English
Related papers

Related papers: Easy-to-Hard Generalization: Scalable Alignment Be…

200 papers

How can "weak teacher models" such as average human annotators or existing AI systems, effectively supervise LLMs to improve performance on hard reasoning tasks, especially those that challenge and requires expertise or daily practice from…

Machine Learning · Computer Science 2025-02-26 Xuan He , Da Yin , Nanyun Peng

How can we train models to perform well on hard test data when hard training data is by definition difficult to label correctly? This question has been termed the scalable oversight problem and has drawn increasing attention as language…

Computation and Language · Computer Science 2024-06-06 Peter Hase , Mohit Bansal , Peter Clark , Sarah Wiegreffe

Two core challenges of alignment are 1) scalable oversight and 2) accounting for the dynamic nature of human values. While solutions like recursive reward modeling address 1), they do not simultaneously account for 2). We sketch a roadmap…

Artificial Intelligence · Computer Science 2025-03-19 Florian Mai , David Kaczér , Nicholas Kluge Corrêa , Lucie Flek

Large language models (LLMs) are now rapidly advancing and surpassing human abilities on many natural language tasks. However, aligning these super-human LLMs with human knowledge remains challenging because the supervision signals from…

Computation and Language · Computer Science 2024-06-28 Yue Guo , Yi Yang

As AI capabilities increasingly surpass human proficiency in complex tasks, current alignment techniques, including SFT and RLHF, face fundamental challenges in ensuring reliable oversight. These methods rely on direct human assessment and…

Artificial Intelligence · Computer Science 2026-01-16 Xueru Wen , Jie Lou , Xinyu Lu , Junjie Yang , Yanjiang Liu , Yaojie Lu , Debing Zhang , Xing Yu

The rapid advancement of artificial intelligence systems has brought the challenge of AI alignment to the forefront of research, particularly in complex decision-making and task execution. As these systems surpass human-level performance in…

Artificial Intelligence · Computer Science 2024-09-12 Mehrdad Zakershahrak , Samira Ghodratnama

As large language models (LLMs) continue to advance, ensuring their alignment with human values becomes increasingly critical. Traditional alignment methods heavily rely on human feedback to fine-tune models. With the emergence of…

Computation and Language · Computer Science 2025-03-26 Ruimeng Ye , Yang Xiao , Bo Hui

Widely used alignment techniques, such as reinforcement learning from human feedback (RLHF), rely on the ability of humans to supervise model behavior - for example, to evaluate whether a model faithfully followed instructions or generated…

We have witnessed superhuman intelligence thanks to the fast development of large language models and multimodal language models. As the application of such superhuman models becomes more and more popular, a critical question arises here:…

Computation and Language · Computer Science 2024-12-24 Minlie Huang , Yingkang Wang , Shiyao Cui , Pei Ke , Jie Tang

Aligning powerful AI models on tasks that surpass human evaluation capabilities is the central problem of \textbf{superalignment}. To address this problem, weak-to-strong generalization aims to elicit the capabilities of strong models…

Machine Learning · Computer Science 2025-03-07 Junhao Shi , Qinyuan Cheng , Zhaoye Fei , Yining Zheng , Qipeng Guo , Xipeng Qiu

Common methods for aligning already-capable models with desired behavior rely on the ability of humans to provide supervision. However, future superhuman models will surpass the capability of humans. Therefore, humans will only be able to…

Computation and Language · Computer Science 2025-01-24 Hao Lang , Fei Huang , Yongbin Li

A leading proposal for aligning artificial superintelligence (ASI) is to use AI agents to automate an increasing fraction of alignment research as capabilities improve. We argue that, even when research agents are not scheming to…

Artificial Intelligence · Computer Science 2026-05-18 Aleksandr Bowkis , Marie Davidsen Buhl , Jacob Pfau , Geoffrey Irving

As artificial intelligence (AI) systems approach and surpass expert human performance across a broad range of tasks, obtaining high-quality human supervision for evaluation and training becomes increasingly challenging. Our focus is on…

Machine Learning · Computer Science 2026-02-25 Ren Yin , Takashi Ishida , Masashi Sugiyama

We investigate how well large language models (LLMs) generalize across different task difficulties, a key question for effective data curation and evaluation. Existing research is mixed regarding whether training on easier or harder data…

Computation and Language · Computer Science 2025-11-27 Yeganeh Kordi , Nihal V. Nayak , Max Zuo , Ilana Nguyen , Stephen H. Bach

Accurate estimation of item (question or task) difficulty is critical for educational assessment but suffers from the cold start problem. While Large Language Models demonstrate superhuman problem-solving capabilities, it remains an open…

Computation and Language · Computer Science 2026-05-12 Ming Li , Han Chen , Yunze Xiao , Jian Chen , Hong Jiao , Tianyi Zhou

The emergence of large language models (LLMs) has sparked the possibility of about Artificial Superintelligence (ASI), a hypothetical AI system surpassing human intelligence. However, existing alignment paradigms struggle to guide such…

Machine Learning · Computer Science 2024-12-30 HyunJin Kim , Xiaoyuan Yi , Jing Yao , Jianxun Lian , Muhua Huang , Shitong Duan , JinYeong Bak , Xing Xie

As AI systems become more intelligent and their behavior becomes more challenging to assess, they may learn to game the flaws of human feedback instead of genuinely striving to follow instructions; however, this risk can be mitigated by…

Artificial Intelligence · Computer Science 2023-12-19 Joshua Clymer , Garrett Baker , Rohan Subramani , Sam Wang

Recent advances in AI -- including generative approaches -- have resulted in technology that can support humans in scientific discovery and forming decisions, but may also disrupt democracies and target individuals. The responsible use of…

As the field progresses toward Artificial General Intelligence (AGI), there is a pressing need for more comprehensive and insightful evaluation frameworks that go beyond aggregate performance metrics. This paper introduces a unified rating…

This paper presents a follow-up study to OpenAI's recent superalignment work on Weak-to-Strong Generalization (W2SG). Superalignment focuses on ensuring that high-level AI systems remain consistent with human values and intentions when…

Computation and Language · Computer Science 2024-02-02 Jitao Sang , Yuhang Wang , Jing Zhang , Yanxu Zhu , Chao Kong , Junhong Ye , Shuyu Wei , Jinlin Xiao
‹ Prev 1 2 3 10 Next ›