Related papers: Becoming self-instruct: introducing early stopping…

Supervised fine-tuning (SFT), supervised instruction tuning (SIT) and in-context learning (ICL) are three alternative, de facto standard approaches to few-shot learning. ICL has gained popularity recently with the advent of LLMs due to its…

Instruction-tuned Large Language Models (LLMs) have achieved remarkable performance across various benchmark tasks. While providing instructions to LLMs for guiding their generations is user-friendly, assessing their instruction-following…

Supervised fine-tuning (SFT) on instruction-following corpus is a crucial approach toward the alignment of large language models (LLMs). However, the performance of LLMs on standard knowledge and reasoning benchmarks tends to suffer from…

Following the initial flourishing of large language models (LLMs), there has been a surge in proposed large vision-language models (LVLMs) that integrate LLMs with vision capabilities. However, it has been observed that LVLMs, after tuning…

The visual representation of a pre-trained model prioritizes the classifiability on downstream tasks, while the widespread applications for pre-trained visual models have posed new requirements for representation interpretability. However,…

Supervised fine-tuning (SFT) is a pivotal approach to adapting large language models (LLMs) for downstream tasks; however, performance often suffers from the ``seesaw phenomenon'', where indiscriminate parameter updates yield progress on…

In this paper, we explore incremental few-shot object detection (iFSD), which incrementally learns novel classes using only a few examples without revisiting base classes. Previous iFSD works achieved the desired results by applying…

Pre-trained large language models (LLMs) can be tailored to adhere to human instructions through instruction tuning. However, due to shifts in the distribution of test-time data, they may not always execute instructions accurately,…

Large language models (LLMs) often exhibit undesirable behaviors, such as safety violations and hallucinations. Although inference-time steering offers a cost-effective way to adjust model behavior without updating its parameters, existing…

Large language models have demonstrated surprising ability to perform in-context learning, i.e., these models can be directly applied to solve numerous downstream tasks by conditioning on a prompt constructed by a few input-output examples.…

Despite more than a decade of reform efforts, students continue to experience difficulty understanding and applying statistical concepts. The predominant focus of reform has been on content, pedagogy, technology and assessment, with little…

As large language models (LLMs) transition from general knowledge retrieval to complex scientific discovery, their evaluation standards must also incorporate the rigorous norms of scientific inquiry. Existing benchmarks exhibit a critical…

In few-shot classification tasks (so-called episodes), a small set of labeled support samples is provided during inference to aid the classification of unlabeled query samples. Metric-based models typically operate by computing similarities…

The ability of Large Language Models (LLMs) to precisely follow complex and fine-grained lexical instructions is a cornerstone of their utility and controllability. However, evaluating this capability remains a significant challenge.…

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…

Supervised fine-tuning (SFT) provides the standard approach for teaching LLMs new behaviors from offline expert demonstrations. However, standard SFT uniformly fits all samples -- including those with low likelihood under the base model --…

Few-shot in-context learning (ICL) enables pre-trained language models to perform a previously-unseen task without any gradient-based training by feeding a small number of training examples as part of the input. ICL incurs substantial…

In this work, we explore "prompt tuning", a simple yet effective mechanism for learning "soft prompts" to condition frozen language models to perform specific downstream tasks. Unlike the discrete text prompts used by GPT-3, soft prompts…

Despite more than a decade of reform efforts, students continue to experience difficulty understanding and applying statistical concepts. The predominant focus of reform has been on content, pedagogy, technology and assessment, with little…

Foundation language models obtain the instruction-following ability through supervised fine-tuning (SFT). Diversity and complexity are considered critical factors of a successful SFT dataset, while their definitions remain obscure and lack…