large language models

An explicit exponential-period representation of the V_quad connection coefficient

Correction notice (v1.3). This version withdraws a claim. Through v1.2 the Discussion stated that this paper “completes part (ii)(a) for d=2” of the master conjecture of the Sakai stratification (10.5281/zenodo.20694840). Part (ii)(a) concerns the Edge–Borel coefficient, a different object from the connection coefficie...

TrafficRAG: A Multimodal RAG Framework for Traffic Accident Liability Determination

Turing-Computability Excludes Phenomenal Consciousness: A Derivation from the NEMS No-Emulation and SIAM Theorems, with Implications for Large Language Models

Derived from two machine-checked theorems in the NEMS (No External Model Selection) framework, we show that no Turing-computable system can satisfy the structural conditions for phenomenal consciousness; large language models are a special case. The argument uses diagonal-capability of the physical universe (weaker tha...

LLM-guided headline rewriting for clickability enhancement without clickbait

Optimizing news headlines for reader engagement is often conflated with clickbait, producing exaggerated or misleading phrasing that erodes editorial trust. We reframe clickbait not as a separate category but as the disproportionate amplification of otherwise legitimate engagement cues, and cast headline rewriting as a...

An explicit coefficient-module proof of Conjecture 3.4(a) in Au's Wilf-Zeilberger seeds paper

K. C. Au, Discovering hypergeometric series with harmonic numbers via Wilf-Zeilberger seeds, arXiv:2602.08721v3, conjectures in Conjecture 3.4(a) that the first summand S1(a,b,c,d,e;n) of Example II has coefficient spaces VN(S1) (the Q-span of the sequences n ↦ [ai1···ei5]S1 with total degree N) whose dimensions have g...

Webis-SR4ALL-26: A cross-disciplinary corpus of systematic reviews

Overview Webis-SR4ALL-26 is a large-scale corpus of 301,871 systematic reviews spanning 27 scientific disciplines. The corpus is derived from OpenAlex and enriched with structured methodological information extracted from full texts where available. The dataset is designed to support research on systematic review retri...

Investigating the Potential of Large Language Models to Support Qualitative Data Analysis - Replication Package

Replication Package: Investigating the Potential of Large Language Models to Support Qualitative Data Analysis This repository gathers the complete replication package and open data regarding the master's dissertation entitled: Investigating the Potential of Large Language Models to Support Qualitative Data Analysis. T...

Meta-Router di Supervisione UMOKWAI — Sistema a caricamento selettivo per la validazione delle bozze tecniche generate da LLM

# AbstractMeta-Router di Supervisione UMOKWAI — Sistema a caricamento selettivo per la validazione delle bozze tecniche generate da LLMIl presente lavoro introduce il Meta-Router di Supervisione, un sistema operativo modulare a caricamento selettivo progettato per controllare le risposte prodotte dal Router Ontologico...

Meta-Router di Supervisione UMOKWAI — Sistema a caricamento selettivo per la validazione delle bozze tecniche generate da LLM

# AbstractMeta-Router di Supervisione UMOKWAI — Sistema a caricamento selettivo per la validazione delle bozze tecniche generate da LLMIl presente lavoro introduce il Meta-Router di Supervisione, un sistema operativo modulare a caricamento selettivo progettato per controllare le risposte prodotte dal Router Ontologico...

Quantization Trade-offs for LLM Inference on Apple Silicon: A Practical Speed, Memory, and Accuracy Benchmark

Deploying large language models (LLMs) on consumer hardware typically requires quantization to reduce memory footprint and increase inference speed, but the practical trade-offs of this compression are not well characterized on Apple Silicon, where most existing benchmarks focus on NVIDIA GPU deployments. We present a...

Diagnostics of Engineering Systems using Large Language Models

Intelligent fault diagnosis underpins the safe and reliable operation of modern engineering systems. While large language models (LLMs) have advanced intelligent diagnostic technologies for industrial scenarios, their real-world deployment remains constrained by domain semantic gaps, insufficient physical fault knowled...

Quantization Trade-offs for LLM Inference on Apple Silicon: A Practical Speed, Memory, and Accuracy Benchmark

Deploying large language models (LLMs) on consumer hardware typically requires quantization to reduce memory footprint and increase inference speed, but the practical trade-offs of this compression are not well characterized on Apple Silicon, where most existing benchmarks focus on NVIDIA GPU deployments. We present a...