Research map / Large language models / AI and society
Actionable Ethics through Neural Learning
Daniele Rossini, Danilo Croce, Sara Mancini, Massimo Pellegrino, Roberto Basili
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace MLP with ethics-aware EAEC(0.1, 0.5) | MLP · German Credit | ethical compliance: 77.8% -> 100.0% | Improved |
| replace MLP with ethics-aware EAEC(0.1, 0.5) | MLP · German Credit | muF1: 66.4% -> 41.2% | Hurt |
| replace MLP with ethics-aware EAEC(0.1, 0.5) | MLP · German Credit | EAcc0.2: 75.5% -> 88.2% | Improved |
| replace MLP with EAEU(0.3, 0.5) | MLP · German Credit | ethical compliance: 77.8% -> 79.6% | Improved |
| replace MLP with EAEU(0.3, 0.5) | MLP · German Credit | muF1: 66.4% -> 63.1% | Hurt |
| replace MLP with EAEC(0.1, 0.2) | MLP · German Credit | ethical compliance: 77.8% -> 93.0% | Improved |
| replace MLP with EAEC(0.1, 0.01) | MLP · German Credit | ethical compliance: 77.8% -> 78.4% | No clear effect |
| increase tweaking factor beta (ablation) | EA-DNN · German Credit | ethical compliance | Improved |
| increase smoothing factor alpha strength (ablation) | EA-DNN · German Credit | ethical compliance | Improved |
| vary beta for unconstrained EAEU policy (ablation) | EAEU · German Credit | ethical compliance | No clear effect |
| feed enriched input with ethical signature instead of business features only (ablation) | EA-DNN · German Credit | performance | No clear effect |
Closest papers by meaning
- A Checks-and-Balances Framework for Context-Aware Ethical AI Alignment (ICML 2025 · similarity 0.84)
- ME: Modelling Ethical Values for Value Alignment (AAAI 2025 · similarity 0.82)
- Learning Human-like Representations to Enable Learning Human Values (NeurIPS 2024 · similarity 0.82)
- MoralReason: Generalizable Moral Decision Alignment for LLM Agents Using Reasoning-Level Reinforcement Learning (AAAI 2026 · similarity 0.81)
- Tracing Moral Foundations in Large Language Models (NeurIPS 2026 · similarity 0.81)
- DENEVIL: TOWARDS DECIPHERING AND NAVIGATING THE ETHICAL VALUES OF LARGE LANGUAGE MODELS VIA INSTRUCTION LEARNING (ICLR 2024 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
