Study Reveals LLMs Struggle to Faithfully Transform Persian Proverbs into Morally Accurate Stories
Researchers introduced a new dataset and task called constrained semantic decompression to evaluate how well large language models can transform abstract Persian proverbs into narratively engaging stories that preserve moral meaning. The study found that current LLMs often produce fluent text but fail to capture the underlying moral and causal structure of the proverbs. The findings suggest that explicit reasoning and iterative refinement can help, indicating the gap stems from difficulty translating abstract meaning rather than lack of knowledge.
Researchers have developed a new evaluation framework to test how well large language models can transform dense, abstract proverbs into engaging narratives while preserving their moral content. The study, which focuses on Persian proverbs, introduces the Proverb Aligned Narrative Dataset (PAND) containing proverbs paired with human-written stories and explicit meanings. Using a hybrid evaluation approach combining human-calibrated LLM-as-a-Judge assessment with structural metrics, the researchers analyzed model performance across different prompting strategies. They identified a persistent "decompression gap" where models achieve strong surface-level fluency but fail to faithfully instantiate the underlying moral and causal logic embedded in proverbs. The research demonstrates that explicit reasoning prompts and iterative refinement can partially address these failures, suggesting the problem lies in translating abstract meaning into narrative form rather than fundamental knowledge deficits. The authors propose this task framework could extend to evaluating how LLMs handle other forms of compressed cultural knowledge.
What different sources said
- arXiv cs.CLCenter
Constrained Semantic Decompression in LLMs through Persian Proverb-Conditioned Story Generation
Related
Topology-Aware Thermodynamics Improves DNA Probe Specificity Design
Researchers developed a new framework for designing DNA probes that accounts for the spatial organization of matched sequences, not just overall thermodynamic stability. Traditional methods rely on scalar measures like melting temperature and free energy, which miss how mismatches are distributed along the probe. The approach could improve diagnostic accuracy in applications like HPV detection and gene expression profiling.
Study Identifies Optimal Thermal Dose for Combining Focused Ultrasound with Immunotherapy in Tumors
Researchers used multimodal PET imaging to identify an optimal thermal dose range for focused ultrasound ablation that destroys tumor tissue while preserving conditions for immunotherapy delivery. The study found that excessive heating collapses blood vessels needed for antibody access, while insufficient heating fails to adequately reduce tumor burden. The findings could guide clinical design of combination treatments pairing thermal ablation with immunotherapies.
Plant MSH1 Protein Functions as Mismatch-Directed Nuclease for Organelle Genome Maintenance
Researchers have identified the precise mechanism by which the AtMSH1 protein in Arabidopsis plants recognizes and cleaves DNA mismatches and lesions, preventing mutations in organellar genomes. The protein combines a DNA mismatch recognition module with a nuclease domain that makes staggered cuts at specific positions relative to DNA damage. This discovery explains how plants maintain unusually low mutation rates in their mitochondrial and chloroplast DNA compared to other eukaryotes.