arXiv preprints from January 1, 2026 through September 14, 2026 — 21:43:36 EST
Reinforcement learning (RL) post-training is a critical stage in modern language model development, playing a key role in improving alignment and reasoning ability. However, several phenomena remain poorly understood, including the reduction in output diversity. To gain a broader understanding of RL post-training, we analyze the...
This study focuses on optimizing path planning for unmanned ground vehicles (UGVs) in precision agriculture using deep reinforcement learning (DRL) techniques in continuous action spaces. The research begins with a review of traditional grid-based methods, such as A* and Dijkstra's algorithms, and discusses their limitations in...
Large language model (LLM)-integrated applications have become increasingly prevalent, yet face critical security vulnerabilities from prompt injection (PI) attacks. Defending against PI attacks faces two major issues: malicious instructions can be injected through diverse vectors, and injected instructions often lack clear semantic...
Uncrewed aerial vehicles (UAVs) play a pivotal role in ensuring seamless connectivity for Internet of Things (IoT) devices, particularly in scenarios where conventional terrestrial networks are constrained or temporarily unavailable. However, traditional coverage-hole detection approaches, such as minimizing drive tests, are costly,...
Biryani, one of India's most celebrated dishes, exhibits remarkable regional diversity in its preparation, ingredients, and presentation. With the growing availability of online cooking videos, there is unprecedented potential to study such culinary variations using computational tools systematically. However, existing video...
Multilingual large language models (LLMs) achieve strong performance across languages, yet how language capabilities are organized at the neuron level remains poorly understood. Prior work has identified language-related neurons mainly through activation-based heuristics, which conflate language preference with functional importance....
Resource autoscaling mechanisms in cloud environments depend on accurate performance metrics to make optimal provisioning decisions. When infrastructure faults including hardware malfunctions, network disruptions, and software anomalies corrupt these metrics, autoscalers may systematically over- or under-provision resources, resulting...
Automated Audio Captioning aims to describe the semantic content of input audio. Recent works have employed large language models (LLMs) as a text decoder to leverage their reasoning capabilities. However, prior approaches that project audio features into the LLM embedding space without considering cross-modal alignment fail to fully...
Communication robots often need to initiate conversations with people in public spaces. At the same time, such robots must not disturb pedestrians. To handle these two requirements, an agent needs to estimate the communication desires of others based on their behavior and then adjust its own communication activities accordingly. In...
This study proposes FlexiVoice, a text-to-speech (TTS) synthesis system capable of flexible style control with zero-shot voice cloning. The speaking style is controlled by a natural-language instruction and the voice timbre is provided by a speech reference in zero-shot manner. FlexiVoice is built with an LLM core, which takes text as...
We present Isabellm, an LLM-powered theorem prover for Isabelle/HOL that performs fully automatic proof synthesis. Isabellm works with any local LLM on Ollama and APIs such as Gemini CLI, and it is designed to run on consumer grade computers. The system combines a stepwise prover, which uses large language models to propose proof...
Generative AI has unleashed the power of content generation and it has also unwittingly opened the pandora box of realistic deepfake causing a number of social hazards and harm to businesses and personal reputation. The investigation & ramification of Generative AI technology across industries, the resolution & hybridization detection...
Recent advances in synergizing large reasoning models (LRMs) with retrieval-augmented generation (RAG) have shown promising results, yet two critical challenges remain: (1) reasoning models typically operate from a single, unchallenged perspective, limiting their ability to conduct deep, self-correcting reasoning over external...
Mechanism design is pivotal to federated learning (FL) for maximizing social welfare by coordinating self-interested clients. Existing mechanisms, however, often overlook the network effects of client participation and the diverse model performance requirements (i.e., generalization error) across applications, leading to suboptimal...
Large-scale multi-tenant retrieval systems generate extensive query logs but lack curated relevance labels for effective domain adaptation, resulting in substantial underutilized "dark data." This challenge is compounded by the high cost of model updates, as jointly fine-tuning query and document encoders requires full corpus...
Retrieval-Augmented Generation (RAG) systems critically depend on effective document chunking strategies to balance retrieval quality, latency, and operational cost. Traditional chunking approaches, such as fixed-size, rule-based, or fully agentic chunking, often suffer from high token consumption, redundant text generation, limited...
The advent of multi-modal language models (MLLMs) has spurred research into their application across various table understanding tasks. However, their performance in credit table understanding (CTU) for financial credit review remains largely unexplored due to the following barriers: low data consistency, high annotation costs...
The deployment of Machine-Generated Text (MGT) detection systems necessitates processing sensitive user data, creating a fundamental conflict between authorship verification and privacy preservation. Standard anonymization techniques often disrupt linguistic fluency, while rigorous Differential Privacy (DP) mechanisms typically...
Time series decomposition into trend, seasonal, and residual components is a fundamental primitive in data mining and analytics pipelines, underpinning anomaly detection, change-point analysis, and forecasting. Most existing methods require a user-specified or estimated season length and assume stable periodic structure. In large,...
AEC drawings encode geometry and semantics through symbols, layout conventions, and dense annotation, yet it remains unclear whether modern multimodal and vision-language models can reliably interpret this graphical language. We present AECV-Bench, a benchmark for evaluating multimodal and vision-language models on realistic AEC...
User devices are the largest contributor to media related global emissions. For web content, dark mode has been widely recommended as an energy-saving measure for certain display types. However, the energy savings achieved by dark mode may be undermined by user behaviour. This pilot study investigates the unintended consequences of...
An information-theoretic privacy mechanism design is studied, where an agent observes useful data $Y$ which is correlated with the private data $X$. The agent wants to reveal the information to a user, hence, the agent utilizes a privacy mechanism to produce disclosed data $U$ that can be revealed. We assume that the agent has no...
The development of category theory in univalent foundations and the formalization thereof is an active field of research. Categories in that setting are often assumed to be univalent which means that identities and isomorphisms of objects coincide. One consequence hereof is that equivalences and identities coincide for univalent...
Permissionless consensus protocols require a scarce resource to regulate leader election and provide Sybil resistance. Existing paradigms such as Proof of Work and Proof of Stake instantiate this scarcity through parallelizable resources like computation or capital. Once acquired, these resources can be subdivided across many...
Reinforcement learning (RL) offers a principled way to enhance the reasoning capabilities of large language models, yet its effectiveness hinges on training signals that remain informative as models evolve. In practice, RL progress often slows when task difficulty becomes poorly aligned with model capability, or when training is...