Image de couverture de livre

Lost-in-the-Middle in Long-Text Generation: Synthetic Dataset, Evaluation Framework, and Mitigation

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Zhang, Junhao, Zhang, Richong, Kong, Fanshuang, Miao, Ziyang, Ye, Yanhan, Zheng, Yaowei
Format:	Preprint
Publié:	2025
Sujets:	Computation and Language Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models
par: Zheng, Yaowei, et autres
Publié: (2024)

EasyRAG: Efficient Retrieval-Augmented Generation Framework for Automated Network Operations
par: Feng, Zhangchi, et autres
Publié: (2024)

LH-Mix: Local Hierarchy Correlation Guided Mixup over Hierarchical Prompt Tuning
par: Kong, Fanshuang, et autres
Publié: (2024)

Activated Parameter Locating via Causal Intervention for Model Merging
par: Kong, Fanshuang, et autres
Publié: (2024)

What Works for 'Lost-in-the-Middle' in LLMs? A Study on GM-Extract and Mitigations
par: Gupte, Mihir, et autres
Publié: (2025)

Never Lost in the Middle: Mastering Long-Context Question Answering with Position-Agnostic Decompositional Training
par: He, Junqing, et autres
Publié: (2023)

What Has Been Lost with Synthetic Evaluation?
par: Gill, Alexander, et autres
Publié: (2025)

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

Dynamic Task Vector Grouping for Efficient Multi-Task Prompt Tuning
par: Zhang, Pieyi, et autres
Publié: (2025)

Easy Dataset: A Unified and Extensible Framework for Synthesizing LLM Fine-Tuning Data from Unstructured Documents
par: Miao, Ziyang, et autres
Publié: (2025)

Generating Synthetic Datasets for Few-shot Prompt Tuning
par: Guo, Xu, et autres
Publié: (2024)

CoRect: Context-Aware Logit Contrast for Hidden State Rectification to Resolve Knowledge Conflicts
par: Ma, Xuhua, et autres
Publié: (2026)

MMGenBench: Fully Automatically Evaluating LMMs from the Text-to-Image Generation Perspective
par: Huang, Hailang, et autres
Publié: (2024)

Lost in Stories: Consistency Bugs in Long Story Generation by LLMs
par: Li, Junjie, et autres
Publié: (2026)

Measuring Diversity in Synthetic Datasets
par: Zhu, Yuchang, et autres
Publié: (2025)

Lost in Translation: Latent Concept Misalignment in Text-to-Image Diffusion Models
par: Zhao, Juntu, et autres
Publié: (2024)

PPTAgent: Generating and Evaluating Presentations Beyond Text-to-Slides
par: Zheng, Hao, et autres
Publié: (2025)

Code-Style In-Context Learning for Knowledge-Based Question Answering
par: Nie, Zhijie, et autres
Publié: (2023)

PROXYQA: An Alternative Framework for Evaluating Long-Form Text Generation with Large Language Models
par: Tan, Haochen, et autres
Publié: (2024)

Your Language Model Can Secretly Write Like Humans: Contrastive Paraphrase Attacks on LLM-Generated Text Detectors
par: Fang, Hao, et autres
Publié: (2025)

Parameterized Synthetic Text Generation with SimpleStories
par: Finke, Lennart, et autres
Publié: (2025)

When Text Embedding Meets Large Language Model: A Comprehensive Survey
par: Nie, Zhijie, et autres
Publié: (2024)

inversedMixup: Data Augmentation via Inverting Mixed Embeddings
par: Kong, Fanshuang, et autres
Publié: (2026)

LongBench Pro: A More Realistic and Comprehensive Bilingual Long-Context Evaluation Benchmark
par: Chen, Ziyang, et autres
Publié: (2026)

DeltaKV: Residual-Based KV Cache Compression via Long-Range Similarity
par: Hao, Jitai, et autres
Publié: (2026)

Bayesian WeakS-to-Strong from Text Classification to Generation
par: Cui, Ziyun, et autres
Publié: (2024)

Style Transfer as Bias Mitigation: Diffusion Models for Synthetic Mental Health Text for Arabic
par: Mankarious, Saad, et autres
Publié: (2026)

Lost in the Middle at Birth: An Exact Theory of Transformer Position Bias
par: Chowdhury, Borun D
Publié: (2026)

Modular Techniques for Synthetic Long-Context Data Generation in Language Model Training and Evaluation
par: Subramanian, Seganrasan, et autres
Publié: (2025)

Equilibrium Dynamics and Mitigation of Gender Bias in Synthetically Generated Data
par: Kattamuri, Ashish, et autres
Publié: (2025)

Cooperative Memory Paging with Keyword Bookmarks for Long-Horizon LLM Conversations
par: Liu, Ziyang
Publié: (2026)

Extract Information from Hybrid Long Documents Leveraging LLMs: A Framework and Dataset
par: Yue, Chongjian, et autres
Publié: (2024)

Synthetic Dialogue Dataset Generation using LLM Agents
par: Abdullin, Yelaman, et autres
Publié: (2024)

ChallengeMe: An Adversarial Learning-enabled Text Summarization Framework
par: Deng, Xiaoyu, et autres
Publié: (2025)

FinRpt: Dataset, Evaluation System and LLM-based Multi-agent Framework for Equity Research Report Generation
par: Jin, Song, et autres
Publié: (2025)

ToxiCraft: A Novel Framework for Synthetic Generation of Harmful Information
par: Hui, Zheng, et autres
Publié: (2024)

Lost in the Source Language: How Large Language Models Evaluate the Quality of Machine Translation
par: Huang, Xu, et autres
Publié: (2024)

Generation of Synthetic Clinical Text: A Systematic Review
par: Alshaikhdeeb, Basel, et autres
Publié: (2025)

Reasoning-Enhanced Self-Training for Long-Form Personalized Text Generation
par: Salemi, Alireza, et autres
Publié: (2025)

Do LLMs Align with My Task? Evaluating Text-to-SQL via Dataset Alignment
par: Rafiei, Davood, et autres
Publié: (2025)