific tasks like transcription or rubric generation, Pensieve supports the entire grading pipeline-from scanned student submissions to final feedback-within a human-in-the-loop interface. Pensieve has been deployed in real-world courses at over 20 institutions and has graded more than 300,000 student responses. We present system details and empirical results across four core STEM disciplines: Computer Science, Mathematics, Physics, and Chemistry. Our findings show that Pensieve reduces grading time by an average of 65%, while maintaining a 95.4% agreement rate with instructor-assigned grades for high-confidence predictions.

Retrieval-augmented Generation (RAG) has primarily been studied in limited settings, such as factoid question answering; more challenging, reasoning-intensive benchmarks have seen limited success from minimal RAG. In this work, we challenge this prevailing view on established, reasoning-intensive benchmarks: MMLU, MMLU Pro, AGI Eval, GPQA, and MATH. We identify a key missing component in prior work: a usable, web-scale datastore aligned with the breadth of pretraining data. To this end, we introduce CompactDS: a diverse, high-quality, web-scale datastore that achieves high retrieval accuracy and subsecond latency on a single-node. The key insights are (1) most web content can be filtered out without sacrificing coverage, and a compact, high-quality subset is sufficient; and (2) combining in-memory approximate nearest neighbor (ANN) retrieval and on-disk exact search balances speed and recall. Using CompactDS, we show that a minimal RAG pipeline achieves consistent accuracy improvements across all benchmarks and model sizes (8B--70B), with relative gains of 10% on MMLU, 33% on MMLU Pro, 14% on GPQA, and 19% on MATH. No single data source suffices alone, highlighting the importance of diversity of sources (web crawls, curated math, academic papers, textbooks). Finally, we show that our carefully designed in-house datastore matches or outperforms web search engines such as Google Search, as well as recently proposed, complex agent-based RAG systems--all while maintaining simplicity, reproducibility, and self-containment. We release CompactDS and our retrieval pipeline, supporting future research exploring retrieval-based AI systems.

The widespread dissemination of toxic content on social media poses a serious threat to both online environments and public discourse, highlighting the urgent need for detoxification methods that effectively remove toxicity while preserving the original semantics. However, existing approaches often struggle to simultaneously achieve strong detoxification performance, semantic preservation, and robustness to out-of-distribution data. Moreover, they typically rely on costly, manually annotated parallel corpora while showing poor data efficiency. To address these challenges, we propose a two-stage training framework that jointly optimizes for data efficiency, semantic preservation, and model generalization. We first perform supervised fine-tuning on a small set of high-quality, filtered parallel data to establish a strong initialization. Then, we leverage unlabeled toxic inputs and a custom-designed reward model to train the LLM using Group Relative Policy Optimization. Experimental results demonstrate that our method effectively mitigates the trade-offs faced by previous work, achieving state-of-the-art performance with improved generalization and significantly reduced dependence on annotated data. Our code is available at: https://github.com/allacnobug/Detoxification-of-Text.

Prompting is the primary method by which we study and control large language models. It is also one of the most powerful: nearly every major capability attributed to LLMs-few-shot learning, chain-of-thought, constitutional AI-was first unlocked through prompting. Yet prompting is rarely treated as science and is frequently frowned upon as alchemy. We argue that this is a category error. If we treat LLMs as a new kind of complex and opaque organism that is trained rather than programmed, then prompting is not a workaround: it is behavioral science. Mechanistic interpretability peers into the neural substrate, prompting probes the model in its native interface: language. We contend that prompting is not inferior, but rather a key component in the science of LLMs.