Guiding Through Complexity: What Makes Good Supervision for Hard Reasoning Tasks?

• URL: http://arxiv.org/abs/2410.20533v2
• Date: Wed, 30 Oct 2024 17:56:22 GMT
• Title: Guiding Through Complexity: What Makes Good Supervision for Hard Reasoning Tasks?
• Authors: Xuan He, Da Yin, Nanyun Peng,
• Abstract summary: We investigate various data-driven strategies that offer supervision data at different quality levels upon tasks of varying complexity. We find that even when the outcome error rate for hard task supervision is high, training on such data can outperform perfectly correct supervision on easier subtasks. Our results also reveal that supplementing hard task supervision with the corresponding subtask supervision can yield notable performance improvements.
• Score: 74.88417042125985
• License:
• Abstract: How can "weak teacher models" such as average human annotators or existing AI systems, effectively supervise LLMs to improve performance on hard reasoning tasks, especially those that challenge and requires expertise or daily practice from the teacher models? In this paper, we seek for empirical answers to this question by investigating various data-driven strategies that offer supervision data at different quality levels upon tasks of varying complexity. Two intuitive strategies emerge for teacher models to provide supervision during alignment training: 1) using lower-quality supervision from complete tasks that match the difficulty of the target reasoning tasks, and 2) leveraging higher-quality supervision from easier subtasks that are less challenging. Interestingly, we find that even when the outcome error rate for hard task supervision is high (e.g., 90\%), training on such data can outperform perfectly correct supervision on easier subtasks on multiple hard math benchmarks. We further identify a more critical factor influencing training performance: step-wise error rates, which indicate the severity of errors in solutions. Specifically, training on hard task supervision with the same outcome error rates but disparate step-wise error rates can lead to a 30\% accuracy gap on MATH benchmark. Our results also reveal that supplementing hard task supervision with the corresponding subtask supervision can yield notable performance improvements than simply combining rephrased hard full task supervision, suggesting new avenues for data augmentation. Data and code are released at \url{https://github.com/hexuan21/Weak-to-Strong}.

Related papers

• Data-CUBE: Data Curriculum for Instruction-based Sentence Representation Learning [85.66907881270785]
  We propose a data curriculum method, namely Data-CUBE, that arranges the orders of all the multi-task data for training. In the task level, we aim to find the optimal task order to minimize the total cross-task interference risk. In the instance level, we measure the difficulty of all instances per task, then divide them into the easy-to-difficult mini-batches for training.
  arXiv  Detail & Related papers  (2024-01-07T18:12:20Z)
• Active Instruction Tuning: Improving Cross-Task Generalization by Training on Prompt Sensitive Tasks [101.40633115037983]
  Instruction tuning (IT) achieves impressive zero-shot generalization results by training large language models (LLMs) on a massive amount of diverse tasks with instructions. How to select new tasks to improve the performance and generalizability of IT models remains an open question. We propose active instruction tuning based on prompt uncertainty, a novel framework to identify informative tasks, and then actively tune the models on the selected tasks.
  arXiv  Detail & Related papers  (2023-11-01T04:40:05Z)
• Task Compass: Scaling Multi-task Pre-training with Task Prefix [122.49242976184617]
  Existing studies show that multi-task learning with large-scale supervised tasks suffers from negative effects across tasks. We propose a task prefix guided multi-task pre-training framework to explore the relationships among tasks. Our model can not only serve as the strong foundation backbone for a wide range of tasks but also be feasible as a probing tool for analyzing task relationships.
  arXiv  Detail & Related papers  (2022-10-12T15:02:04Z)
• Task-Agnostic Continual Reinforcement Learning: Gaining Insights and Overcoming Challenges [27.474011433615317]
  Continual learning (CL) enables the development of models and agents that learn from a sequence of tasks. We investigate the factors that contribute to the performance differences between task-agnostic CL and multi-task (MTL) agents.
  arXiv  Detail & Related papers  (2022-05-28T17:59:00Z)
• Boosting Supervised Learning Performance with Co-training [15.986635379046602]
  We propose a new light-weight self-supervised learning framework that could boost supervised learning performance with minimum additional cost. Our results show that both self-supervised tasks can improve the accuracy of the supervised task and, at the same time, demonstrates strong domain adaption capability.
  arXiv  Detail & Related papers  (2021-11-18T17:01:17Z)
• URLB: Unsupervised Reinforcement Learning Benchmark [82.36060735454647]
  We introduce the Unsupervised Reinforcement Learning Benchmark (URLB) URLB consists of two phases: reward-free pre-training and downstream task adaptation with extrinsic rewards. We provide twelve continuous control tasks from three domains for evaluation and open-source code for eight leading unsupervised RL methods.
  arXiv  Detail & Related papers  (2021-10-28T15:07:01Z)
• Hierarchical Reinforcement Learning as a Model of Human Task Interleaving [60.95424607008241]
  We develop a hierarchical model of supervisory control driven by reinforcement learning. The model reproduces known empirical effects of task interleaving. The results support hierarchical RL as a plausible model of task interleaving.
  arXiv  Detail & Related papers  (2020-01-04T17:53:28Z)

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.