中文
相关论文

相关论文: Do as I Say, Not as I Do: Instruction-Induction Co…

200 篇论文

Large language models' behavior is often shaped by instructions such as system prompts, refusal boundaries, privacy constraints, and tool-use rules that must hold at inference time. Yet in practice these constraints can be violated under…

计算与语言 · 计算机科学 2026-03-27 Vitoria Guardieiro , Avishree Khare , Adam Stein , Eric Wong

Recent work exploring the capabilities of pre-trained large language models (LLMs) has demonstrated their ability to act as general pattern machines by completing complex token sequences representing a wide array of tasks, including…

计算机与社会 · 计算机科学 2024-03-25 Seyed Parsa Neshaei , Richard Lee Davis , Adam Hazimeh , Bojan Lazarevski , Pierre Dillenbourg , Tanja Käser

Current instruction-tuned language models are exclusively trained with textual preference data and thus are often not aligned with the unique requirements of other modalities, such as speech. To better align language models with the speech…

In this paper, we introduce the Instruction Following Score (IFS), a metric that detects language models' ability to follow instructions. The metric has a dual purpose. First, IFS can be used to distinguish between base and instruct models.…

计算与语言 · 计算机科学 2023-07-10 Waseem AlShikh , Manhal Daaboul , Kirk Goddard , Brock Imel , Kiran Kamble , Parikshith Kulkarni , Melisa Russak

Training large language models to follow instructions makes them perform better on a wide range of tasks and generally become more helpful. However, a perfectly helpful model will follow even the most malicious instructions and readily…

计算与语言 · 计算机科学 2024-03-20 Federico Bianchi , Mirac Suzgun , Giuseppe Attanasio , Paul Röttger , Dan Jurafsky , Tatsunori Hashimoto , James Zou

Aligned instruction following models can better fulfill user requests than their unaligned counterparts. However, it has been shown that there is a length bias in evaluation of such models, and that training algorithms tend to exploit this…

计算与语言 · 计算机科学 2024-06-26 Weizhe Yuan , Ilia Kulikov , Ping Yu , Kyunghyun Cho , Sainbayar Sukhbaatar , Jason Weston , Jing Xu

Instruction tuning plays a crucial role in shaping the outputs of language models (LMs) to desired styles. In this work, we propose a simple yet effective method, Instruction Modelling (IM), which trains LMs by applying a loss function to…

计算与语言 · 计算机科学 2024-10-04 Zhengyan Shi , Adam X. Yang , Bin Wu , Laurence Aitchison , Emine Yilmaz , Aldo Lipani

A typical approach developers follow to influence an LLM's behavior in an application is through careful manipulation of the prompt, such as by adding or modifying instructions. However, merely adding more instructions provides little…

人工智能 · 计算机科学 2025-10-17 Ben Elder , Evelyn Duesterwald , Vinod Muthusamy

Large Language Models (LLMs) have demonstrated exceptional performance in code generation tasks and have become indispensable programming assistants for developers. However, existing code generation benchmarks primarily assess the…

软件工程 · 计算机科学 2025-11-25 Peiding Wang , Li Zhang , Fang Liu , Lin Shi , Minxiao Li , Bo Shen , An Fu

Reliably ensuring Large Language Models (LLMs) follow complex instructions is a critical challenge, as existing benchmarks often fail to reflect real-world use or isolate compliance from task success. We introduce MOSAIC (MOdular Synthetic…

人工智能 · 计算机科学 2026-01-27 Alberto Purpura , Li Wang , Sahil Badyal , Eugenio Beaufrand , Adam Faulkner

Large Language Models (LLMs) for public use require continuous pre-training to remain up-to-date with the latest data. The models also need to be fine-tuned with specific instructions to maintain their ability to follow instructions…

计算与语言 · 计算机科学 2024-10-15 Ishan Jindal , Chandana Badrinath , Pranjal Bharti , Lakkidi Vinay , Sachin Dev Sharma

Instruction tuning aligns the response of large language models (LLMs) with human preferences. Despite such efforts in human--LLM alignment, we find that instruction tuning does not always make LLMs human-like from a cognitive modeling…

计算与语言 · 计算机科学 2024-04-16 Tatsuki Kuribayashi , Yohei Oseki , Timothy Baldwin

Large language models should follow hierarchical instructions where system prompts override user inputs, yet recent work shows they often ignore this rule while strongly obeying social cues such as authority or consensus. We extend these…

计算与语言 · 计算机科学 2025-12-19 Siqi Zeng

Instruction tuning has been widely adopted to ensure large language models (LLMs) follow user instructions effectively. The resulting instruction-following capabilities of LLMs heavily rely on the instruction datasets used for tuning.…

人工智能 · 计算机科学 2025-02-27 Zhangchen Xu , Fengqing Jiang , Luyao Niu , Bill Yuchen Lin , Radha Poovendran

Recent advances in applying reinforcement learning (RL) to large language models (LLMs) have led to substantial progress. In particular, a series of remarkable yet often counterintuitive phenomena have been reported in LLMs, exhibiting…

机器学习 · 计算机科学 2025-09-03 Haoze Wu , Cheng Wang , Wenshuo Zhao , Junxian He

Instruction-following made modern large language models (LLMs) helpful assistants. However, the key to taming LLMs on complex instructions remains mysterious, for that there are huge gaps between models trained by open-source community and…

计算与语言 · 计算机科学 2025-09-30 Kaikai An , Li Sheng , Ganqu Cui , Shuzheng Si , Ning Ding , Yu Cheng , Baobao Chang

While Large Language Models (LLMs) are widely documented to be sensitive to minor prompt perturbations and prone to sycophantic alignment, their robustness in consequential, rule-bound decision-making remains under-explored. We uncover a…

人工智能 · 计算机科学 2026-04-07 Jon Chun , Katherine Elkins

Large Language Models (LLMs) have achieved remarkable success in tasks requiring complex reasoning, such as code generation, mathematical problem solving, and algorithmic synthesis -- especially when aided by reasoning tokens and…

计算与语言 · 计算机科学 2025-06-13 Jaechul Roh , Varun Gandhi , Shivani Anilkumar , Arin Garg

Recent advances in reasoning with large language models (LLMs) have demonstrated strong performance on complex mathematical tasks, including combinatorial optimization. Techniques such as Chain-of-Thought and In-Context Learning have…

人工智能 · 计算机科学 2025-09-17 Marylou Fauchard , Florian Carichon , Margarida Carvalho , Golnoosh Farnadi

Large Language Models (LLMs) are increasingly deployed in agentic and retrieval-augmented generation (RAG) systems, where they must execute user-specified tasks over externally provided reference text. In practice, such context is often…

人工智能 · 计算机科学 2026-05-29 Zeli Su , Zhankai Xu , Tianlei Chen , Longfei Zheng , Xiaolu Zhang , Jun Zhou , Wentao Zhang