Work

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

Year: 2025

Type: preprint

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.

Source: arXiv (Cornell University)

DOI: 10.48550/arxiv.2501.17338

Authors: Ma, Mingyu Derek, Ding, Yanna, Huang, Zijie, Gao, Jianxi, Sun, Yizhou, Wang, Wei

Institutions: University of California, Los Angeles, Rensselaer Polytechnic Institute

Language: English

Topic: Topic Modeling

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funders: National Science Foundation, Rensselaer Polytechnic Institute, National Institutes of Health, National Human Genome Research Institute

Awards: CAREER: Network Resilience: Theories, Algorithms, and Applications, III: Medium: Collaborative Research: Collaborative Machine-Learning-Centric Data Analytics at Scale, PIPP Phase I: An End-to-End Pandemic Early Warning System by Harnessing Open-source Intelligence, Collaborative Research: III: Medium: VirtualLab: Integrating Deep Graph Learning and Causal Inference for Multi-Agent Dynamical Systems, Building BRIDGES: Coordinating the Dissemination and Training for Biomedical Artificial Intelligence

OpenAlex ID: W4406975362

Cites:

Cited by:

Related to: 10