Related papers: Mind the Gap: Data Rewriting for Stable Off-Policy…

Supervised Fine-Tuning (SFT) followed by Reinforcement Learning (RL) has emerged as the standard post-training paradigm for large language models (LLMs). However, the conventional SFT process, driven by Cross-Entropy (CE) loss, often…

Large language model fine-tuning techniques typically depend on extensive labeled data, external guidance, and feedback, such as human alignment, scalar rewards, and demonstration. However, in practical application, the scarcity of specific…

Large language models (LLMs) have achieved remarkable success across various tasks but face deployment challenges due to their massive computational demands. While post-training pruning methods like SparseGPT and Wanda can effectively…

Supervised Fine-Tuning (SFT) adapts pre-trained Large Language Models (LLMs) to domain-specific instructions by training on a carefully curated subset of high-quality instruction-response pairs, typically drawn from a larger dataset that…

Large language models (LLMs) with one or more fine-tuning phases have become a necessary step to unlock various capabilities, enabling LLMs to follow natural language instructions or align with human preferences. However, it carries the…

Continual learning, enabling models to acquire new skills and knowledge without degrading existing capabilities, remains a fundamental challenge for foundation models. While on-policy reinforcement learning can reduce forgetting, it…

Although LLMs have demonstrated improved performance by scaling parallel test-time compute, doing so relies on generating reasoning paths that are both diverse and accurate. For challenging problems, the forking tokens that trigger diverse…

Data curation is a critical yet under-explored area in large language model (LLM) training. Existing methods, such as data selection and mixing, operate in an offline paradigm, detaching themselves from training. This separation introduces…

Catastrophic forgetting remains a major challenge when fine-tuning large language models (LLMs) on narrow, task-specific data, often degrading their general knowledge and reasoning abilities. We propose SA-SFT, a lightweight…

Recent advances in natural language processing (NLP) have opened up greater opportunities to enable fine-tuned large language models (LLMs) to behave as more powerful interactive agents through improved instruction-following ability.…

Widely used language models (LMs) are typically built by scaling up a two-stage training pipeline: a pre-training stage that uses a very large, diverse dataset of text and a fine-tuning (sometimes, 'alignment') stage that uses targeted…

Supervised Fine-Tuning (SFT) is essential for aligning Large Language Models (LLMs) with user intent, yet it is believed to suppress generative diversity. Although this reduction is frequently referenced, formal empirical testing of the…

Unconstrained fine-tuning of flow-matching Vision-Language-Action (VLA) models drives dense parameter overwrites, degrading pre-trained capabilities. We present Conservative Supervised Fine-Tuning (ConSFT), an optimization objective that…

Fine-tuning Large Language Models (LLMs) has emerged as a common practice for tailoring models to individual needs and preferences. The choice of datasets for fine-tuning can be diverse, introducing safety concerns regarding the potential…

The transition from fitting empirical data to achieving true human utility is fundamentally constrained by a granularity mismatch, where fine-grained autoregressive generation is often supervised by coarse or uniform signals. This position…

By pretraining on trillions of tokens, an LLM gains the capability of text generation. However, to enhance its utility and reduce potential harm, SFT and alignment are applied sequentially to the pretrained model. Because SFT and alignment…

Packing, initially utilized in the pre-training phase, is an optimization technique designed to maximize hardware resource efficiency by combining different training sequences to fit the model's maximum input length. Although it has…

Large Language Models (LLMs) typically rely on Supervised Fine-Tuning (SFT) to specialize in downstream tasks, with the Cross Entropy (CE) loss being the de facto choice. However, CE maximizes the likelihood of observed data without…

In natural language processing tasks, pure reinforcement learning (RL) fine-tuning methods often suffer from inefficient exploration and slow convergence; while supervised fine-tuning (SFT) methods, although efficient in training, have…

Supervised fine-tuning (SFT) data selection is commonly formulated as instance ranking: score each example and retain a top-$k$ subset. However, effective SFT training subsets are often produced through ordered curation recipes, where…