Context-Parametric Inversion: Why Instruction Finetuning May Not Actually Improve Context Reliance

• URL: http://arxiv.org/abs/2410.10796v2
• Date: Tue, 22 Oct 2024 17:35:03 GMT
• Title: Context-Parametric Inversion: Why Instruction Finetuning May Not Actually Improve Context Reliance
• Authors: Sachin Goyal, Christina Baek, J. Zico Kolter, Aditi Raghunathan,
• Abstract summary: In-principle, one would expect models to adapt to the user context better after instruction finetuning. We observe a surprising failure mode: during instruction tuning, the context reliance under knowledge conflicts initially increases as expected, but then gradually decreases.
• Score: 68.56701216210617
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: A standard practice when using large language models is for users to supplement their instruction with an input context containing new information for the model to process. However, models struggle to reliably follow the input context, especially when it conflicts with their parametric knowledge from pretraining. In-principle, one would expect models to adapt to the user context better after instruction finetuning, particularly when handling knowledge conflicts. However, we observe a surprising failure mode: during instruction tuning, the context reliance under knowledge conflicts initially increases as expected, but then gradually decreases as instruction finetuning progresses. This happens while the performance on standard benchmarks keeps on increasing far after this drop. We call this phenomenon context-parametric inversion and observe it across multiple general purpose instruction tuning datasets such as TULU, Alpaca and Ultrachat, across different model families like Llama, Mistral, and Pythia. We perform various controlled studies and theoretical analysis to show that context-parametric inversion occurs due to examples in the instruction finetuning data where the input context provides information that aligns with model's parametric knowledge. Our analysis suggests some natural mitigation strategies with limited but insightful gains, and serves as a useful starting point in addressing this deficiency in instruction finetuning.

Related papers

• Context-aware Prompt Tuning: Advancing In-Context Learning with Adversarial Methods [69.36397993451742]
  This work introduces Context-aware Prompt Tuning (CPT), a method inspired by ICL, PT, and adversarial attacks. We modify specific context tokens, considering the unique structure of input and output formats. Inspired by adversarial attacks, we adjust the input based on the labels present in the context, focusing on minimizing, rather than maximizing, the loss.
  arXiv  Detail & Related papers  (2024-10-22T17:45:47Z)
• Information Guided Regularization for Fine-tuning Language Models [11.831883526217942]
  We argue that a more surgical approach to regularization needs to exist for smoother transfer learning. We devise a novel approach to dropout for improved model regularization and better downstream generalization.
  arXiv  Detail & Related papers  (2024-06-20T05:18:37Z)
• Disperse-Then-Merge: Pushing the Limits of Instruction Tuning via Alignment Tax Reduction [75.25114727856861]
  Large language models (LLMs) tend to suffer from deterioration at the latter stage ofSupervised fine-tuning process. We introduce a simple disperse-then-merge framework to address the issue. Our framework outperforms various sophisticated methods such as data curation and training regularization on a series of standard knowledge and reasoning benchmarks.
  arXiv  Detail & Related papers  (2024-05-22T08:18:19Z)
• Studying Large Language Model Behaviors Under Context-Memory Conflicts With Real Documents [54.953320616069654]
  Retrieval-augmented generation mitigates many problems of fully parametric language models. In RAG, the model's knowledge can be updated from documents provided in context. We present a framework for studying such knowledge conflicts in a realistic setup.
  arXiv  Detail & Related papers  (2024-04-24T17:59:36Z)
• Robust and Scalable Model Editing for Large Language Models [75.95623066605259]
  We propose EREN (Edit models by REading Notes) to improve the scalability and robustness of LLM editing. Unlike existing techniques, it can integrate knowledge from multiple edits, and correctly respond to syntactically similar but semantically unrelated inputs.
  arXiv  Detail & Related papers  (2024-03-26T06:57:23Z)
• R-Tuning: Instructing Large Language Models to Say `I Don't Know' [66.11375475253007]
  Large language models (LLMs) have revolutionized numerous domains with their impressive performance but still face their challenges. Previous instruction tuning methods force the model to complete a sentence no matter whether the model knows the knowledge or not. We present a new approach called Refusal-Aware Instruction Tuning (R-Tuning) Experimental results demonstrate R-Tuning effectively improves a model's ability to answer known questions and refrain from answering unknown questions.
  arXiv  Detail & Related papers  (2023-11-16T08:45:44Z)
• Influence Tuning: Demoting Spurious Correlations via Instance Attribution and Instance-Driven Updates [26.527311287924995]
  influence tuning can help deconfounding the model from spurious patterns in data. We show that in a controlled setup, influence tuning can help deconfounding the model from spurious patterns in data.
  arXiv  Detail & Related papers  (2021-10-07T06:59:46Z)

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.