Metraipolis › archive › dc_2bc06c7b9e2e

QMUL_AK at AISA-ArabicFC: Schema-Constrained Function Calling with ALLaM — arXiv:2610.37995

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.37995
License (as archived): CC0-1.0
Archived: 2026-10-31T06:16:45.065Z
SHA-256 of the archived text: 4656d7ac4cf33f8991b50a95e4ed7aa4f47b515da7e28e30ef9d7f38f2191b0d
Stable URL of this page: https://metraipolis.com/documento/dc_2bc06c7b9e2e

How to cite

QMUL_AK at AISA-ArabicFC: Schema-Constrained Function Calling with ALLaM — arXiv:2610.37995 — arXiv (en), https://arxiv.org/abs/2610.37995 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2bc06c7b9e2e (Metraipolis archive dc_2bc06c7b9e2e, sha256 4656d7ac4cf33f8991b50a95e4ed7aa4f47b515da7e28e30ef9d7f38f2191b0d)

BibTeX

@misc{dc2bc06c7b9e2e,
  title = {QMUL_AK at AISA-ArabicFC: Schema-Constrained Function Calling with ALLaM — arXiv:2610.37995},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.37995},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2bc06c7b9e2e; sha256 4656d7ac4cf33f8991b50a95e4ed7aa4f47b515da7e28e30ef9d7f38f2191b0d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.37995. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We describe the QMUL_AK submission to AISA-ArabicFC 2026. The system adapts the Arabic-English ALLaM-7B-Instruct model to function calling with low-rank adaptation (LoRA). It combines an Arabic schema-aware prompt, dialect and no-call resampling, a token-weighted training objective, and deterministic validation of generated calls against the candidate tool schema. The same output supports Track A function calling and Track B reasoning-augmented calling. Training uses only the official task data. On the development set, the system achieves 0.9982 function accuracy and 0.8920 argument exact match, corresponding to overall A and B scores of 0.9345 and 0.9454. On the blind test set, the system reaches overall A and B scores of 0.9253 and 0.9380, ranking 3rd on both submission tracks and 1st of 21 teams on the Track C dialect-robustness diagnostic with a macro-averaged score of 0.9075 and the smallest dialect gap in the shared task (0.077). The results show that function selection is nearly saturated, while argument extraction remains the main source of error.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.37995 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-31T06:07:23.320Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).