Writerslogic at the CLEF 2026 SimpleText Track: Multi-Candidate LLM Simplification and Stacked Complexity Spotting

Abstract:We describe the Writerslogic team's participation in the CLEF 2026 SimpleText shared task, addressing Task 1 (text simplification) and Task 2 (complexity spotting). For Task 1, we develop a multi-candidate generation pipeline using GPT-4o-mini that produces five simplification candidates per sentence at varying temperatures, then selects the best candidate using a reference-free scoring heuristic that rewards compression, source word retention, Cochrane Plain Language Summary vocabulary usage, and lexical simplicity. On Task 1.1 (sentence-level simplification), our Claude Sonnet 4 submission achieves SARI 47.43 and BLEU 14.21, the top-ranked sentence-level system (3rd on the combined Task 1 leaderboard, behind two document-level submissions). For Task 2, we fine-tune a DeBERTa-v3-large NLI model on 350K labeled (source, sentence) pairs, framing hallucination detection as natural language inference. The model reads the most relevant source sentence as premise and the candidate as hypothesis, directly learning to distinguish grounded from hallucinated content. On Task 2.1 (binary overgeneration identification), our fine-tuned DeBERTa system achieves 0.8081 document-level macro F1 (0.8085 in our best ensemble), the top-ranked entry within the identification track and 2nd among teams overall, behind AIIR Lab (0.8197). On Task 2.2 (multi-class error classification), our best submission reaches 0.804 multiclass accuracy, ranking 2nd among unique teams behind AIIR Lab (0.827). We evaluate both tasks on English and multilingual biomedical text from Cochrane systematic reviews.

11 pages, 3 tables. Notebook for the SimpleText Lab at CLEF 2026. Code: this https URL

Computation and Language (cs.CL)

arXiv:2610.03567 [cs.CL]

(or arXiv:2610.03567v1 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2610.03567 Focus to learn more arXiv-issued DOI via DataCite (pending registration)

CLEF 2026 Working Notes, CEUR Workshop Proceedings, pp. 6068-6078

simpletext-clef2026/README.md

simpletext-clef2026/task1/batch_haiku.py

simpletext-clef2026/task1/batch_opus.py

simpletext-clef2026/task1/batch_sonnet.py

simpletext-clef2026/task1/ensemble_best.py

simpletext-clef2026/task1/ensemble_strategies.py

simpletext-clef2026/task1/eval_local.py

simpletext-clef2026/task1/eval_scraped.py

simpletext-clef2026/task1/pyproject.toml

simpletext-clef2026/task1/scrape_cochrane.py

simpletext-clef2026/task1/simplify.py

simpletext-clef2026/task1/simplify_ensemble.py

simpletext-clef2026/task1/simplify_local.py

simpletext-clef2026/task1/simplify_ollama.py

simpletext-clef2026/task1/simplify_pls_grounded.py

simpletext-clef2026/task1/simplify_pls_task12.py

simpletext-clef2026/task1/simplify_rag.py

simpletext-clef2026/task1/simplify_regex.py

simpletext-clef2026/task1/simplify_v2.py

simpletext-clef2026/task1/simplify_v3.py

simpletext-clef2026/task1/simplify_v4.py

simpletext-clef2026/task1/simplify_v5.py

simpletext-clef2026/task1/simplify_v6.py

simpletext-clef2026/task2/build_portfolio.py

simpletext-clef2026/task2/classify.py

simpletext-clef2026/task2/classify_v2.py

simpletext-clef2026/task2/classify_v3.py

simpletext-clef2026/task2/classify_v4.py

simpletext-clef2026/task2/feature_extraction.py

simpletext-clef2026/task2/llm_grounding.py

simpletext-clef2026/task2/llm_grounding_v2.py

simpletext-clef2026/task2/main.py

simpletext-clef2026/task2/modal_finetune_deberta.py

simpletext-clef2026/task2/modal_finetune_shidey.py

simpletext-clef2026/task2/modal_nli_inference.py

simpletext-clef2026/task2/predict.py

simpletext-clef2026/task2/predict_v6.py

simpletext-clef2026/task2/predict_v7.py

simpletext-clef2026/task2/pyproject.toml

simpletext-clef2026/task2/train.py

simpletext-clef2026/task2/train_v7.py

(38 additional files not shown) You must enabled JavaScript to view entire file list.

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.