E-CONAN (Entailment, CONtradition And Neutral) Diagnostics Dataset Investigating Linguistic Phenomena in Arabic Natural Language Understanding

arXiv:2609.33530 · cs.CL, cs.AI, cs.HC, cs.LG, cs.LO · Submitted 2026-09-27 · Read on arXiv

cs.CL, cs.AI, cs.HC, cs.LG, cs.LO

Submitted: 2026-09-27

Updated: 2026-09-27

Terminology

• Textual Entailment — 18× in this paper
• Alternation — 11× in this paper · explained in 1 episode
• Benchmark — 11× in this paper · explained in 2 episodes
• Natural Language Inference — 10× in this paper
• Core — 8× in this paper · explained in 2 episodes
• Recognizing Textual Entailment — 8× in this paper
• Natural Language Understanding — 7× in this paper
• Redundancy — 5× in this paper · explained in 2 episodes
• Generalization — 3× in this paper · explained in 13 episodes
• Hierarchical Framework — 3× in this paper · explained in 2 episodes
• Actor — 2× in this paper · explained in 3 episodes
• Adaptation — 2× in this paper · explained in 1 episode
• Benchmarking — 2× in this paper · explained in 3 episodes
• General Language Understanding Evaluation — 2× in this paper
• Natural Language Inference (NLI) — 2× in this paper · explained in 2 episodes
• Pragmatics — 2× in this paper · explained in 2 episodes
• Transformer — 2× in this paper · explained in 2 episodes
• Abstraction — 1× in this paper · explained in 1 episode
• Distillation — 1× in this paper · explained in 2 episodes
• Few-Shot Learning — 1× in this paper · explained in 4 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 6 episodes
• Mixture-of-Experts (MoE) — 1× in this paper · explained in 7 episodes
• RAG — 1× in this paper · explained in 3 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Representation Learning — 1× in this paper · explained in 2 episodes
• Retrieval Augmented Generation (RAG) — 1× in this paper · explained in 4 episodes
• Transfer Learning — 1× in this paper · explained in 10 episodes

Sources

• First Train to Generate, then Generate to Train: UnitedSynT5 for Few-Shot NLI
• Not another Negation Benchmark: The NaN-NLI Test Suite for Sub-clausal Negation
• Survey of NLU Benchmarks Diagnosing Linguistic Phenomena: Why not Standardize Diagnostics Benchmarks?
• Qwen2 Technical Report
• Qwen2.5 Technical Report
• Command R7B Arabic: A Small, Enterprise Focused, Multilingual, and Culturally Aware Arabic LLM
• DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering