small language model

Ontology-Guided Prompting for Zero-Shot Information Extraction: Data, Code, Annotations, and Evaluation Artifacts

This record contains the data, code, annotations, and evaluation artifacts for a controlled study of ontology-guided prompting in zero-shot information extraction. The experiment compares a generic extraction prompt with an ontology-guided variant using 20 synthetic single-entity texts constructed for four selected ont...

A BERT BiLSTM and BiTCN model for suicidal ideation identification in Spanish texts

This study proposes a computational model for detecting suicidal ideation in Spanish-language social media posts that combines Transformer-based representations with BiLSTM and BiTCN neural network layers to capture contextual and sequential linguistic patterns.

Why This Language? Historical Symmetry Breaking through Cultural Transmission: code, pre-registrations, results and manuscript

Code, pre-registrations with frozen thresholds, result summaries and manuscript for a minimal controlled model of cultural transmission in a Lewis referential game. Two small neural agents invent a language over 64 objects; each generation is a fresh pair that first learns a limited set of its parent's (object, message...

Evidence and Metadata Dataset for "Large Language Models for Small-Molecule Discovery: Model Taxonomy, Chemical Grounding, and Experimental Evidence

This dataset supports the review “Large Language Models for Small-Molecule Discovery: Model Taxonomy, Chemical Grounding, and Experimental Evidence”. It provides a structured, machine-readable evidence base for the literature synthesis, including: (i) a literature inventory and search/screening seed; (ii) model metadat...

Cell line screening of AI-generated drug repurposing predictions for NF2-related schwannomas

Non-NF2 schwannomatosis (SWN) is a rare syndrome that causes nerve sheath tumors with complex genetic variations and poorly understood modifiers. With no approved pharmacologic interventions, there is an urgent need for treatments that prevent or reduce tumor growth, or that ameliorate schwannoma-associated pain. Many...

The Truthometer: A Structured Cited Verdict, a Rulebook for the Claims No Register Settles, and a Distilled Judge

Paper 5 shows that for the commercial web, truth is a register join rather than a judgement, and that grounding in retrieved evidence is what makes verification work. This paper is the system that follows, and its contribution is threefold: the score, the rulebook, and the distillation boundary. **The Veracity Score.**...

The Truthometer: A Structured Cited Verdict, a Rulebook for the Claims No Register Settles, and a Distilled Judge

Paper 5 shows that for the commercial web, truth is a register join rather than a judgement, and that grounding in retrieved evidence is what makes verification work. This paper is the system that follows, and its contribution is threefold: the score, the rulebook, and the distillation boundary. **The Veracity Score.**...

Full-Layer Memory Injection via Compile-Time Key-Value Caches: An External-Knowledge Mechanism for Small-Parameter Language Models

Knowledge of large language models (LLMs) is frozen in parameters after pretraining and hard to update at low cost. This paper proposes a full-layer memory injection method based on compile-time key-value (KV) caches. Experiments on Qwen3-0.6B show full-layer token-level KV injection raises private-specification QA sco...

Evaluation of Vanilla and RAG-Supported LLM Comprehension of the EU AI Act

A comparative benchmarking study evaluating the regulatory comprehension of four open-source large language models, TinyLlama-1.1B-Chat, Gemma-2B-Instruct, Llama-3.1-8B-Instruct, and Mistral-7B-Instruct-v0.3, contributes an empirical characterization of small and medium language model regulatory comprehension and a reu...

Interpretable Subgroup Discovery with Abstention in Small, Heterogeneous Clinical Trials: A Retrospective Multi-Dataset Study

Small, heterogeneous clinical datasets pose a challenge for whole-cohort prediction because clinically meaningful treatment or response patterns may be diluted across biologically diverse patients. We describe and evaluate NetraAI, an interpretable dynamical-systems framework for selective subgroup discovery that uses...

Large language models in healthcare semantics

AI for Recruitment Administration

Measuring the waits a hiring process creates, and keeping the record it owes Four working days of internal approval are three weeks of silence on the other side, and almost every complaint about hiring is a complaint about that gap rather than about the decision at the end of it. Recruitment administration is a queue w...