Inferring from Logits: Exploring Best Practices for Decoding-Free Generative Candidate Selection

• URL: http://arxiv.org/abs/2501.17338v1
• Date: Tue, 28 Jan 2025 23:21:28 GMT
• Title: Inferring from Logits: Exploring Best Practices for Decoding-Free Generative Candidate Selection
• Authors: Mingyu Derek Ma, Yanna Ding, Zijie Huang, Jianxi Gao, Yizhou Sun, Wei Wang,
• Abstract summary: Generative Language Models rely on autoregressive decoding to produce the output sequence token by token. We introduce an evaluation of a comprehensive collection of decoding-free candidate selection approaches on a comprehensive set of tasks.
• Score: 37.54564513506548
• License:
• Abstract: Generative Language Models rely on autoregressive decoding to produce the output sequence token by token. Many tasks such as preference optimization, require the model to produce task-level output consisting of multiple tokens directly by selecting candidates from a pool as predictions. Determining a task-level prediction from candidates using the ordinary token-level decoding mechanism is constrained by time-consuming decoding and interrupted gradients by discrete token selection. Existing works have been using decoding-free candidate selection methods to obtain candidate probability from initial output logits over vocabulary. Though these estimation methods are widely used, they are not systematically evaluated, especially on end tasks. We introduce an evaluation of a comprehensive collection of decoding-free candidate selection approaches on a comprehensive set of tasks, including five multiple-choice QA tasks with a small candidate pool and four clinical decision tasks with a massive amount of candidates, some with 10k+ options. We evaluate the estimation methods paired with a wide spectrum of foundation LMs covering different architectures, sizes and training paradigms. The results and insights from our analysis inform the future model design.

Related papers

• Multi-Draft Speculative Sampling: Canonical Architectures and Theoretical Limits [26.220189807865548]
  We consider multi-draft speculative sampling, where the proposal sequences are sampled independently from different draft models. We show that the optimal scheme can be decomposed into a two-step solution.
  arXiv  Detail & Related papers  (2024-10-23T19:28:34Z)
• On Speeding Up Language Model Evaluation [48.51924035873411]
  Development of prompt-based methods with Large Language Models (LLMs) requires making numerous decisions. We propose a novel method to address this challenge. We show that it can identify the top-performing method using only 5-15% of the typically needed resources.
  arXiv  Detail & Related papers  (2024-07-08T17:48:42Z)
• Multi-Candidate Speculative Decoding [82.05519287513444]
  Large language models have shown impressive capabilities across a variety of NLP tasks, yet their generating text autoregressively is time-consuming. One way to speed them up is speculative decoding, which generates candidate segments from a fast draft model that is then verified in parallel by the target model. This paper proposes sampling multiple candidates from a draft model and then organising them in batches for verification. We design algorithms for efficient multi-candidate verification while maintaining the distribution of the target model.
  arXiv  Detail & Related papers  (2024-01-12T17:15:23Z)
• Large Language Models Are Not Robust Multiple Choice Selectors [117.72712117510953]
  Multiple choice questions (MCQs) serve as a common yet important task format in the evaluation of large language models (LLMs) This work shows that modern LLMs are vulnerable to option position changes due to their inherent "selection bias" We propose a label-free, inference-time debiasing method, called PriDe, which separates the model's prior bias for option IDs from the overall prediction distribution.
  arXiv  Detail & Related papers  (2023-09-07T17:44:56Z)
• Probabilistic Planning with Prioritized Preferences over Temporal Logic Objectives [26.180359884973566]
  We study temporal planning in probabilistic environments, modeled as labeled Markov decision processes (MDPs) This paper introduces a new specification language, termed prioritized qualitative choice linear temporal logic on finite traces. We formulate and solve a problem of computing an optimal policy that minimizes the expected score of dissatisfaction given user preferences.
  arXiv  Detail & Related papers  (2023-04-23T13:03:27Z)
• 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)
• Biological Sequence Design with GFlowNets [75.1642973538266]
  Design of de novo biological sequences with desired properties often involves an active loop with several rounds of molecule ideation and expensive wet-lab evaluations. This makes the diversity of proposed candidates a key consideration in the ideation phase. We propose an active learning algorithm leveraging uncertainty estimation and the recently proposed GFlowNets as a generator of diverse candidate solutions.
  arXiv  Detail & Related papers  (2022-03-02T15:53:38Z)
• Online Active Model Selection for Pre-trained Classifiers [72.84853880948894]
  We design an online selective sampling approach that actively selects informative examples to label and outputs the best model with high probability at any round. Our algorithm can be used for online prediction tasks for both adversarial and streams.
  arXiv  Detail & Related papers  (2020-10-19T19:53:15Z)
• Optimal Multiple Stopping Rule for Warm-Starting Sequential Selection [9.625436987364909]
  We present the Warm-starting Dynamic Thresholding algorithm, developed using dynamic programming, for a variant of the standard online selection problem. The problem allows job positions to be either free or already occupied at the beginning of the process. Throughout the selection process, the decision maker interviews one after the other the new candidates and reveals a quality score for each of them.
  arXiv  Detail & Related papers  (2020-02-12T14:04:43Z)

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.