paper-with-me

홈 › Papers

SoFA: Shielded On-the-fly Alignment via Priority Rule Following

2024-02-27 · Xinyu Lu, Bowen Yu, Yaojie Lu, Hongyu Lin, Haiyang Yu, Le Sun, Xianpei Han, Yongbin Li

The alignment problem in Large Language Models (LLMs) involves adapting them to the broad spectrum of human values. This requirement challenges existing alignment methods due to diversity of preferences and regulatory standards. This paper introduces a novel alignment paradigm, priority rule following, which defines rules as the primary control mechanism in each dialog, prioritizing them over user instructions. Our preliminary analysis reveals that even the advanced LLMs, such as GPT-4, exhibit shortcomings in understanding and prioritizing the rules. Therefore, we present PriorityDistill, a semi-automated approach for distilling priority following signals from LLM simulations to ensure robust rule integration and adherence. Our experiments show that this method not only effectively minimizes misalignments utilizing only one general rule but also adapts smoothly to various unseen rules, ensuring they are shielded from hijacking and that the model responds appropriately.

📄 PDF Abstract BibTeX arXiv:2402.17358

Code (1)

luxinyu1/sofa 공식 구현

Tasks

Diversity

Methods 이 논문이 사용한 방법론

Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Multi-Head Attention 설명 없음
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Adam 설명 없음

Similar Papers 제목 키워드 기반

CARVE-Q: Quantum-Proposed, Classically Certified Interactive Driving Repair

2026-06-03 · Yifan Wang arxiv

The critical question after a correct driving veto is not only whether a maneuver is unsafe, but whether the blocked interaction admits a lawful, auditable, and responsibility-bounded repair. Prediction and game-theoreti…

Fair Allocation of Vaccines, Ventilators and Antiviral Treatments: Leaving No Ethical Value Behind in Health Care Rationing

2020-08-02 · Parag A. Pathak, Tayfun Sönmez, M. Utku Ünver, M. Bumin Yenmez

A priority system has traditionally been the protocol of choice for the allocation of scarce life-saving resources during public health emergencies. Covid-19 revealed the limitations of this allocation rule. Many argue t…

Cross-center Early Sepsis Recognition by Medical Knowledge Guided Collaborative Learning for Data-scarce Hospitals

2023-02-11 · Ruiqing Ding, Fangjie Rong, Xiao Han, Leye Wang

There are significant regional inequities in health resources around the world. It has become one of the most focused topics to improve health services for data-scarce hospitals and promote health equity through knowledg…

Transfer Learning

Modeling and Control of a Pneumatic Morphing Soft Quadrotor based on the SOFA Framework for Dynamic Soft Robotic Simulation

2026-05-20 · F. Labra Caso, V. Sumathy, P. Ferrentino, B. Vanderborght 외 arxiv

This article presents a novel SOFA based finite element method for the soft body modeling and the corresponding dynamic simulation and control of a pneumatic morphing soft quadrotor. The proposed modeling preserves the p…

VISA: Value Injection via Shielded Adaptation for Personalized LLM Alignment

2026-03-05 · Jiawei Chen, Tianzhuo Yang, Guoxi Zhang, Jiaming Ji 외 arxiv

Aligning Large Language Models (LLMs) with nuanced human values remains a critical challenge, as existing methods like Reinforcement Learning from Human Feedback (RLHF) often handle only coarse-grained attributes. In pra…

Reinforcement Learning