Pith. sign in

REVIEW 6 cited by

Direct Preference Optimization with an Offset

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 2402.10571 v2 pith:DJDPY7ME submitted 2024-02-16 cs.CL cs.AIcs.LG

Direct Preference Optimization with an Offset

classification cs.CL cs.AIcs.LG
keywords preferenceresponseoffsetpreferreddispreferredlanguageodpoaligning
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Direct preference optimization (DPO) is a successful fine-tuning strategy for aligning large language models with human preferences without the need to train a reward model or employ reinforcement learning. DPO, as originally formulated, relies on binary preference data and fine-tunes a language model to increase the likelihood of a preferred response over a dispreferred response. However, not all preference pairs are equal. Sometimes, the preferred response is only slightly better than the dispreferred one. In other cases, the preference is much stronger. For instance, if a response contains harmful or toxic content, the annotator will have a strong preference for that response. In this paper, we propose a generalization of DPO, termed DPO with an offset (ODPO), that does not treat every preference pair equally during fine-tuning. Intuitively, ODPO requires the difference between the likelihood of the preferred and dispreferred response to be greater than an offset value. The offset is determined based on the extent to which one response is preferred over another. Our experiments on various tasks suggest that ODPO significantly outperforms DPO in aligning language models, especially when the number of preference pairs is limited.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 6 Pith papers

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

  1. Towards Order Fairness: Mitigating LLMs Order Sensitivity through Dual Group Advantage Optimization

    cs.LG 2026-05 unverdicted novelty 7.0

    DGAO uses reinforcement learning to optimize LLMs for both accuracy and order stability by balancing intra-group accuracy advantages and inter-group stability advantages.

  2. MASS-DPO: Multi-negative Active Sample Selection for Direct Policy Optimization

    cs.LG 2026-05 unverdicted novelty 7.0

    MASS-DPO derives a Plackett-Luce-specific log-determinant Fisher information objective to select non-redundant negative samples, matching or exceeding multi-negative DPO performance with substantially fewer negatives ...

  3. Mid-Training with Self-Generated Data Improves Reinforcement Learning in Language Models

    cs.AI 2026-05 unverdicted novelty 5.0

    Mid-training LLMs on self-generated diverse reasoning paths improves subsequent RL performance on mathematical benchmarks and OOD tasks.

  4. ARGUS: Policy-Adaptive Ad Governance via Evolving Reinforcement with Adversarial Umpiring

    cs.CL 2026-05 unverdicted novelty 5.0

    ARGUS uses a Prosecutor-Defender-Umpire multi-agent setup plus RAG and chain-of-thought rewards to adapt ad policy enforcement to new regulations using minimal fresh labels.

  5. Enhancing Speech Large Language Models through Reinforced Behavior Alignment

    cs.CL 2025-08 unverdicted novelty 5.0

    Reinforced Behavior Alignment (RBA) uses self-synthesized data from a teacher LLM and reinforcement learning to close the instruction-following gap in SpeechLMs, outperforming distillation and reaching SOTA on spoken ...

  6. Towards Large Reasoning Models: A Survey of Reinforced Reasoning with Large Language Models

    cs.AI 2025-01 unverdicted novelty 3.0

    The paper surveys reinforced reasoning techniques for LLMs, covering automated data construction, learning-to-reason methods, and test-time scaling as steps toward Large Reasoning Models.