# Where is the answer? Investigating Positional Bias in Language Model Knowledge Extraction

arXiv: https://arxiv.org/abs/2402.12170 · HTML: https://mlsanity.com/paper/2402.12170/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 35 of its 46 references that are in the map.
- First posted on arXiv 16 Feb 2024.

## Builds on (most cited first, 30 of 35)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Finetuned Language Models Are Zero-Shot Learners](https://mlsanity.com/paper/2109.01652/) (2021)
- [Lost in the Middle: How Language Models Use Long Contexts](https://mlsanity.com/paper/2307.03172/) (2023)
- [Locating and Editing Factual Associations in GPT](https://mlsanity.com/paper/2202.05262/) (2022)
- [The Pile: An 800GB Dataset of Diverse Text for Language Modeling](https://mlsanity.com/paper/2101.00027/) (2020)
- [G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment](https://mlsanity.com/paper/2303.16634/) (2023)
- [HuggingFace's Transformers: State-of-the-art Natural Language Processing](https://mlsanity.com/paper/1910.03771/) (2019)
- [REALM: Retrieval-Augmented Language Model Pre-Training](https://mlsanity.com/paper/2002.08909/) (2020)
- [BLOOM: A 176B-Parameter Open-Access Multilingual Language Model](https://mlsanity.com/paper/2211.05100/) (2022)
- [LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day](https://mlsanity.com/paper/2306.00890/) (2023)
- [Mass-Editing Memory in a Transformer](https://mlsanity.com/paper/2210.07229/) (2022)
- [Decoupling Representation and Classifier for Long-Tailed Recognition](https://mlsanity.com/paper/1910.09217/) (2019)
- [The RefinedWeb Dataset for Falcon LLM: Outperforming Curated Corpora with Web Data, and Web Data Only](https://mlsanity.com/paper/2306.01116/) (2023)
- [Fast Model Editing at Scale](https://mlsanity.com/paper/2110.11309/) (2021)
- [LongLLMLingua: Accelerating and Enhancing LLMs in Long Context Scenarios via Prompt Compression](https://mlsanity.com/paper/2310.06839/) (2023)
- [Memory-Based Model Editing at Scale](https://mlsanity.com/paper/2206.06520/) (2022)
- [MedAlpaca -- An Open-Source Collection of Medical Conversational AI Models and Training Data](https://mlsanity.com/paper/2304.08247/) (2023)
- [Physics of Language Models: Part 3.1, Knowledge Storage and Extraction](https://mlsanity.com/paper/2309.14316/) (2023)
- [Knowledge Editing for Large Language Models: A Survey](https://mlsanity.com/paper/2310.16218/) (2023)
- [Towards Continual Knowledge Learning of Language Models](https://mlsanity.com/paper/2110.03215/) (2021)
- [What Language Model Architecture and Pretraining Objective Work Best for Zero-Shot Generalization?](https://mlsanity.com/paper/2204.05832/) (2022)
- [A Question-Entailment Approach to Question Answering](https://mlsanity.com/paper/1901.08079/) (2019)
- [Attention Sorting Combats Recency Bias In Long Context Language Models](https://mlsanity.com/paper/2310.01427/) (2023)
- [Look at the First Sentence: Position Bias in Question Answering](https://mlsanity.com/paper/2004.14602/) (2020)
- [Training Trajectories of Language Models Across Scales](https://mlsanity.com/paper/2212.09803/) (2022)

## Built on by (most cited first, 6 of 6)

- [Self-Tuning: Instructing LLMs to Effectively Acquire New Knowledge through Self-Teaching](https://mlsanity.com/paper/2406.06326/) (2024)
- ["Lost-in-the-Later": Framework for Quantifying Contextual Grounding in Large Language Models](https://mlsanity.com/paper/2507.05424/) (2025)
- [Enhancing elusive clues in knowledge learning by contrasting attention of language models](https://mlsanity.com/paper/2409.17954/) (2024)
- [Review of Hallucination Understanding in Large Language and Vision Models](https://mlsanity.com/paper/2510.00034/) (2025)
- [Self-Foveate: Enhancing Diversity and Difficulty of Synthesized Instructions from Unsupervised Text via Multi-Level Foveation](https://mlsanity.com/paper/2507.23440/) (2025)
- [Can LLMs Ask Good Questions?](https://mlsanity.com/paper/2501.03491/) (2025)

## Abstract (arXiv)

Large language models require updates to remain up-to-date or adapt to new domains by fine-tuning them with new documents. One key is memorizing the latest information in a way that the memorized information is extractable with a query prompt. However, LLMs suffer from a phenomenon called perplexity curse; despite minimizing document perplexity during fine-tuning, LLMs struggle to extract information through a prompt sentence. In this new knowledge acquisition and extraction, we find a very intriguing fact that LLMs can accurately answer questions about the first sentence, but they struggle to extract information described in the middle or end of the documents used for fine-tuning. Our study suggests that the auto-regressive training causes this issue; each token is prompted by reliance on all previous tokens, which hinders the model from recalling information from training documents by question prompts. To conduct the in-depth study, we publish both synthetic and real datasets, enabling the evaluation of the QA performance w.r.t. the position of the corresponding answer in a document. Our investigation shows that even a large model suffers from the perplexity curse, but regularization such as denoising auto-regressive loss can enhance the information extraction from diverse positions. These findings will be (i) a key to improving knowledge extraction from LLMs and (ii) new elements to discuss the trade-off between RAG and fine-tuning in adapting LLMs to a new domain.