Pith. sign in

REVIEW 5 cited by

WPO: Enhancing RLHF with Weighted Preference Optimization

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2406.11827 v2 pith:MPKPUFEX submitted 2024-06-17 cs.CL cs.AIcs.LG

classification cs.CLcs.AIcs.LG
keywords preferenceoptimizationdataoff-policymethodmodelspolicyalpaca
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Reinforcement learning from human feedback (RLHF) is a promising solution to align large language models (LLMs) more closely with human values. Off-policy preference optimization, where the preference data is obtained from other models, is widely adopted due to its cost efficiency and scalability. However, off-policy preference optimization often suffers from a distributional gap between the policy used for data collection and the target policy, leading to suboptimal optimization. In this paper, we propose a novel strategy to mitigate this problem by simulating on-policy learning with off-policy preference data. Our Weighted Preference Optimization (WPO) method adapts off-policy data to resemble on-policy data more closely by reweighting preference pairs according to their probability under the current policy. This method not only addresses the distributional gap problem but also enhances the optimization process without incurring additional costs. We validate our method on instruction following benchmarks including Alpaca Eval 2 and MT-bench. WPO not only outperforms Direct Preference Optimization (DPO) by up to 5.6% on Alpaca Eval 2 but also establishes a remarkable length-controlled winning rate against GPT-4-turbo of 76.7% based on Gemma-2-9b-it. We release the code and models at https://github.com/wzhouad/WPO.

Discussion (0). Sign in to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Multi-Turn On-Policy Distillation with Prefix Replay

    cs.LG 2026-07 conditional novelty 6.0 of 10

    ReOPD offline-distills multi-turn agentic LLMs via teacher-prefix replay plus step-decay sampling, matching online OPD accuracy at ≥4× speed with zero tool calls.

  2. Multiplayer Nash Preference Optimization

    cs.AI 2025-09 unverdicted novelty 6.0 of 10

    MNPO extends NLHF to multiplayer Nash games, inheriting equilibrium guarantees while showing empirical gains on instruction-following benchmarks under diverse preferences.

  3. Multi-Turn On-Policy Distillation with Prefix Replay

    cs.LG 2026-07 conditional novelty 5.0 of 10

    By replaying teacher prefixes with a step-decay schedule, multi-turn on-policy distillation can run without live environment interaction, matching or slightly beating online OPD accuracy.

  4. Normalized Rewards for Preference Optimization

    cs.LG 2026-06 conditional novelty 5.0 of 10

    A regularization term that conserves the combined length-normalized probability of chosen and rejected responses reduces likelihood displacement in DPO/SimPO, improves AlpacaEval and benchmark outcomes, and acts prima...

  5. Margin Adaptive DPO: Leveraging Reward Model for Granular Control in Preference Optimization

    cs.LG 2025-10 conditional novelty 5.0 of 10

    MADPO replaces DPO's fixed temperature with an instance-level weight derived from a trained reward model, amplifying low-margin preference pairs and dampening high-margin pairs.

Pith tools