Buchumschlag

A Necessary Step toward Faithfulness: Measuring and Improving Consistency in Free-Text Explanations

Fuente: arXiv

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	Zhao, Lingjun, Daumé III, Hal
Format:	Preprint
Veröffentlicht:	2025
Schlagworte:	Computation and Language Artificial Intelligence
Online-Zugang:	Acceder al recurso
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Ähnliche Einträge

Successfully Guiding Humans with Imperfect Instructions by Highlighting Potential Errors and Suggesting Corrections
von: Zhao, Lingjun, et al.
Veröffentlicht: (2024)

SMARTER: A Data-efficient Framework to Improve Toxicity Detection with Explanation via Self-augmenting Large Language Models
von: Nghiem, Huy, et al.
Veröffentlicht: (2025)

Pragmatics Meets Culture: Culturally-adapted Artwork Description Generation and Evaluation
von: Zhao, Lingjun, et al.
Veröffentlicht: (2026)

HateCOT: An Explanation-Enhanced Dataset for Generalizable Offensive Speech Detection via Large Language Models
von: Nghiem, Huy, et al.
Veröffentlicht: (2024)

Can Hallucination Correction Improve Video-Language Alignment?
von: Zhao, Lingjun, et al.
Veröffentlicht: (2025)

Steering Safely or Off a Cliff? Rethinking Specificity and Robustness in Inference-Time Interventions
von: Goyal, Navita, et al.
Veröffentlicht: (2026)

Language Models Predict Empathy Gaps Between Social In-groups and Out-groups
von: Hou, Yu, et al.
Veröffentlicht: (2025)

"You Gotta be a Doctor, Lin": An Investigation of Name-Based Bias of Large Language Models in Employment Recommendations
von: Nghiem, Huy, et al.
Veröffentlicht: (2024)

Do great minds think alike? Investigating Human-AI Complementarity in Question Answering with CAIMIRA
von: Gor, Maharshi, et al.
Veröffentlicht: (2024)

Toxicity Detection is NOT all you Need: Measuring the Gaps to Supporting Volunteer Content Moderators
von: Cao, Yang Trista, et al.
Veröffentlicht: (2023)

The Probabilities Also Matter: A More Faithful Metric for Faithfulness of Free-Text Explanations in Large Language Models
von: Siegel, Noah Y., et al.
Veröffentlicht: (2024)

When Stereotypes GTG: The Impact of Predictive Text Suggestions on Gender Bias in Human-AI Co-Writing
von: Baumler, Connor, et al.
Veröffentlicht: (2024)

Bias in the Tails: How Name-conditioned Evaluative Framing in Resume Summaries Destabilizes LLM-based Hiring
von: Nghiem, Huy, et al.
Veröffentlicht: (2026)

Reheat Nachos for Dinner? Evaluating AI Support for Cross-Cultural Communication of Neologisms
von: Ki, Dayeon, et al.
Veröffentlicht: (2026)

The Impact of Explanations on Fairness in Human-AI Decision-Making: Protected vs Proxy Features
von: Goyal, Navita, et al.
Veröffentlicht: (2023)

Balancing Safety and Helpfulness in Healthcare AI Assistants through Iterative Preference Alignment
von: Nghiem, Huy, et al.
Veröffentlicht: (2025)

Walk the Talk? Measuring the Faithfulness of Large Language Model Explanations
von: Matton, Katie, et al.
Veröffentlicht: (2025)

Local Explanations and Self-Explanations for Assessing Faithfulness in black-box LLMs
von: Fragkathoulas, Christos, et al.
Veröffentlicht: (2024)

Comparing Explanation Faithfulness between Multilingual and Monolingual Fine-tuned Language Models
von: Zhao, Zhixue, et al.
Veröffentlicht: (2024)

Faithfulness and the Notion of Adversarial Sensitivity in NLP Explanations
von: Manna, Supriya, et al.
Veröffentlicht: (2024)

Faithfulness Serum: Mitigating the Faithfulness Gap in Textual Explanations of LLM Decisions via Attribution Guidance
von: Alon, Bar, et al.
Veröffentlicht: (2026)

Improving Faithfulness of Large Language Models in Summarization via Sliding Generation and Self-Consistency
von: Li, Taiji, et al.
Veröffentlicht: (2024)

FaithLM: Towards Faithful Explanations for Large Language Models
von: Chuang, Yu-Neng, et al.
Veröffentlicht: (2024)

Call Me When Necessary: LLMs can Efficiently and Faithfully Reason over Structured Environments
von: Cheng, Sitao, et al.
Veröffentlicht: (2024)

Faithful and Robust LLM-Driven Theorem Proving for NLI Explanations
von: Quan, Xin, et al.
Veröffentlicht: (2025)

Faithful and Plausible Natural Language Explanations for Image Classification: A Pipeline Approach
von: Wojciechowski, Adam, et al.
Veröffentlicht: (2024)

Illocutionary Explanation Planning for Source-Faithful Explanations in Retrieval-Augmented Language Models
von: Sovrano, Francesco, et al.
Veröffentlicht: (2026)

CAuSE: Decoding Multimodal Classifiers using Faithful Natural Language Explanation
von: Bandyopadhyay, Dibyanayan, et al.
Veröffentlicht: (2025)

Incorporating Attribution Importance for Improving Faithfulness Metrics
von: Zhao, Zhixue, et al.
Veröffentlicht: (2023)

Towards Consistent Natural-Language Explanations via Explanation-Consistency Finetuning
von: Chen, Yanda, et al.
Veröffentlicht: (2024)

Towards Faithful Knowledge Graph Explanation Through Deep Alignment in Commonsense Question Answering
von: Zhai, Weihe, et al.
Veröffentlicht: (2023)

ASL STEM Wiki: Dataset and Benchmark for Interpreting STEM Articles
von: Yin, Kayo, et al.
Veröffentlicht: (2024)

VinaBench: Benchmark for Faithful and Consistent Visual Narratives
von: Gao, Silin, et al.
Veröffentlicht: (2025)

Generating Faithful Text From a Knowledge Graph with Noisy Reference Text
von: Hashem, Tahsina, et al.
Veröffentlicht: (2023)

GeoFaith: A Spatio-Temporal Dual View of Faithful Chain-of-Thought
von: Lv, Weijiang, et al.
Veröffentlicht: (2026)

FaithLens: Detecting and Explaining Faithfulness Hallucination
von: Si, Shuzheng, et al.
Veröffentlicht: (2025)

Learning from Sufficient Rationales: Analysing the Relationship Between Explanation Faithfulness and Token-level Regularisation Strategies
von: Kamp, Jonathan, et al.
Veröffentlicht: (2025)

Verbosity Tradeoffs and the Impact of Scale on the Faithfulness of LLM Self-Explanations
von: Siegel, Noah Y., et al.
Veröffentlicht: (2025)

MAMM-Refine: A Recipe for Improving Faithfulness in Generation with Multi-Agent Collaboration
von: Wan, David, et al.
Veröffentlicht: (2025)

PR-CAD: Progressive Refinement for Unified Controllable and Faithful Text-to-CAD Generation with Large Language Models
von: An, Jiyuan, et al.
Veröffentlicht: (2026)