English
Related papers

Related papers: Super(ficial)-alignment: Strong Models May Deceive…

200 papers

The paradigm of Weak-to-Strong Generalization (W2SG) suggests that a pre-trained strong model can surpass its weak supervisor, yet the decisive role of pre-training remains theoretically and empirically under-explored. In this work, we…

Machine Learning · Computer Science 2026-05-08 Wei Yao , Wang Zhaoyang , Gengze Xu , Chen Qian , Dongrui Liu , Ziqiao Wang , Yong Liu , Yunbei Xu

Large language models (LLMs) are increasingly deployed with hierarchical instruction schemes, where certain instructions (e.g., system-level directives) are expected to take precedence over others (e.g., user messages). Yet, we lack a…

Computation and Language · Computer Science 2026-03-23 Yilin Geng , Haonan Li , Honglin Mu , Xudong Han , Timothy Baldwin , Omri Abend , Eduard Hovy , Lea Frermann

As large language models (LLMs) are rapidly advancing and achieving near-human capabilities on specific tasks, aligning them with human values is becoming more urgent. In scenarios where LLMs outperform humans, we face a weak-to-strong…

Computation and Language · Computer Science 2025-03-04 Yougang Lyu , Lingyong Yan , Zihan Wang , Dawei Yin , Pengjie Ren , Maarten de Rijke , Zhaochun Ren

The rapid proliferation of generative AI, especially large language models, has led to their integration into a variety of applications. A key phenomenon known as weak-to-strong generalization - where a strong model trained on a weak…

Machine Learning · Computer Science 2025-01-03 Martin Pawelczyk , Lillian Sun , Zhenting Qi , Aounon Kumar , Himabindu Lakkaraju

Aligning language models (LMs) with human preferences has become a key area of research, enabling these models to meet diverse user needs better. Inspired by weak-to-strong generalization, where a strong LM fine-tuned on labels generated by…

Computation and Language · Computer Science 2025-03-07 Wenhong Zhu , Zhiwei He , Xiaofeng Wang , Pengfei Liu , Rui Wang

This paper introduces the counter-intuitive generalization results of overfitting pre-trained large language models (LLMs) on very small datasets. In the setting of open-ended text generation, it is well-documented that LLMs tend to…

Computation and Language · Computer Science 2025-02-27 Fredrik Carlsson , Fangyu Liu , Daniel Ward , Murathan Kurfali , Joakim Nivre

With Large Language Models (LLMs) rapidly approaching and potentially surpassing human-level performance, it has become imperative to develop approaches capable of effectively supervising and enhancing these powerful models using smaller,…

Machine Learning · Computer Science 2025-06-06 Aakriti Agrawal , Mucong Ding , Zora Che , Chenghao Deng , Anirudh Satheesh , Bang An , Bayan Bruss , John Langford , Furong Huang

Recent studies on the safety alignment of large language models (LLMs) have revealed that existing approaches often operate superficially, leaving models vulnerable to various adversarial attacks. Despite their significance, these studies…

Cryptography and Security · Computer Science 2025-06-02 Jianwei Li , Jung-Eun Kim

Online education platforms, leveraging the internet to distribute education resources, seek to provide convenient education but often fall short in real-time communication with students. They often struggle to address the diverse obstacles…

Artificial Intelligence · Computer Science 2024-04-29 Qingyao Li , Lingyue Fu , Weiming Zhang , Xianyu Chen , Jingwei Yu , Wei Xia , Weinan Zhang , Ruiming Tang , Yong Yu

While the wide adoption of refusal training in large language models (LLMs) has showcased improvements in model safety, recent works have highlighted shortcomings due to the shallow nature of these alignment methods. To this end, the work…

Machine Learning · Computer Science 2026-04-17 Pankayaraj Pathmanathan , Furong Huang

Weak-to-strong generalization (W2SG) refers to the phenomenon where a strong student model, trained on a dataset labeled by a weak teacher, ultimately outperforms the teacher on the target task. Recent studies attribute this performance…

Machine Learning · Computer Science 2025-09-30 Gengze Xu , Wei Yao , Ziqiao Wang , Yong Liu

The rapid development of large language models (LLMs) has not only provided numerous opportunities but also presented significant challenges. This becomes particularly evident when LLMs inadvertently generate harmful or toxic content,…

Computation and Language · Computer Science 2024-02-20 Kai Chen , Chunwei Wang , Kuo Yang , Jianhua Han , Lanqing Hong , Fei Mi , Hang Xu , Zhengying Liu , Wenyong Huang , Zhenguo Li , Dit-Yan Yeung , Lifeng Shang , Xin Jiang , Qun Liu

A hallmark property of explainable AI models is the ability to teach other agents, communicating knowledge of how to perform a task. While Large Language Models perform complex reasoning by generating explanations for their predictions, it…

Computation and Language · Computer Science 2023-11-15 Swarnadeep Saha , Peter Hase , Mohit Bansal

Natural Language Inference (NLI) models are known to learn from biases and artefacts within their training data, impacting how well they generalise to other unseen datasets. Existing de-biasing approaches focus on preventing the models from…

Computation and Language · Computer Science 2022-05-03 Joe Stacey , Yonatan Belinkov , Marek Rei

Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities in processing both visual and textual information. However, the critical challenge of alignment between visual and textual representations is not fully…

Computer Vision and Pattern Recognition · Computer Science 2025-09-24 Dong Shu , Haiyan Zhao , Jingyu Hu , Weiru Liu , Ali Payani , Lu Cheng , Mengnan Du

Large language models exhibit systematic vulnerabilities to adversarial attacks despite extensive safety alignment. We provide a mechanistic analysis revealing that position-dependent gradient weakening during autoregressive training…

Machine Learning · Computer Science 2025-11-18 Thong Bach , Dung Nguyen , Thao Minh Le , Truyen Tran

Large language models (LLMs) have exhibited impressive performance and surprising emergent properties. However, their effectiveness remains limited by the fixed context window of the transformer architecture, posing challenges for…

Computation and Language · Computer Science 2025-06-16 Tianqi Du , Haotian Huang , Yifei Wang , Yisen Wang

Recent years have witnessed remarkable progress made in large language models (LLMs). Such advancements, while garnering significant attention, have concurrently elicited various concerns. The potential of these models is undeniably vast;…

Computation and Language · Computer Science 2023-09-27 Tianhao Shen , Renren Jin , Yufei Huang , Chuang Liu , Weilong Dong , Zishan Guo , Xinwei Wu , Yan Liu , Deyi Xiong

Large Language Models (LLMs) have shown powerful performance and development prospects and are widely deployed in the real world. However, LLMs can capture social biases from unprocessed training data and propagate the biases to downstream…

Computation and Language · Computer Science 2024-02-22 Yingji Li , Mengnan Du , Rui Song , Xin Wang , Ying Wang

Misalignment between model predictions and intended usage can be detrimental for the deployment of computer vision models. The issue is exacerbated when the task involves complex structured outputs, as it becomes harder to design procedures…

Computer Vision and Pattern Recognition · Computer Science 2023-02-17 André Susano Pinto , Alexander Kolesnikov , Yuge Shi , Lucas Beyer , Xiaohua Zhai