NaCTeM at EMNLP 2026

2026-08-25

This year, NaCTeM has five papers accepted at EMNLP 2026, four at the Main Conference and one in Findings:

• Reasoning Confidence Is Not Policy-Invariant: Calibrating Prefix Success under Model and Budget Shift (Main)
• XCR-Bench: Benchmarking Cross-Cultural Reasoning in LLMs via Culture-Specific Items and Hall's Triad (Main)
• DITING: Evaluating Sentence-Level Cultural Meaning in Web Novel Translation (Main)
• SiMing-Bench: Evaluating Procedural Correctness from Continuous Interactions in Clinical Skill Videos (Main)
• Bias Before the Model: Detecting and Removing Cognitive Bias Triggers in Financial Text (Findings)

We are also chairing FinNLP 2026, the 11th Workshop on Financial Technology and Natural Language Processing, on Wednesday 28 October in Budapest, invited talks, oral and poster sessions, the FinEval shared task and a panel discussion.

Work has been done in collaboration with The Fin AI, ACL SIG-FinTech and CLAIN at Wuhan University.

See you in Budapest, 24–29 October!