Image de couverture de livre

When Answers Stray from Questions: Hallucination Detection via Question-Answer Orthogonal Decomposition

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Yao, Siyang, Feng, Erhu, Xia, Yubin
Format:	Preprint
Publié:	2026
Sujets:	Machine Learning Artificial Intelligence Computation and Language
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

MultiQ&A: An Analysis in Measuring Robustness via Automated Crowdsourcing of Question Perturbations and Answers
par: Cho, Nicole, et autres
Publié: (2025)

Measuring and Reducing LLM Hallucination without Gold-Standard Answers
par: Wei, Jiaheng, et autres
Publié: (2024)

Anchored Answers: Unravelling Positional Bias in GPT-2's Multiple-Choice Questions
par: Li, Ruizhe, et autres
Publié: (2024)

The Challenge of Achieving Attributability in Multilingual Table-to-Text Generation with Question-Answer Blueprints
par: Haussmann, Aden
Publié: (2025)

Explicit Diversity Conditions for Effective Question Answer Generation with Large Language Models
par: Yadav, Vikas, et autres
Publié: (2024)

Multilingual Non-Factoid Question Answering with Answer Paragraph Selection
par: Mishra, Ritwik, et autres
Publié: (2024)

MeDiSumQA: Patient-Oriented Question-Answer Generation from Discharge Letters
par: Dada, Amin, et autres
Publié: (2025)

AQUA-Bench: Beyond Finding Answers to Knowing When There Are None in Audio Question Answering
par: Kuan, Chun-Yi, et autres
Publié: (2026)

Hallucination-Free Automatic Question & Answer Generation for Intuitive Learning
par: Wang, Nicholas X., et autres
Publié: (2026)

Rewarding Intellectual Humility Learning When Not To Answer In Large Language Models
par: Jha, Abha, et autres
Publié: (2026)

DefAn: Definitive Answer Dataset for LLMs Hallucination Evaluation
par: Rahman, A B M Ashikur, et autres
Publié: (2024)

Verif.ai: Towards an Open-Source Scientific Generative Question-Answering System with Referenced and Verifiable Answers
par: Košprdić, Miloš, et autres
Publié: (2024)

When Does a Language Model Commit? A Finite-Answer Theory of Pre-Verbalization Commitment
par: Zhang, Long, et autres
Publié: (2026)

Answer, Assemble, Ace: Understanding How LMs Answer Multiple Choice Questions
par: Wiegreffe, Sarah, et autres
Publié: (2024)

No Answer Needed: Predicting LLM Answer Accuracy from Question-Only Linear Probes
par: Cencerrado, Iván Vicente Moreno, et autres
Publié: (2025)

From Answers to Questions: EQGBench for Evaluating LLMs' Educational Question Generation
par: Zhou, Chengliang, et autres
Publié: (2025)

Proving that Cryptic Crossword Clue Answers are Correct
par: Andrews, Martin, et autres
Publié: (2024)

ExpertQA: Expert-Curated Questions and Attributed Answers
par: Malaviya, Chaitanya, et autres
Publié: (2023)

The Strongest Teacher Is Not Always the Best Teacher: Student-Centric Answer Selection
par: Hu, Zhengyu, et autres
Publié: (2026)

General Table Question Answering via Answer-Formula Joint Generation
par: Wang, Zhongyuan, et autres
Publié: (2025)

Graph Guided Question Answer Generation for Procedural Question-Answering
par: Pham, Hai X., et autres
Publié: (2024)

Augmenting Math Word Problems via Iterative Question Composing
par: Liu, Haoxiong, et autres
Publié: (2024)

Question Answering with LLMs and Learning from Answer Sets
par: Borroto, Manuel, et autres
Publié: (2025)

Clarify, Abstain or Answer? Strategising in Conversation with Belief-Augmented Generation
par: Baan, Joris, et autres
Publié: (2026)

Answer Matching Outperforms Multiple Choice for Language Model Evaluation
par: Chandak, Nikhil, et autres
Publié: (2025)

SelfReflect: Can LLMs Communicate Their Internal Answer Distribution?
par: Kirchhof, Michael, et autres
Publié: (2025)

ASAG2024: A Combined Benchmark for Short Answer Grading
par: Meyer, Gérôme, et autres
Publié: (2024)

When Single Answer Is Not Enough: Rethinking Single-Step Retrosynthesis Benchmarks for LLMs
par: Zagribelnyy, Bogdan, et autres
Publié: (2026)

Retrieval Augmented Question Answering: When Should LLMs Admit Ignorance?
par: Wang, Dingmin, et autres
Publié: (2025)

Putting People in LLMs' Shoes: Generating Better Answers via Question Rewriter
par: Chen, Junhao, et autres
Publié: (2024)

Inference-Time Intervention: Eliciting Truthful Answers from a Language Model
par: Li, Kenneth, et autres
Publié: (2023)

The Detection-Extraction Gap: Models Know the Answer Before They Can Say It
par: Wang, Hanyang, et autres
Publié: (2026)

Model Internals-based Answer Attribution for Trustworthy Retrieval-Augmented Generation
par: Qi, Jirui, et autres
Publié: (2024)

Train Once, Answer All: Many Pretraining Experiments for the Cost of One
par: Bordt, Sebastian, et autres
Publié: (2025)

Crystal-KV: Efficient KV Cache Management for Chain-of-Thought LLMs via Answer-First Principle
par: Wang, Zihan, et autres
Publié: (2026)

DiffuSpeech: Silent Thought, Spoken Answer via Unified Speech-Text Diffusion
par: Lou, Yuxuan, et autres
Publié: (2026)

Clinical QA 2.0: Multi-Task Learning for Answer Extraction and Categorization
par: Pattnayak, Priyaranjan, et autres
Publié: (2025)

ProfBench: Multi-Domain Rubrics requiring Professional Knowledge to Answer and Judge
par: Wang, Zhilin, et autres
Publié: (2025)

Seeing Isn't Knowing: Do VLMs Know When Not to Answer Spatial Questions (and Why)?
par: Zhang, Yue, et autres
Publié: (2026)

Grounded Cache Routing for Retrieval-Augmented Generation: When Is It Safe to Reuse an Answer?
par: Shah, Syed Huma
Publié: (2026)