No article text was recorded for this version.
scalar reward reductionism
Version 1 of 139 papers40 conceptsfrozen 2026-10-04
Papers in this version
- Blog: Quality Is a Decision, Not a Measurement: A Safety Rubric Reorders UltraFeedback's Binarized Labels
- Blog: Self-Play Preference Optimization for Language Model Alignment
- Blog: Multi-Objective SPIBB: Seldonian Offline Policy Improvement with Safety Constraints in Finite MDPs
- Blog: Imitation Learning via Focused Satisficing
- Blog: Beyond Bradley-Terry Models: A General Preference Model for Language Model Alignment
- Blog: Simultaneous Multi-objective Alignment Across Verifiable and Non-verifiable Rewards
- Blog: Rethinking Diverse Human Preference Learning through Principal Component Analysis
- Blog: Scalar reward is not enough: a response to Silver, Singh, Precup and Sutton (2021)
- Blog: Energy-Based Reward Models for Robust Language Model Alignment
- Blog: Position: Alignment Needs Rule-Class Routing Before Preference Learning
- Blog: Reinforcement Recommendation with User Multi-aspect Preference
- Blog: Controllable Preference Optimization: Toward Controllable Multi-Objective Alignment
- Blog: Extracting Reward Functions from Diffusion Models
- Blog: Beyond Reward: Offline Preference-guided Policy Optimization
- Blog: How Aligned are Different Alignment Metrics?
- Blog: Panacea: Pareto Alignment via Preference Adaptation for LLMs
- Blog: 2D-DPO: Scaling Direct Preference Optimization with 2-Dimensional Supervision
- Blog: Alternating Reinforcement Learning for Rubric-Based Reward Modeling in Non-Verifiable LLM Post-Training
- Blog: Sequence to Sequence Reward Modeling: Improving RLHF by Language Feedback
- Blog: Iterative Nash Policy Optimization: Aligning LLMs with General Preferences via No-Regret Learning
- Blog: Arithmetic Control of LLMs for Diverse User Preferences: Directional Preference Alignment with Multi-Objective Rewards
- Blog: Rethinking AI Alignment: From Static Rewards to Social Reinforcement Learning
- Blog: SteerLM: Attribute Conditioned SFT as an (User-Steerable) Alternative to RLHF
- Blog: General Preference Modeling with Preference Representations for Aligning Language Models
- Blog: Learning Multi-Dimensional Human Preference for Text-to-Image Generation
- Blog: Pluralistic AI Alignment Requires Inference-Time Multi-Objective Control
- Blog: Safe RLHF: Safe Reinforcement Learning from Human Feedback
- Blog: Aligning Generative Language Models with Human Values
- Blog: Pareto Multi-Objective Alignment for Language Models
- Blog: Three Dogmas of Reinforcement Learning
- Blog: Improving Instruction Following in Language Models through Proxy-Based Uncertainty Estimation
- Blog: DMoERM: Recipes of Mixture-of-Experts for Effective Reward Modeling
- Blog: Checklists Are Better Than Reward Models For Aligning Language Models
- Blog: DRESS: Instructing Large Vision-Language Models to Align and Interact with Humans via Natural Language Feedback
- Blog: Latent Preference Coding: Aligning Large Language Models via Discrete Latent Codes
- Blog: From Lists to Emojis: How Format Bias Affects Model Alignment
- Blog: Preference-Guided Reflective Sampling for Aligning Language Models
- Blog: Sequential Preference Optimization: Multi-Dimensional Preference Alignment with Implicit Reward Modeling
- Blog: Text2Grad: Reinforcement Learning from Natural Language Feedback
History & citation
Current · Permanent link
- This direction split from Multi-Dimensional Preference Alignment for Language Models (79% overlap)
Cite this direction
Latest version (a living page):
@misc{lacuna_scalar_reward_reductionism_0a83da28,
title = {scalar reward reductionism},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/scalar-reward-reductionism/rdr_0a83da28e19a423c97ec74b879747a4f},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-20},
}This exact version (version 1, frozen text):
@misc{lacuna_scalar_reward_reductionism_0a83da28_v1,
title = {scalar reward reductionism},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/scalar-reward-reductionism/rdr_0a83da28e19a423c97ec74b879747a4f/v/1},
note = {Version 1, frozen 2026-10-04. Accessed 2026-11-20},
}