English
Related papers

Related papers: Teaching an Old LLM Secure Coding: Localized Prefe…

200 papers

Direct Preference Optimization (DPO), which derives reward signals directly from pairwise preference data, has shown its effectiveness on aligning Large Language Models (LLMs) with human preferences. Despite its widespread use across…

Computation and Language · Computer Science 2024-04-09 Duanyu Feng , Bowen Qin , Chen Huang , Zheng Zhang , Wenqiang Lei

We introduce ORPO-Distill, a general-purpose method for cross-architecture LLM distillation that formulates the problem as a preference optimization task. Unlike standard CoT distillation, the approach transfers knowledge through diverse…

Machine Learning · Computer Science 2025-09-30 Aasheesh Singh , Vishal Vaddina , Dagnachew Birru

Large Language Models (LLMs) excel at understanding natural language but struggle with optimisation tasks involving multiple constraints and user-defined preferences, which commonly arise in domains such as robotics. We propose a hybrid…

Artificial Intelligence · Computer Science 2026-05-29 Pedro Orvalho , Marta Kwiatkowska , Guillem Alenyà , Felip Manyà

Large language models (LLMs) achieve remarkable performance in code generation tasks. However, a significant performance disparity persists between popular programming languages (e.g., Python, C++) and others. To address this capability…

Computation and Language · Computer Science 2025-12-05 Haoyuan Wu , Rui Ming , Jilong Gao , Hangyu Zhao , Xueyi Chen , Yikai Yang , Haisheng Zheng , Zhuolun He , Bei Yu

While recent code-specific large language models (LLMs) have greatly enhanced their code generation capabilities, the safety of these models remains under-explored, posing potential risks as insecure code generated by these models may…

Cryptography and Security · Computer Science 2025-06-09 Xiangzhe Xu , Zian Su , Jinyao Guo , Kaiyuan Zhang , Zhenting Wang , Xiangyu Zhang

Many algorithms for aligning LLMs with human preferences assume that human preferences are binary and deterministic. However, human preferences can vary across individuals, and therefore should be represented distributionally. In this work,…

Machine Learning · Computer Science 2024-12-31 Hiroki Furuta , Kuang-Huei Lee , Shixiang Shane Gu , Yutaka Matsuo , Aleksandra Faust , Heiga Zen , Izzeddin Gur

We consider the problem of revealing/sharing data in an efficient and secure way via a compact representation. The representation should ensure reliable reconstruction of the desired features/attributes while still preserve privacy of the…

Information Theory · Computer Science 2016-05-09 Kittipong Kittichokechai , Giuseppe Caire

Direct Preference Optimization (DPO) has emerged as a de-facto approach for aligning language models with human preferences. Recent work has shown DPO's effectiveness relies on training data quality. In particular, clear quality differences…

Machine Learning · Computer Science 2025-01-28 Nirav Diwan , Tolga Ergen , Dongsub Shim , Honglak Lee

Large Language Models (LLMs) have demonstrated unprecedented generative capabilities, yet their alignment with human values remains critical for ensuring helpful and harmless deployments. While Reinforcement Learning from Human Feedback…

Direct Preference Optimization (DPO) has become a widely used training method for the instruction fine-tuning of large language models (LLMs). In this work, we explore an under-investigated aspect of DPO - its dependency on the reference…

Computation and Language · Computer Science 2024-08-23 Yixin Liu , Pengfei Liu , Arman Cohan

Constrained codes are used to prevent errors from occurring in various data storage and data transmission systems. They can help in increasing the storage density of magnetic storage devices, in managing the lifetime of electronic storage…

Information Theory · Computer Science 2022-09-07 Ahmed Hareedy , Beyza Dabak , Robert Calderbank

A major challenge in aligning large language models (LLMs) with human preferences is the issue of distribution shift. LLM alignment algorithms rely on static preference datasets, assuming that they accurately represent real-world user…

Machine Learning · Computer Science 2026-01-16 Zaiyan Xu , Sushil Vemuri , Kishan Panaganti , Dileep Kalathil , Rahul Jain , Deepak Ramachandran

Adapting large language models (LLMs) for specific tasks usually involves fine-tuning through reinforcement learning with human feedback (RLHF) on preference data. While these data often come from diverse labelers' groups (e.g., different…

Computation and Language · Computer Science 2024-05-31 Shyam Sundhar Ramesh , Yifan Hu , Iason Chaimalas , Viraj Mehta , Pier Giuseppe Sessa , Haitham Bou Ammar , Ilija Bogunovic

Traditional preference tuning methods for LLMs/Visual Generative Models often rely solely on reward model labeling, which can be opaque, offer limited insights into the rationale behind preferences, and are prone to issues such as reward…

Machine Learning · Computer Science 2026-01-13 Hanyang Zhao , Haoxian Chen , Yucheng Guo , Genta Indra Winata , Tingting Ou , Ziyu Huang , David D. Yao , Wenpin Tang

As large language models (LLMs) are increasingly applied across various domains, enhancing safety while maintaining the helpfulness of LLMs has become a critical challenge. Recent studies solve this problem through safety-constrained online…

Computation and Language · Computer Science 2025-06-04 Yupeng Qi , Ziyu Lyu , Min Yang , Yanlin Wang , Lu Bai , Lixin Cui

We argue that when it comes to producing secure code with AI, the prevailing "fighting fire with fire" approach -- using probabilistic AI-based checkers or attackers to secure probabilistically generated code -- fails to address the long…

Cryptography and Security · Computer Science 2026-02-10 Benjamin Livshits

Code Large Language Models (Code LLMs) have been increasingly used by developers to boost productivity, but they often generate vulnerable code. Thus, there is an urgent need to ensure that code generated by Code LLMs is correct and secure.…

Cryptography and Security · Computer Science 2024-07-23 Yanjun Fu , Ethan Baker , Yu Ding , Yizheng Chen

The ability of LLMs to represent diverse perspectives is critical as they increasingly impact society. However, recent studies reveal that alignment algorithms such as RLHF and DPO significantly reduce the diversity of LLM outputs. Not only…

Computation and Language · Computer Science 2025-11-13 Stewart Slocum , Asher Parker-Sartori , Dylan Hadfield-Menell

Human preference alignment is critical in building powerful and reliable large language models (LLMs). However, current methods either ignore the multi-dimensionality of human preferences (e.g. helpfulness and harmlessness) or struggle with…

Machine Learning · Computer Science 2024-10-14 Xingzhou Lou , Junge Zhang , Jian Xie , Lifeng Liu , Dong Yan , Kaiqi Huang

Preference alignment methods are increasingly critical for steering large language models (LLMs) to generate outputs consistent with human values. While recent approaches often rely on synthetic data generated by LLMs for scalability and…

Computation and Language · Computer Science 2025-10-21 Mingye Zhu , Yi Liu , Zheren Fu , Yongdong Zhang , Zhendong Mao