English
Related papers

Related papers: Self-Supervised Alignment with Mutual Information:…

200 papers

Instruction fine-tuning is crucial for today's large language models (LLMs) to learn to follow instructions and align with human preferences. Conventionally, supervised data, including the instruction and the correct response, is required…

Computation and Language · Computer Science 2024-09-04 Juncheng Xie , Shensian Syu , Hung-yi Lee

We investigate the extent to which modern, neural language models are susceptible to structural priming, the phenomenon whereby the structure of a sentence makes the same structure more probable in a follow-up sentence. We explore how…

Computation and Language · Computer Science 2022-06-30 Arabella Sinclair , Jaap Jumelet , Willem Zuidema , Raquel Fernández

Large language models (LLMs) have been significantly improved by instruction fine-tuning, but still lack transparency and the ability to utilize up-to-date knowledge and information. In this work, we propose search-augmented instruction…

Computation and Language · Computer Science 2023-06-27 Hongyin Luo , Yung-Sung Chuang , Yuan Gong , Tianhua Zhang , Yoon Kim , Xixin Wu , Danny Fox , Helen Meng , James Glass

Direct alignment algorithms (DAAs), such as direct preference optimization (DPO), have become popular alternatives for Reinforcement Learning from Human Feedback (RLHF) due to their simplicity, efficiency, and stability. However, the…

Machine Learning · Computer Science 2024-10-15 Jongwoo Ko , Saket Dingliwal , Bhavana Ganesh , Sailik Sengupta , Sravan Bodapati , Aram Galstyan

Aligning large language models (LLMs) with human preferences becomes a key component to obtaining state-of-the-art performance, but it yields a huge cost to construct a large human-annotated preference dataset. To tackle this problem, we…

Machine Learning · Computer Science 2025-03-05 Dongyoung Kim , Kimin Lee , Jinwoo Shin , Jaehyung Kim

Preference optimization methods have been successfully applied to improve not only the alignment of large language models (LLMs) with human values, but also specific natural language tasks such as summarization and stylistic continuations.…

Machine Learning · Computer Science 2025-02-06 Salem Lahlou , Abdalgader Abubaker , Hakim Hacid

Large language models (LLMs) have demonstrated remarkable capabilities in handling complex dialogue tasks without requiring use case-specific fine-tuning. However, analyzing live dialogues in real-time necessitates low-latency processing…

Computation and Language · Computer Science 2025-03-10 Xuanqing Liu , Luyang Kong , Wei Niu , Afshin Khashei , Belinda Zeng , Steve Johnson , Jon Jay , Davor Golac , Matt Pope

Despite being trained specifically to follow user instructions, today's instructiontuned language models perform poorly when instructed to produce random outputs. For example, when prompted to pick a number uniformly between one and ten…

Computation and Language · Computer Science 2024-08-09 Yiming Zhang , Avi Schwarzschild , Nicholas Carlini , Zico Kolter , Daphne Ippolito

Large Language Models (LLMs) have achieved excellent performances in various tasks. However, fine-tuning an LLM requires extensive supervision. Human, on the other hand, may improve their reasoning abilities by self-thinking without…

Computation and Language · Computer Science 2022-10-26 Jiaxin Huang , Shixiang Shane Gu , Le Hou , Yuexin Wu , Xuezhi Wang , Hongkun Yu , Jiawei Han

We introduce Synthetic Bootstrapped Pretraining (SBP), a language model (LM) pretraining procedure that first learns a model of relations between documents from the pretraining dataset and then leverages it to synthesize a vast new corpus…

Computation and Language · Computer Science 2025-12-16 Zitong Yang , Aonan Zhang , Hong Liu , Tatsunori Hashimoto , Emmanuel Candès , Chong Wang , Ruoming Pang

Model-based evaluation is at the heart of successful model development -- as a reward model for training, and as a replacement for human evaluation. To train such evaluators, the standard approach is to collect a large amount of human…

Controlling the syntactic structure of text generated by language models is valuable for applications requiring clarity, stylistic consistency, or interpretability, yet it remains a challenging task. In this paper, we argue that sampling…

Computation and Language · Computer Science 2025-06-10 Vicky Xefteri , Tim Vieira , Ryan Cotterell , Afra Amini

Aligning Large Language Models (LLMs) with human intentions and values is crucial yet challenging. Current methods primarily rely on human preferences, which are costly and insufficient in capturing nuanced feedback expressed in natural…

Computation and Language · Computer Science 2024-06-12 Chi Hu , Yimin Hu , Hang Cao , Tong Xiao , Jingbo Zhu

Despite its flexibility to learn diverse inductive biases in machine learning programs, meta learning (i.e., learning to learn) has long been recognized to suffer from poor scalability due to its tremendous compute/memory costs, training…

Machine Learning · Computer Science 2023-10-24 Sang Keun Choe , Sanket Vaibhav Mehta , Hwijeen Ahn , Willie Neiswanger , Pengtao Xie , Emma Strubell , Eric Xing

The field of preference optimization has made outstanding contributions to the alignment of language models with human preferences. Despite these advancements, recent methods still rely heavily on substantial paired (labeled) feedback data,…

Machine Learning · Computer Science 2026-02-20 Seonggyun Lee , Sungjun Lim , Seojin Park , Soeun Cheon , Kyungwoo Song

Aligning large language models (LLMs) with human values is a vital task for LLM practitioners. Current alignment techniques have several limitations: (1) requiring a large amount of annotated data; (2) demanding heavy human involvement; (3)…

Computation and Language · Computer Science 2024-01-17 Hongyi Guo , Yuanshun Yao , Wei Shen , Jiaheng Wei , Xiaoying Zhang , Zhaoran Wang , Yang Liu

A crucial consideration when developing and deploying Large Language Models (LLMs) is the human values to which these models are aligned. In the constitutional framework of alignment models are aligned to a set of principles (the…

Machine Learning · Computer Science 2026-01-27 Henry Bell , Lara Neubauer da Costa Schertel , Bochu Ding , Brandon Fain

As AI systems approach superhuman capabilities, scalable oversight increasingly relies on LLM-as-a-judge frameworks where models evaluate and guide each other's training. A core assumption is that binary preference labels provide only…

Machine Learning · Computer Science 2026-03-13 Isotta Magistrali , Frédéric Berdoz , Sam Dauncey , Roger Wattenhofer

Preference optimization, particularly through Reinforcement Learning from Human Feedback (RLHF), has achieved significant success in aligning Large Language Models (LLMs) to adhere to human intentions. Unlike offline alignment with a fixed…

Machine Learning · Computer Science 2024-11-06 Shenao Zhang , Donghan Yu , Hiteshi Sharma , Han Zhong , Zhihan Liu , Ziyi Yang , Shuohang Wang , Hany Hassan , Zhaoran Wang

To obtain high-quality sentence embeddings from pretrained language models (PLMs), they must either be augmented with additional pretraining objectives or finetuned on a large set of labeled text pairs. While the latter approach typically…

Computation and Language · Computer Science 2021-10-05 Timo Schick , Hinrich Schütze