Evolutionary Contrastive Distillation for Language Model Alignment

• URL: http://arxiv.org/abs/2410.07513v1
• Date: Thu, 10 Oct 2024 01:04:03 GMT
• Title: Evolutionary Contrastive Distillation for Language Model Alignment
• Authors: Julian Katz-Samuels, Zheng Li, Hyokun Yun, Priyanka Nigam, Yi Xu, Vaclav Petricek, Bing Yin, Trishul Chilimbi,
• Abstract summary: Evolutionary Contrastive Distillation (ECD) is a novel method for generating high-quality synthetic preference data. Our method yields a 7B model that exceeds the complex instruction-following performance of current SOTA 7B models.
• Score: 35.94171633370035
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The ability of large language models (LLMs) to execute complex instructions is essential for their real-world applications. However, several recent studies indicate that LLMs struggle with challenging instructions. In this paper, we propose Evolutionary Contrastive Distillation (ECD), a novel method for generating high-quality synthetic preference data designed to enhance the complex instruction-following capability of language models. ECD generates data that specifically illustrates the difference between a response that successfully follows a set of complex instructions and a response that is high-quality, but nevertheless makes some subtle mistakes. This is done by prompting LLMs to progressively evolve simple instructions to more complex instructions. When the complexity of an instruction is increased, the original successful response to the original instruction becomes a "hard negative" response for the new instruction, mostly meeting requirements of the new instruction, but barely missing one or two. By pairing a good response with such a hard negative response, and employing contrastive learning algorithms such as DPO, we improve language models' ability to follow complex instructions. Empirically, we observe that our method yields a 7B model that exceeds the complex instruction-following performance of current SOTA 7B models and is competitive even with open-source 70B models.

Related papers

• Constraint Back-translation Improves Complex Instruction Following of Large Language Models [55.60192044049083]
  Large language models (LLMs) struggle to follow instructions with complex constraints in format, length, etc. Previous works conduct post-training on complex instruction-response pairs generated by feeding complex instructions to advanced LLMs. We propose a novel data generation technique, constraint back-translation.
  arXiv  Detail & Related papers  (2024-10-31T17:42:26Z)
• TaCIE: Enhancing Instruction Comprehension in Large Language Models through Task-Centred Instruction Evolution [27.949846287419998]
  TaCIE redefines instruction evolution from merely evolving seed instructions to a more dynamic and comprehensive combination of elements. Applying TaCIE across multiple domains, LLMs fine-tuned with these evolved instructions have substantially outperformed those tuned with conventional methods.
  arXiv  Detail & Related papers  (2024-09-18T10:06:28Z)
• Phased Instruction Fine-Tuning for Large Language Models [12.037895935630882]
  Phased Instruction Fine-Tuning (Phased IFT) is proposed, based on the idea that learning to follow instructions is a gradual process. It assesses instruction difficulty using GPT-4, divides the instruction data into subsets of increasing difficulty, and uptrains the model sequentially on these subsets. Experiments with Llama-2 7B/13B/70B, Llama3 8/70B and Mistral-7B models using Alpaca data show that Phased IFT significantly outperforms One-off IFT.
  arXiv  Detail & Related papers  (2024-06-01T04:25:26Z)
• Conifer: Improving Complex Constrained Instruction-Following Ability of Large Language Models [23.17547206140014]
  We introduce Conifer, an instruction tuning dataset for large language models. We train models with Conifer to follow instructions with complex constraints. On several instruction-following benchmarks, our 7B model outperforms the state-of-the-art open-source 7B models.
  arXiv  Detail & Related papers  (2024-04-03T15:55:39Z)
• EasyInstruct: An Easy-to-use Instruction Processing Framework for Large Language Models [37.80143756214926]
  EasyInstruct is an easy-to-use instruction processing framework for Large Language Models (LLMs) EasyInstruct modularizes instruction generation, selection, and prompting, while also considering their combination and interaction.
  arXiv  Detail & Related papers  (2024-02-05T14:33:56Z)
• Can Large Language Models Understand Real-World Complex Instructions? [54.86632921036983]
  Large language models (LLMs) can understand human instructions, but struggle with complex instructions. Existing benchmarks are insufficient to assess LLMs' ability to understand complex instructions. We propose CELLO, a benchmark for evaluating LLMs' ability to follow complex instructions systematically.
  arXiv  Detail & Related papers  (2023-09-17T04:18:39Z)
• Instruction Position Matters in Sequence Generation with Large Language Models [67.87516654892343]
  Large language models (LLMs) are capable of performing conditional sequence generation tasks, such as translation or summarization. We propose enhancing the instruction-following capability of LLMs by shifting the position of task instructions after the input sentences.
  arXiv  Detail & Related papers  (2023-08-23T12:36:57Z)
• Improving Long-Horizon Imitation Through Instruction Prediction [93.47416552953075]
  In this work, we explore the use of an often unused source of auxiliary supervision: language. Inspired by recent advances in transformer-based models, we train agents with an instruction prediction loss that encourages learning temporally extended representations that operate at a high level of abstraction. In further analysis we find that instruction modeling is most important for tasks that require complex reasoning, while understandably offering smaller gains in environments that require simple plans.
  arXiv  Detail & Related papers  (2023-06-21T20:47:23Z)
• The Wisdom of Hindsight Makes Language Models Better Instruction Followers [84.9120606803906]
  Reinforcement learning has seen wide success in finetuning large language models to better align with instructions via human feedback. In this paper, we consider an alternative approach: converting feedback to instruction by relabeling the original one and training the model for better alignment in a supervised manner. We propose Hindsight Instruction Relabeling (HIR), a novel algorithm for aligning language models with instructions.
  arXiv  Detail & Related papers  (2023-02-10T12:16:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.