相关论文: Instructions Shape Production of Language, not Pro…

Language models are trained to follow instructions, but they are also powerful pattern completers. What happens when these two objectives conflict? We construct conversations in which a user instruction to behave in a target way T (e.g.,…

We analyze reasoning in language models during task-specific fine-tuning and draws parallel between reasoning tokens--intermediate steps generated while solving problem and the human working memory. Drawing from cognitive science, we align…

Human explanations of natural language, rationales, form a tool to assess whether models learn a label for the right reasons or rely on dataset-specific shortcuts. Sufficiency is a common metric for estimating the informativeness of…

Modality following is the ability to selectively leverage multimodal contexts based on user instructions. It is fundamental to the safety and reliability of multimodal large language models (MLLMs) in real-world deployments. However, the…

Large language models are able to perform a task by conditioning on a few input-output demonstrations - a paradigm known as in-context learning. We show that language models can explicitly infer an underlying task from a few demonstrations…

Instruction tuning commonly means finetuning a language model on instruction-response pairs. We discover two forms of adaptation (tuning) that are deficient compared to instruction tuning, yet still yield instruction following; we call this…

Large Language Models (LLMs) have achieved remarkable success, where instruction tuning is the critical step in aligning LLMs with user intentions. In this work, we investigate how the instruction tuning adjusts pre-trained models with a…

How do language models learn to make predictions during pre-training? To study this, we extract learning curves from five autoregressive English language model pre-training runs, for 1M unseen tokens in context. We observe that the language…

Learning to construct text representations in end-to-end systems can be difficult, as natural languages are highly compositional and task-specific annotated datasets are often limited in size. Methods for directly supervising language…

Language models are often trained to maximize the likelihood of the next token given past tokens in the training dataset. However, during inference time, they are utilized differently, generating text sequentially and auto-regressively by…

Recent language models have achieved impressive performance in natural language tasks by incorporating instructions with task input during fine-tuning. Since all samples in the same natural language task can be explained with the same task…

Instruction tuning is commonly assumed to endow language models with a domain-general ability to follow instructions, yet the underlying mechanism remains poorly understood. Does instruction-following rely on a universal mechanism or…

Large language models (LLMs) have shown impressive performance in following natural language instructions to solve unseen tasks. However, it remains unclear whether models truly understand task definitions and whether the human-written…

Despite the fact that large language models (LLMs) show exceptional skill in instruction following tasks, this strength can turn into a vulnerability when the models are required to disregard certain instructions. Instruction-following…

Instruction-tuning is a widely adopted finetuning method that enables large language models (LLMs) to generate output that more closely resembles human responses. However, no studies have shown that instruction-tuning actually teaches LLMs…

Instruction tuning enables pretrained language models to perform new tasks from inference-time natural language descriptions. These approaches rely on vast amounts of human supervision in the form of crowdsourced datasets or user…

The ability to follow instructions is crucial for numerous real-world applications of language models. In pursuit of deeper insights and more powerful capabilities, we derive instruction-specific vector representations from language models…

Instruction tuning -- tuning large language models on instruction-output pairs -- is a promising technique for making models better adapted to the real world. Yet, the key factors driving the model's capability to understand and follow…

Despite the recent success of instruction-tuned language models and their ubiquitous usage, very little is known of how models process instructions internally. In this work, we address this gap from a mechanistic point of view by…

Language model-based instruction-following systems have lately shown increasing performance on many benchmark tasks, demonstrating the capability of adapting to a broad variety of instructions. However, such systems are often not designed…