arXiv AI
Jul 21
Normalized Rewards for Preference Optimization
arXiv:2607. 16240v1 Announce Type: cross Abstract: Direct Alignment Algorithms (DAAs) such as DPO have become a common way to post-train and align LLMs with human preferences.
Patterning in Practice: Debiasing Reward Models with Susceptibilities
The paper introduces patterning, a reweighting technique that adjusts preference pairs based on their susceptibility to bias, to debias a Gemma 2 9B Instruct reward model trained on Skywork-Reward-Preference v0.2. Using this method, the authors achieve a +14.2 ± 1.2 percentage point improvement on the RM‑Bench Hard split while maintaining overall accuracy, matching the best reported Hard‑split gain from a comparable model. The study also demonstrates that the learned weights are interpretable, transferable across Gemma variants, and partially effective on Llama 3.1 8B.
Related stories
arXiv Machine Learning
Jun 18
The Reward Was in Your Data All Along: Correcting Flow Matching with Discriminator-Guided RL
arXiv:2606. 19162v1 Announce Type: new Abstract: Score- and flow-matching models often rely on preference-based reinforcement learning for two purposes: aligning with subjective preferences and, surprisingly, recovering properties such as visual realism and coherent object structure that matching-based training is intended to learn from the data itself.
arXiv AI
Jun 12
Boosting Direct Preference Optimization with Penalization
arXiv:2606. 12505v1 Announce Type: cross Abstract: Offline preference optimization has become a practical substitute for reinforcement learning from human feedback, but pairwise objectives such as Direct Preference Optimization (DPO) and its variants use only the chosen and rejected responses stored in a static dataset.
arXiv Machine Learning
2d ago
Uncertainty-Normalized Margins for Direct Preference Optimization
arXiv:2609.38647v1 Announce Type: new Abstract: Direct preference optimization (DPO) models binary preferences through a Bradley-Terry model with a common noise scale, without explicitly accounting f...
arXiv AI
Jun 2
Margin Adaptive DPO: Leveraging Reward Model for Granular Control in Preference Optimization
arXiv:2510. 05342v2 Announce Type: replace-cross Abstract: Direct Preference Optimization (DPO) has emerged as a simple and effective method for aligning large language models.
arXiv AI
Jul 24
Towards Disentangled Preference Optimization Dynamics: Suppress the Loser, Preserve the Winner
arXiv:2604. 18239v4 Announce Type: replace-cross Abstract: Preference optimization is widely used to align large language models (LLMs) with human preferences.
arXiv AI
Aug 24
Share the Judge, Learn the Deferral: Where Specialization Helps LLM Evaluation
The paper investigates two strategies for improving large language model (LLM) evaluation: specialized judge weights and rule‑based deferral policies. Experiments on nearly 100,000 rubric‑conditioned samples show that correct rubrics boost accuracy, while incorrect ones hurt it, and that splitting training data into criterion‑specific experts can severely degrade performance unless the experts are warm‑started from a unified model. The authors demonstrate that lightweight deferral cascades can match or exceed the accuracy of larger standalone judges at a fraction of the compute cost, and they provide practical design rules for building efficient, reliable LLM evaluators.
arXiv AI
Aug 21
Truncate Bad, Upweight Good: BoN-Style Distillation via Rank-Based Classification
arXiv:2608. 19748v1 Announce Type: cross Abstract: Inference-time selection methods, such as Best-of-N, improve generation by sampling a pool of candidates and selecting the top-ranked completion according to a reward model.
arXiv AI
Jun 30
Pessimism's Paradox: Conservative Offline Training Amplifies Reward Hacking During Online Adaptation in Reasoning Models
arXiv:2606. 30627v1 Announce Type: cross Abstract: Conservative offline training is widely advocated as a safe foundation for subsequent online adaptation: if a policy stays close to well-supported behaviour, the argument goes, it is less likely to exploit imperfections in a learned reward model.
arXiv AI
Aug 25
XTC: Head-Aware Sampling by Excluding Top Choices
XTC (Exclude Top Choices) is a lightweight, head‑aware decoding operator that improves diversity in autoregressive language models by removing overly probable tokens that dominate the next‑token distribution. It works by identifying tokens above a plausibility threshold, probabilistically excluding the dominant choices, and renormalizing the remaining distribution. Across 60 experiments on models such as Gemma 3 and DeepSeek R1, XTC boosts Distinct‑2 scores by 11–15 % and cuts repeat trigrams by 27–47 %, while a Mechanical Turk study shows a 62.3 % preference for XTC‑generated text without loss of fluency.
arXiv AI
Jul 14
Length Penalties Make Chain-of-Thought Less Monitorable
arXiv:2607. 09786v1 Announce Type: new Abstract: Length-penalized reinforcement learning can shorten chain-of-thought reasoning while hiding an influence that drives the model's answer.
Hugging Face Trending Papers
Aug 24
XTC: Head-Aware Sampling by Excluding Top Choices
Standard decoding rules for autoregressive language models promote diversity by rescaling the full next-token distribution or truncating its low-probability tail. These strategies overlook a common re...