paper-with-me

홈 › Papers

Humanline: Online Alignment as Perceptual Loss

2025-09-29 · Sijia Liu, Niklas Muennighoff, Kawin Ethayarajh arxiv

Online alignment (e.g., GRPO) is generally more performant than offline alignment (e.g., DPO) -- but why? Drawing on prospect theory from behavioral economics, we propose a human-centric explanation. We prove that online on-policy sampling better approximates the human-perceived distribution of what the model can produce, and PPO/GRPO-style clipping -- originally introduced to just stabilize training -- recovers a perceptual bias in how humans perceive probability. In this sense, PPO/GRPO act as perceptual losses already. Our theory further suggests that the online/offline dichotomy is itself incidental to maximizing human utility, since we can achieve the same effect by selectively training on any data in a manner that mimics human perception, rather than restricting ourselves to online on-policy data. Doing so would allow us to post-train more quickly, cheaply, and flexibly without sacrificing performance. To this end, we propose a design pattern that explicitly incorporates perceptual distortions of probability into objectives like DPO/KTO/GRPO, creating humanline variants of them. Surprisingly, we find that these humanline variants, even when trained with offline off-policy data, can match the performance of their online counterparts (on both verifiable and unverifiable tasks) while running up to 6x faster.

📄 PDF Abstract BibTeX arXiv:2509.24207

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

OARS: Process-Aware Online Alignment for Generative Real-World Image Super-Resolution

2026-03-13 · Shijie Zhao, Xuanyu Zhang, Bin Chen, Weiqi Li 외 arxiv

Aligning generative real-world image super-resolution models with human visual preference is challenging due to the perception--fidelity trade-off and diverse, unknown degradations. Prior approaches rely on offline prefe…

Image Super-Resolution

Improving Image Autoencoder Embeddings with Perceptual Loss

2020-01-10 · Gustav Grund Pihlgren, Fredrik Sandin, Marcus Liwicki

Autoencoders are commonly trained using element-wise loss. However, element-wise loss disregards high-level structures in the image which can lead to embeddings that disregard them as well. A recent improvement to autoen…

Bridging the Perceptual Gap: Residual-Enhanced Downscaling and Manifold-Aware Perception Alignment Adaptation for NR-IQA

2026-09-15 · Yu Li, Zhengran Shen, Yachun Mi, Puchao Zhou 외 arxiv

Leveraging Large Vision-Language Models like CLIP has recently set new benchmarks for No-Reference Image Quality Assessment (NR-IQA). However, the contrastive pretraining of CLIP inherently prioritizes semantic invarianc…

No-Reference Image Quality Assessment

Perceptual Quality Optimization of Image Super-Resolution

2026-02-25 · Wei Zhou, Yixiao Li, Hadi Amirpour, Xiaoshuai Hao 외 arxiv

Single-image super-resolution (SR) has achieved remarkable progress with deep learning, yet most approaches rely on distortion-oriented losses or heuristic perceptual priors, which often lead to a trade-off between fidel…

Image Super-Resolution

Structured Uncertainty Similarity Score (SUSS): Learning a Probabilistic, Interpretable, Perceptual Metric Between Images

2025-12-03 · Paula Seidler, Neill D. F. Campbell, Ivor J A Simpson arxiv

Perceptual similarity scores that align with human vision are critical for both training and evaluating computer vision models. Deep perceptual losses, such as LPIPS, achieve good alignment but rely on complex, highly no…