AILS-NTUA at SemEval-2026 Task 12: Graph-Based Retrieval and Reflective Prompting for Abductive Event Reasoning
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914368719945728 |
|---|---|
| author | Karafyllis, Nikolas Lymperaiou, Maria Filandrianos, Giorgos Voulodimos, Athanasios Stamou, Giorgos |
| author_facet | Karafyllis, Nikolas Lymperaiou, Maria Filandrianos, Giorgos Voulodimos, Athanasios Stamou, Giorgos |
| contents | We present a winning three-stage system for SemEval 2026 Task~12: Abductive Event Reasoning that combines graph-based retrieval, LLM-driven abductive reasoning with prompt design optimized through reflective prompt evolution, and post-hoc consistency enforcement; our system ranks first on the evaluation-phase leaderboard with an accuracy score of 0.95. Cross-model error analysis across 14 models (7~families) reveals three shared inductive biases: causal chain incompleteness, proximate cause preference, and salience bias, whose cross-family convergence (51\% cause-count reduction) indicates systematic rather than model-specific failure modes in multi-label causal reasoning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_04319 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | AILS-NTUA at SemEval-2026 Task 12: Graph-Based Retrieval and Reflective Prompting for Abductive Event Reasoning Karafyllis, Nikolas Lymperaiou, Maria Filandrianos, Giorgos Voulodimos, Athanasios Stamou, Giorgos Computation and Language We present a winning three-stage system for SemEval 2026 Task~12: Abductive Event Reasoning that combines graph-based retrieval, LLM-driven abductive reasoning with prompt design optimized through reflective prompt evolution, and post-hoc consistency enforcement; our system ranks first on the evaluation-phase leaderboard with an accuracy score of 0.95. Cross-model error analysis across 14 models (7~families) reveals three shared inductive biases: causal chain incompleteness, proximate cause preference, and salience bias, whose cross-family convergence (51\% cause-count reduction) indicates systematic rather than model-specific failure modes in multi-label causal reasoning. |
| title | AILS-NTUA at SemEval-2026 Task 12: Graph-Based Retrieval and Reflective Prompting for Abductive Event Reasoning |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2603.04319 |