large language models

Characterizing the Ecosystem of Open-Source Uncensored Large Language Models: Supply Chain, Governance, and Risks

From Metrics to Insights: Towards High-Precision and Explainable Query-Analysis System in Oil and Gas Operations with Large Language Models

Abstract Natural language-based data query and analysis are crucial for enterprise operation and management. Currently, querying across a large number of metrics faces two main challenges: low query accuracy and the lack of an interpretable analytical process in query results. This paper constructs a natural language q...

SRFE: Measuring Chain-of-Thought Faithfulness with Independent Stepwise Visual Probes

Using stepwise probe as an external reference or Chain-of-Thought (CoT) to evaluate visual-language model (VLM) analysis ability is common, but the faithfulness of PASS/FAIL verdicts in VLM produced by explicit CoT and independent SRFE probe needs further work to figure out. Prior works on CoT faithfulness are mainly a...

From Prompt to Plot: An MCP Framework for LLM-Driven Engineering Automation

The wellbore schematic is one of the most critical diagrams in drilling engineering design. Currently, these schematics are predominantly generated manually, a process that is tedious, inefficient, and lacks standardization, leading to significant inconsistencies among diagrams produced by different engineers. To add...

Flowvision AI Context Aware Production Forecasting Via Virtual Flow Metering and Operational Text Mining

Abstract This study, completed during the 2025 GOTECH® hackathon, presents a data-driven framework to identify production-optimization opportunities and to forecast incremental production. The approach was applied to an offshore platform with wells operating under high, low, and very low pressure, where planned and rea...

Research on the Impact Mechanism of the Fusion Model on Stock Valuation Correction and Market Risk Forecasting in an Uncertain Environment

Under geopolitical tensions and monetary policy shifts, global stock market volatility has intensified. Traditional valuation models and risk forecasting models, relying on historical static parameters and overlooking non-quantitative information such as management tone, suffer from systematic biases. This study uses t...

The Three-Stage Trajectory of Large Language Models: 2022 to 2026

This literature review puts forward a three-stage model of Large Language Models (LLMs), delineates the research goals of LLM in three stages and predicts future trends. Against the rapid, fragmented advancement of large-language-model technologies, systematic sorting of their evolutionary trajectories helps researcher...

The Psychological Mechanisms and Social Implications of Over-Trust in AI Chat-Based on GPT-4 Persuasion Experiments

As large language models have been increasingly applied to daily life in information seeking and decision-making, public trust in AI output has greatly improved. This brings about the worry that human critical thinking could become impaired. Using Salvi et al.'s GPT-4 conversational persuasion experiment as a start, th...

Analysis of Intelligent Operation and Maintenance Technology Strategy of Communication Networks Based on Large Language Models

With the continuous improvement of communication network scale and business complexity, the traditional operation and maintenance methods relying on manual experience and rules face certain limitations in multi-source information processing and complex fault analysis. The ability of large language model in natural lang...

Exploring the Synergistic Effects of LLM-Based Conversational AI and Teacher Intervention on L2 Learners' Interactional Competence

This study investigates how Large Language Model (LLM)-based conversational AI and human teacher intervention affect the development of Second Language (L2) Interactional Competence (IC). Employing a two-phase qualitative research design, the study first analyzes interaction texts and reflective journals from five coll...

Ollama in the Wild: A Longitudinal Measurement of Exposed Ollama LLM Endpoints at Internet Scale

Self-hosted large language model (LLM) serving is emerging as a distinct category of Internet service, but we still know little about how these deployments appear and change on the public Internet. We present a 365-day longitudinal measurement of exposed Ollama endpoints (port 11434) from February 2025 to February 2026...

Artifacts of "Grounding Large Language Models in a Symbolic Formalization Environment for Natural-Language Planning"

Experimental results for the paper "Grounding Large Language Models in a Symbolic Formalization Environment for Natural-Language Planning". The archive contains the results of 5,100 formalization sessions. Qwen 3.5 (122B) and GPT-OSS (120B) each formalized the 2,550 Blocksworld, Logistics and Rovers problems of the ben...