analiza aiqbench
This research introduces Pingala, a decoding strategy specifically engineered to generate semantically coherent Sanskrit poetry while strictly adhering to complex prosodic constraints like syllable weight patterns. By segmenting verses into grouped lines and utilizing phonetically aware transliteration (SLP1), the method significantly improves metrical alignment without sacrificing semantic quality in instruction-tuned models such as Phi-4. The study also proposes a novel reference-free evaluation framework using cross-encoders to better assess generated poetry instances against true literary standards.
Pingala: Prosody-Aware Decoding for Sanskrit Poetry Generation
arXiv:2603.24413v2 Announce Type: replace Abstract: Poetry generation in Sanskrit typically requires the verse to be semantically coherent and adhere to strict prosodic rules. In Sanskrit prosody, every line of a verse is typically a fixed length sequence of syllables adhering to prescribed binary patterns of syllable weights. We observe that instead of treating a verse as a monolithic sequence, segmenting them as grouped-lines leads to significant improvement in semantic coherence by 10\% with comparable metrical adherence. Specifically, Pingala, our proposed decoding approach is designed to encourage every line to have well-formed words and our token selection biases the model towards it by preferring longer tokens. Writing in Sanskrit follows phonemic orthography, hence using a phonetically aware transliteration scheme, SLP1, increased the metrical alignment by 46\% with comparable semantic similarity, for a instruction fine-tuned large language models like Phi-4. We also introduce a new approach for reference-free evaluation using cross-encoders which achieved better alignment with true poetry instances.
Abstract:Poetry generation in Sanskrit typically requires the verse to be semantically coherent and adhere to strict prosodic rules. In Sanskrit prosody, every line of a verse is typically a fixed length sequence of syllables adhering to prescribed binary patterns of syllable weights. We observe that instead of treating a verse as a monolithic sequence, segmenting them as grouped-lines leads to significant improvement in semantic coherence by 10\% with comparable metrical adherence. Specifically, Pingala, our proposed decoding approach is designed to encourage every line to have well-formed words and our token selection biases the model towards it by preferring longer tokens. Writing in Sanskrit follows phonemic orthography, hence using a phonetically aware transliteration scheme, SLP1, increased the metrical alignment by 46\% with comparable semantic similarity, for a instruction fine-tuned large language models like Phi-4. We also introduce a new approach for reference-free evaluation using cross-encoders which achieved better alignment with true poetry instances.
Computation and Language (cs.CL)
arXiv:2603.24413 [cs.CL]
(or arXiv:2603.24413v2 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2603.24413 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 245 cuvinte
Subiecte conexe
J++ Lens: Jacobian Filtering Enables More Faithful Workspace Lenses
LessWrong · 8 oct. 2026
GEPARD - Generative, Prosody-aware, Autoregressive text-to-speech model for Realtime Dialogue
arXiv · 7 sept. 2026
How to Build In-Vehicle AI Agents with NVIDIA: From Cloud to Car
NVIDIA · 5 mai 2026
Computational Social Linguistics for Telugu Cultural Preservation: Novel Algorithms for Chandassu Metrical Pattern Recognition
arXiv · 14 oct. 2026