Conformal Language Modeling

• URL: http://arxiv.org/abs/2306.10193v2
• Date: Sat, 1 Jun 2024 21:40:33 GMT
• Title: Conformal Language Modeling
• Authors: Victor Quach, Adam Fisch, Tal Schuster, Adam Yala, Jae Ho Sohn, Tommi S. Jaakkola, Regina Barzilay,
• Abstract summary: We propose a novel approach to conformal prediction for generative language models (LMs) Standard conformal prediction produces prediction sets with rigorous, statistical guarantees. We demonstrate the promise of our approach on multiple tasks in open-domain question answering, text summarization, and radiology report generation.
• Score: 61.94417935386489
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We propose a novel approach to conformal prediction for generative language models (LMs). Standard conformal prediction produces prediction sets -- in place of single predictions -- that have rigorous, statistical performance guarantees. LM responses are typically sampled from the model's predicted distribution over the large, combinatorial output space of natural language. Translating this process to conformal prediction, we calibrate a stopping rule for sampling different outputs from the LM that get added to a growing set of candidates until we are confident that the output set is sufficient. Since some samples may be low-quality, we also simultaneously calibrate and apply a rejection rule for removing candidates from the output set to reduce noise. Similar to conformal prediction, we prove that the sampled set returned by our procedure contains at least one acceptable answer with high probability, while still being empirically precise (i.e., small) on average. Furthermore, within this set of candidate responses, we show that we can also accurately identify subsets of individual components -- such as phrases or sentences -- that are each independently correct (e.g., that are not "hallucinations"), again with statistical guarantees. We demonstrate the promise of our approach on multiple tasks in open-domain question answering, text summarization, and radiology report generation using different LM variants.

Related papers

• Conformal Generative Modeling with Improved Sample Efficiency through Sequential Greedy Filtering [55.15192437680943]
  Generative models lack rigorous statistical guarantees for their outputs. We propose a sequential conformal prediction method producing prediction sets that satisfy a rigorous statistical guarantee. This guarantee states that with high probability, the prediction sets contain at least one admissible (or valid) example.
  arXiv  Detail & Related papers  (2024-10-02T15:26:52Z)
• Balancing Diversity and Risk in LLM Sampling: How to Select Your Method and Parameter for Open-Ended Text Generation [60.493180081319785]
  We propose a systematic way to estimate the intrinsic capacity of a truncation sampling method by considering the trade-off between diversity and risk at each decoding step. Our work provides a comprehensive comparison between existing truncation sampling methods, as well as their recommended parameters as a guideline for users.
  arXiv  Detail & Related papers  (2024-08-24T14:14:32Z)
• Language Models with Conformal Factuality Guarantees [44.767328168194815]
  Conformal factuality is a framework that can ensure high probability correctness guarantees for language model (LM) outputs. We show that conformal prediction in language models corresponds to a back-off algorithm that provides high probability correctness guarantees.
  arXiv  Detail & Related papers  (2024-02-15T18:31:53Z)
• Non-Exchangeable Conformal Language Generation with Nearest Neighbors [12.790082627386482]
  Non-exchangeable conformal nucleus sampling is a novel extension of the conformal prediction framework to generation based on nearest neighbors. Our method can be used post-hoc for an arbitrary model without extra training and supplies token-level, calibrated prediction sets equipped with statistical guarantees.
  arXiv  Detail & Related papers  (2024-02-01T16:04:04Z)
• Conformal Nucleus Sampling [67.5232384936661]
  We assess whether a top-$p$ set is indeed aligned with its probabilistic meaning in various linguistic contexts. We find that OPT models are overconfident, and that calibration shows a moderate inverse scaling with model size.
  arXiv  Detail & Related papers  (2023-05-04T08:11:57Z)
• Arithmetic Sampling: Parallel Diverse Decoding for Large Language Models [65.52639709094963]
  Methods such as beam search and Gumbel top-k sampling can guarantee a different output for each element of the beam, but are not easy to parallelize. We present a framework for sampling according to an arithmetic code book implicitly defined by a large language model.
  arXiv  Detail & Related papers  (2022-10-18T22:19:41Z)
• Selection by Prediction with Conformal p-values [7.917044695538599]
  We study screening procedures that aim to select candidates whose unobserved outcomes exceed user-specified values. We develop a method that wraps around any prediction model to produce a subset of candidates while controlling the proportion of falsely selected units.
  arXiv  Detail & Related papers  (2022-10-04T06:34:49Z)
• Conformal Prediction Sets with Limited False Positives [43.596058175459746]
  We develop a new approach to multi-label conformal prediction in which we aim to output a precise set of promising prediction candidates with a bounded number of incorrect answers. We demonstrate the effectiveness of this approach across a number of classification tasks in natural language processing, computer vision, and computational chemistry.
  arXiv  Detail & Related papers  (2022-02-15T18:52:33Z)
• Individual Calibration with Randomized Forecasting [116.2086707626651]
  We show that calibration for individual samples is possible in the regression setup if the predictions are randomized. We design a training objective to enforce individual calibration and use it to train randomized regression functions.
  arXiv  Detail & Related papers  (2020-06-18T05:53:10Z)

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.