Tag
ReacTOD proposes a bounded neuro-symbolic architecture for zero-shot dialogue state tracking, using a self-correcting ReAct loop with deterministic validation. It achieves state-of-the-art results on MultiWOZ and Schema-Guided Dialogue benchmarks, improving joint goal accuracy by up to 14 percentage points.
Introduces ANNEAL, a neuro-symbolic agent that converts recurring failures into governed symbolic edits of a process knowledge graph without modifying model weights, achieving persistent structural repairs and eliminating recurring failures in tested settings.
Proposes Logic-GNN, a neuro-symbolic framework that uses temporal graph neural networks and graph Kolmogorov complexity to induce a symbolic grammar for clinical records, enabling detection and correction of data entry errors as grammatical violations. The system achieves an F1-score of 0.94 on a large healthcare dataset, outperforming state-of-the-art methods by 12%.
This paper empirically studies LLMs' legal reasoning in tax law, showing that data contamination inflates performance and that neuro-symbolic hybrid systems offer more reliable and robust generalization than monolithic LLMs.
This paper presents NSPI, a neuro-symbolic framework that combines LLMs and symbolic computation to prove polynomial inequalities. It uses LLM-generated sum-of-squares conjectures, refines them symbolically, and formally verifies the proofs in Lean, demonstrating scalability on polynomials with up to 10 variables.
This paper identifies a systematic gap between legal interpretation and formal logic in AI legal reasoning, proposes a neuro-symbolic approach to bridge it, and demonstrates substantial label shifts when re-annotating legal NLI data under strict formal entailment.
This paper introduces a neural network architecture that learns lifted action schemas from fully observed state traces with unobserved action arguments, aiming to enable robust learning of planning domains for neuro-symbolic models.
Introduces Ethical Immanence, a new AI alignment paradigm that embeds ethical behavior into model architecture via loss function regularization and metacognitive detection, promising lower costs and inherent stability for open-source LLMs.
ReaComp compiles LLM reasoning traces into reusable symbolic program synthesizers that achieve strong accuracy on program synthesis benchmarks while eliminating LLM calls at test time, significantly reducing computational cost.
TabularMath introduces a benchmark and AutoT2T framework for evaluating LLMs' mathematical reasoning over tabular data, revealing that table complexity, data quality, and modality significantly impact model performance. The study addresses a gap in LLM evaluation by systematically assessing robustness to incomplete or inconsistent table information in real-world scenarios.