Self-Improving Transformers Overcome Easy-to-Hard and Length Generalization Challenges

• URL: http://arxiv.org/abs/2502.01612v2
• Date: Thu, 13 Feb 2025 05:32:54 GMT
• Title: Self-Improving Transformers Overcome Easy-to-Hard and Length Generalization Challenges
• Authors: Nayoung Lee, Ziyang Cai, Avi Schwarzschild, Kangwook Lee, Dimitris Papailiopoulos,
• Abstract summary: Large language models often struggle with length generalization and solving complex problem instances beyond their training distribution.<n>We present a self-improvement approach where models iteratively generate and learn from their own solutions.<n>Our results demonstrate how controlled weak-to-strong curricula can systematically teach a model logical extrapolation.
• Score: 15.975023196507841
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language models often struggle with length generalization and solving complex problem instances beyond their training distribution. We present a self-improvement approach where models iteratively generate and learn from their own solutions, progressively tackling harder problems while maintaining a standard transformer architecture. Across diverse tasks including arithmetic, string manipulation, and maze solving, self-improving enables models to solve problems far beyond their initial training distribution-for instance, generalizing from 10-digit to 100-digit addition without apparent saturation. We observe that in some cases filtering for correct self-generated examples leads to exponential improvements in out-of-distribution performance across training rounds. Additionally, starting from pretrained models significantly accelerates this self-improvement process for several tasks. Our results demonstrate how controlled weak-to-strong curricula can systematically teach a model logical extrapolation without any changes to the positional embeddings, or the model architecture.

Related papers

• Can Large Reasoning Models Self-Train? [58.953117118687096]
  Scaling the performance of large language models increasingly depends on methods that reduce reliance on human supervision.<n>We propose an online self-training reinforcement learning algorithm that leverages the model's self-consistency to infer correctness signals and train without any ground-truth supervision.
  arXiv  Detail & Related papers  (2025-05-27T17:16:00Z)
• Iterative Deepening Sampling for Large Language Models [27.807695570974644]
  Training models to achieve effective self-correction and self-correction remains a significant challenge. We propose a novel iterative sampling algorithm framework designed to enhance self-correction and generate higher-quality samples.
  arXiv  Detail & Related papers  (2025-02-08T04:39:51Z)
• Self-supervised Analogical Learning using Language Models [59.64260218737556]
  We propose SAL, a self-supervised analogical learning framework. SAL mimics the human analogy process and trains models to explicitly transfer high-quality symbolic solutions. We show that the resulting models outperform base language models on a wide range of reasoning benchmarks.
  arXiv  Detail & Related papers  (2025-02-03T02:31:26Z)
• RL + Transformer = A General-Purpose Problem Solver [0.0]
  In this study, we demonstrate that a pre-trained transformer fine-tuned with reinforcement learning over multiple episodes develops the ability to solve problems that it has never encountered before. This powerful meta-learner not only excels in solving unseen in-distribution environments with remarkable sample efficiency, but also shows strong performance in out-of-distribution environments.
  arXiv  Detail & Related papers  (2025-01-24T01:55:20Z)
• Multiagent Finetuning: Self Improvement with Diverse Reasoning Chains [114.76612918465948]
  Large language models (LLMs) have achieved remarkable performance in recent years but are fundamentally limited by the underlying training data.<n>We propose a complementary approach towards self-improvement where finetuning is applied to a multiagent society of language models.
  arXiv  Detail & Related papers  (2025-01-10T04:35:46Z)
• Provable unlearning in topic modeling and downstream tasks [36.571324268874264]
  Provable guarantees for unlearning are often limited to supervised learning settings. We provide the first theoretical guarantees for unlearning in the pre-training and fine-tuning paradigm. We show that it is easier to unlearn pre-training data from models that have been fine-tuned to a particular task, and one can unlearn this data without modifying the base model.
  arXiv  Detail & Related papers  (2024-11-19T16:04:31Z)
• SRA-MCTS: Self-driven Reasoning Augmentation with Monte Carlo Tree Search for Code Generation [14.786100203787194]
  Large language models demonstrate exceptional performance in simple code generation tasks but face challenges in tackling complex problems. We propose a reasoning-augmented data generation process, SRA-MCTS, which guides the model to autonomously generate high-quality intermediate reasoning paths. Our method operates entirely through the model itself without requiring additional supervision.
  arXiv  Detail & Related papers  (2024-11-17T12:31:04Z)
• AdaMerging: Adaptive Model Merging for Multi-Task Learning [68.75885518081357]
  This paper introduces an innovative technique called Adaptive Model Merging (AdaMerging) It aims to autonomously learn the coefficients for model merging, either in a task-wise or layer-wise manner, without relying on the original training data. Compared to the current state-of-the-art task arithmetic merging scheme, AdaMerging showcases a remarkable 11% improvement in performance.
  arXiv  Detail & Related papers  (2023-10-04T04:26:33Z)
• Domain Generalization via Balancing Training Difficulty and Model Capability [61.053202176230904]
  Domain generalization (DG) aims to learn domain-generalizable models from one or multiple source domains that can perform well in unseen target domains. Despite its recent progress, most existing work suffers from the misalignment between the difficulty level of training samples and the capability of contemporarily trained models. We design MoDify, a Momentum Difficulty framework that tackles the misalignment by balancing the seesaw between the model's capability and the samples' difficulties.
  arXiv  Detail & Related papers  (2023-09-02T07:09:23Z)
• Self-Supervised Reinforcement Learning that Transfers using Random Features [41.00256493388967]
  We propose a self-supervised reinforcement learning method that enables the transfer of behaviors across tasks with different rewards. Our method is self-supervised in that it can be trained on offline datasets without reward labels, but can then be quickly deployed on new tasks.
  arXiv  Detail & Related papers  (2023-05-26T20:37:06Z)
• Improving Non-autoregressive Generation with Mixup Training [51.61038444990301]
  We present a non-autoregressive generation model based on pre-trained transformer models. We propose a simple and effective iterative training method called MIx Source and pseudo Target. Our experiments on three generation benchmarks including question generation, summarization and paraphrase generation, show that the proposed framework achieves the new state-of-the-art results.
  arXiv  Detail & Related papers  (2021-10-21T13:04:21Z)
• Conditional Generative Modeling via Learning the Latent Space [54.620761775441046]
  We propose a novel framework for conditional generation in multimodal spaces. It uses latent variables to model generalizable learning patterns. At inference, the latent variables are optimized to find optimal solutions corresponding to multiple output modes.
  arXiv  Detail & Related papers  (2020-10-07T03:11:34Z)
• Joint learning of variational representations and solvers for inverse problems with partially-observed data [13.984814587222811]
  In this paper, we design an end-to-end framework allowing to learn actual variational frameworks for inverse problems in a supervised setting. The variational cost and the gradient-based solver are both stated as neural networks using automatic differentiation for the latter. This leads to a data-driven discovery of variational models.
  arXiv  Detail & Related papers  (2020-06-05T19:53:34Z)

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.