AI Paper Atlas

Research map / Reinforcement learning / RL from feedback

Why DPO is a Misspecified Estimator and How to Fix It

Aditya Gopalan, Sayak Ray Chowdhury, Debangshu Banerjee

ICLR 2026 · Reinforcement learning › RL from feedback · approach: Reward modeling versus direct optimization · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add auxiliary nullspace variables to the DPO loss (AuxDPO)DPO, Llama3.1-8B · MMLU-Pro (in-domain)% accuracy change vs base: 57.14 -> 63.26Improved
add auxiliary nullspace variables to the DPO loss (AuxDPO)DPO, Llama3.1-8B · RewardBench v2 (OOD)% accuracy change vs base: 14.31 -> 32.44Improved
add auxiliary nullspace variables to the DPO loss (AuxDPO)DPO, Qwen3-0.6B · RewardBench v2 (OOD)% accuracy change vs base: -8.16 -> 18.36Improved
add auxiliary nullspace variables to the DPO loss (AuxDPO)DPO, Llama3.2-1B · RewardBench v2 (OOD)% accuracy change vs base: 14.11 -> 43.27Improved
use AuxDPO instead of IPOLlama3.2-1B · MMLU-Pro (OOD)% accuracy change vs base: 14.58 (IPO) vs 12.52 (AuxDPO)Hurt
add auxiliary nullspace variables to the DPO loss (AuxDPO)DPO, Llama3.1-8B · MMLU-Pro overall accuracy (OOD)accuracy (%): 27.06 -> 39.26Improved
add auxiliary nullspace variables to the DPO loss (AuxDPO)DPO, Llama3.1-8B · MMLU-Pro overall accuracy (in-domain)accuracy (%): 46.60 -> 51.95Improved
reduce trainable parameters (LoRA r=4 or last layer only) (ablation)AuxDPO vs DPO · MMLU-Pro and RewardBench v2 (ID)accuracyImproved
increase trainable fraction above 75% (ablation)DPO and AuxDPO · MMLU-Pro and RewardBench v2accuracyNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy