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

• URL: http://arxiv.org/abs/2402.12170v2
• Date: Thu, 23 May 2024 06:52:31 GMT
• Title: Where is the answer? Investigating Positional Bias in Language Model Knowledge Extraction
• Authors: Kuniaki Saito, Kihyuk Sohn, Chen-Yu Lee, Yoshitaka Ushiku,
• Abstract summary: Large language models require updates to remain up-to-date or adapt to new domains. One key is memorizing the latest information in a way that the memorized information is extractable with a query prompt. Despite minimizing document perplexity during fine-tuning, LLMs struggle to extract information through a prompt sentence.
• Score: 36.40833517478628
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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.

Related papers

• Best Practices for Distilling Large Language Models into BERT for Web Search Ranking [14.550458167328497]
  Large Language Models (LLMs) can generate a ranked list of potential documents. We transfer the ranking expertise of LLMs to a more compact model like BERT, using a ranking loss to enable the deployment of less resource-intensive models. Our model has been successfully integrated into a commercial web search engine as of February 2024.
  arXiv  Detail & Related papers  (2024-11-07T08:54:46Z)
• Utilize the Flow before Stepping into the Same River Twice: Certainty Represented Knowledge Flow for Refusal-Aware Instruction Tuning [68.57166425493283]
  We introduce Certainty Represented Knowledge Flow for Refusal-Aware Instructions Construction (CRaFT) CRaFT incorporates response certainty to selectively filter and modify data, reducing static conflicts. We conducted extensive experiments on open-ended question answering and multiple-choice question task.
  arXiv  Detail & Related papers  (2024-10-09T14:12:51Z)
• Peering into the Mind of Language Models: An Approach for Attribution in Contextual Question Answering [9.86691461253151]
  We introduce a novel method for attribution in contextual question answering, leveraging the hidden state representations of large language models (LLMs) Our approach bypasses the need for extensive model retraining and retrieval model overhead, offering granular attributions and preserving the quality of generated answers. We present Verifiability-granular, an attribution dataset which has token level annotations for LLM generations in the contextual question answering setup.
  arXiv  Detail & Related papers  (2024-05-28T09:12:44Z)
• LLMs Know What They Need: Leveraging a Missing Information Guided Framework to Empower Retrieval-Augmented Generation [6.676337039829463]
  We propose a Missing Information Guided Retrieve-Extraction-Solving paradigm (MIGRES) We leverage the identification of missing information to generate a targeted query that steers the subsequent knowledge retrieval. Extensive experiments conducted on multiple public datasets reveal the superiority of the proposed MIGRES method.
  arXiv  Detail & Related papers  (2024-04-22T09:56:59Z)
• Instruction-tuned Language Models are Better Knowledge Learners [106.38526595116961]
  We propose pre-instruction-tuning (PIT), a method that instruction-tunes on questions prior to training on documents. Extensive experiments and ablation studies demonstrate that pre-instruction-tuning significantly enhances the ability of LLMs to absorb knowledge from new documents.
  arXiv  Detail & Related papers  (2024-02-20T09:20:32Z)
• Small Models, Big Insights: Leveraging Slim Proxy Models To Decide When and What to Retrieve for LLMs [60.40396361115776]
  This paper introduces a novel collaborative approach, namely SlimPLM, that detects missing knowledge in large language models (LLMs) with a slim proxy model. We employ a proxy model which has far fewer parameters, and take its answers as answers. Heuristic answers are then utilized to predict the knowledge required to answer the user question, as well as the known and unknown knowledge within the LLM.
  arXiv  Detail & Related papers  (2024-02-19T11:11:08Z)
• DIVKNOWQA: Assessing the Reasoning Ability of LLMs via Open-Domain Question Answering over Knowledge Base and Text [73.68051228972024]
  Large Language Models (LLMs) have exhibited impressive generation capabilities, but they suffer from hallucinations when relying on their internal knowledge. Retrieval-augmented LLMs have emerged as a potential solution to ground LLMs in external knowledge.
  arXiv  Detail & Related papers  (2023-10-31T04:37:57Z)
• FreshLLMs: Refreshing Large Language Models with Search Engine Augmentation [92.43001160060376]
  We study the factuality of large language models (LLMs) in the context of answering questions that test current world knowledge. We introduce FreshQA, a novel dynamic QA benchmark encompassing a diverse range of question and answer types. We benchmark a diverse array of both closed and open-source LLMs under a two-mode evaluation procedure that allows us to measure both correctness and hallucination. Motivated by these results, we present FreshPrompt, a simple few-shot prompting method that substantially boosts the performance of an LLM on FreshQA.
  arXiv  Detail & Related papers  (2023-10-05T00:04:12Z)
• Enhancing In-Context Learning with Answer Feedback for Multi-Span Question Answering [9.158919909909146]
  In this paper, we propose a novel way of employing labeled data such as it informs LLM of some undesired output. Experiments on three multi-span question answering datasets and a keyphrase extraction dataset show that our new prompting strategy consistently improves LLM's in-context learning performance.
  arXiv  Detail & Related papers  (2023-06-07T15:20:24Z)
• Can LMs Learn New Entities from Descriptions? Challenges in Propagating Injected Knowledge [72.63368052592004]
  We study LMs' abilities to make inferences based on injected facts (or propagate those facts) We find that existing methods for updating knowledge show little propagation of injected knowledge. Yet, prepending entity definitions in an LM's context improves performance across all settings.
  arXiv  Detail & Related papers  (2023-05-02T17:59:46Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.