Papers › Extensive Self-Contrast Enables Feedback-Free Language Model Alignment

Extensive Self-Contrast Enables Feedback-Free Language Model Alignment

31 Mar 2024arXiv:2404.00604archive 2025-07-28

Xiao Liu, Xixuan Song, Yuxiao Dong, Jie Tang

Reinforcement learning from human feedback (RLHF) has been a central technique for recent large language model (LLM) alignment. However, its heavy dependence on costly human or LLM-as-Judge preference feedback could stymie its wider applications. In this work, we introduce Self-Contrast, a feedback-free large language model alignment method via exploiting extensive self-generated negatives. With only supervised fine-tuning (SFT) targets, Self-Contrast leverages the LLM itself to generate massive diverse candidates, and harnesses a pre-trained embedding model to filter multiple negatives according to text similarity. Theoretically, we illustrate that in this setting, merely scaling negative responses can still effectively approximate situations with more balanced positive and negative preference annotations. Our experiments with direct preference optimization (DPO) on three datasets show that, Self-Contrast could consistently outperform SFT and standard DPO training by large margins. And as the number of self-generated negatives increases, the performance of Self-Contrast continues to grow. Code and data are available at https://github.com/THUDM/Self-Contrast.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

hiyouga/llama-factory officialmentioned in papermentioned on GitHubpytorchApache-2.0 report
thudm/self-contrast officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Language ModelingLanguage ModellingLarge Language Modeltext similarity

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

DPOSFT

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections