An official website of the United States government
Existing continual learning (CL) methods mainly rely on fine-tuning or adapting large language mod- els (LLMs). They still suffer from catastrophic for- getting (CF). Little work has been done to exploit in-context learning (ICL) to leverage the extensive knowledge within LLMs for CL without updating any parameters. However, incrementally learning each new task in ICL necessitates adding training examples from each class of the task to the prompt, which hampers scalability as the prompt length in- creases. This issue not only leads to excessively long prompts that exceed the input token limit of the underlying LLM but also degrades the model’s performance due to the overextended context. To address this, we introduce InCA, a novel approach that integrates an external continual learner (ECL) with ICL to enable scalable CL without CF. The ECL is built incrementally to pre-select a small subset of likely classes for each test instance. By restricting the ICL prompt to only these selected classes, InCA prevents prompt lengths from becom- ing excessively long, while maintaining high per- formance. Experimental results demonstrate that InCA significantly outperforms existing CL base- lines, achieving substantial performance gains.
more »
« less
This content will become publicly available on January 19, 2026
Continual Learning Using Only Large Language Model Prompting
We introduce CLOB, a novel continual learning (CL) paradigm wherein a large language model (LLM) is regarded as a black box. Learning is done incrementally via only verbal prompting. CLOB does not fine-tune any part of the LLM or add any trainable parameters to it. It is particularly suitable for LLMs that are accessible via APIs. We also propose a new CL technique, called CIS, based on incremental summarization that also overcomes the LLM’s input length limit. Experiments show CIS outperforms baselines by a very large margin.
more »
« less
- Award ID(s):
- 2229876
- PAR ID:
- 10577384
- Publisher / Repository:
- The 31st International Conference on Computational Linguistics (COLING-2025)
- Date Published:
- Format(s):
- Medium: X
- Location:
- Abu Dhabi, UAE
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
In this work, we theoretically investigate why large language model (LLM)-empowered agents can solve decision-making problems in the physical world. We consider a hierarchical reinforcement learning (RL) model where the LLM Planner handles high-level task planning and the Actor performs low-level execution. Within this model, the LLM Planner operates in a partially observable Markov decision process (POMDP), iteratively generating language-based subgoals through prompting. Assuming appropriate pretraining data, we prove that the pretrained LLM Planner effectively conducts Bayesian aggregated imitation learning (BAIL) via in-context learning. We also demonstrate the need for exploration beyond the subgoals produced by BAIL, showing that naively executing these subgoals results in linear regret. To address this, we propose an ε-greedy exploration strategy for BAIL, which we prove achieves sublinear regret when pretraining error is low. Finally, we extend our theoretical framework to cases where the LLM Planner acts as a world model to infer the environment’s transition model and to multi-agent settings, facilitating coordination among multiple Actors.more » « less
-
This paper studies continual learning (CL) of a sequence of aspect sentiment classification (ASC) tasks. Although some CL techniques have been proposed for document sentiment classification, we are not aware of any CL work on ASC. A CL system that incrementally learns a sequence of ASC tasks should address the following two issues: (1) transfer knowledge learned from previous tasks to the new task to help it learn a better model, and (2) maintain the performance of the models for previous tasks so that they are not forgotten. This paper proposes a novel capsule network based model called B-CL to address these issues. B-CL markedly improves the ASC performance on both the new task and the old tasks via forward and backward knowledge transfer. The effectiveness of B-CL is demonstrated through extensive experiments.more » « less
-
In this talk, we have presented the use of LLM to study the embodied learning. We have discovered that the use of LLM is really effective to summarize the videos of embodied learning and conventional learning. LLM is also very effective in summarize the sentiment of the user comments of the two different videos. Correlation between the user comments and video content was also made. It is discovered that the use of the embodied learning is rather effective to engage learners in learning roboticsmore » « less
-
Cyber-Physical Systems (CPS) integrate computational elements with physical processes via sensors and actuators. While CPS is expected to have human-level intelligence, traditional machine learning which is trained on specific and isolated datasets seems insufficient to meet such expectation. In recent years, Large Language Models (LLMs), like GPT-4, have experienced explosive growth and show significant improvement in reasoning and language comprehension capabilities which promotes LLM-enabled CPS. In this paper, we present a comprehensive review of these studies about LLM-enabled CPS. First, we overview LLM-enabled CPS and the roles that LLM plays in CPS. Second, we categorize existing works in terms of the application domain and discuss their key contributions. Third, we present commonly-used metrics and benchmarks for LLM-enabled CPS evaluation. Finally, we discuss future research opportunities and corresponding challenges of LLM-enabled CPS.more » « less
