Request for adding paper: Eliminating Inductive Bias in Reward Models with Information-Theoretic Guidance#82
Open
BIRlz wants to merge 2 commits into
Open
Request for adding paper: Eliminating Inductive Bias in Reward Models with Information-Theoretic Guidance#82BIRlz wants to merge 2 commits into
BIRlz wants to merge 2 commits into
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Dear Authors,
We sincerely thank you for maintaining awesome-RLHF code repo. We would like to recommend our work, DIR (Debiasing via Information optimization), for inclusion: "Eliminating Inductive Bias in Reward Models with Information-Theoretic Guidance" https://arxiv.org/abs/2512.23461.
In this paper, we propose a novel method called DIR (Debiasing via Information optimization). Inspired by the Information Bottleneck principle, DIR decouples preferences from biased attributes by maximizing mutual information (MI) with human feedback while minimizing MI with spurious features. Beyond the theoretical framework, DIR has been rigorously validated in large-scale industrial scenarios and successfully integrated into our production development.
We believe this information-theoretic perspective offers a robust solution to RM bias, and we would be honored if you could include it in your repo. Thank you for considering this request.