Results show that fine-tuned lightweight LLM models outperform the earlier transformer-based NL2ASP approach and can offer a practical alternative to substantially larger models.
The central claim of this extended abstract is that, despite its lightweight syntax, an LLMASP-LGX configuration specifies a constructive interaction between LLM and ASP: facts extracted so far, together with facts derived by ASP, decide which LLM queries are admissible at each step, and LLM responses update the logica...
Mario Alviano, Lorenzo Grillo, N. Leone et al.· 0 citations
Answer Set Programming (ASP) and Large Language Models (LLMs) offer complementary strengths in symbolic reasoning and natural language understanding. In this work, we present an integrated framework for reliable structured knowledge extraction from natural language that combines the syntactic capabilities of LLMs with...
Mario Alviano, Matteo Capalbo, G. Gottlob et al.· Journal of Artificial Intell...· 0 citations
A neuro-symbolic, specification-guided decoding framework that combines LLM generation with symbolic reasoning, and presents a forward-only variant that guarantees specification satisfaction by construction, and a backtracking variant that revises earlier feasible choices when their downstream consequences leave only v...
Alessandro Previti, K. Vandikas, Mario Alviano et al.· 0 citations
A comparative study of Transformer-based and State-Space Model designs when it comes to long-context reasoning finds Transformer-based models have a pronounced tendency to forget distant information while StateSpace Model-based designs retain information more stably and yield better results for tasks that involve long...
Sheyasree Ghosh, Akanksha Srivastava, Soumen Bhowmik et al.· International Journal of Inn...· 0 citations
Rigorous evaluation of AI-assisted modeling remains an open challenge, particularly for low-resource domain-specific languages where training data is scarce and methodological correctness matters as much as syntactic validity. We present an evaluation framework and protocol-driven architecture for this setting, instant...
Sokhna Amar, M. Elaasar, Sambit Bhattacharya· Proceedings of the ACM/IEEE...· 0 citations
Neural combinatorial optimization (NCO) rapidly produces high-quality solutions for problem structures encountered during training, but operational requirements may change after deployment and be communicated in natural language. We propose LangNCO, a framework that compiles instance-specific requirements into structur...
Hironori Ohigashi· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.